Research ArticleFiscal Decentralization and Regional Financial EfficiencyAn Empirical Analysis of Spatial Durbin Model

Jianmin Liu12 Xiaomei Hu1 and Hongli Tang1

1College of Economics and Trade Hunan University Changsha Hunan 410079 China2Hunan University of Finance and Economics Changsha Hunan 410025 China

Correspondence should be addressed to Xiaomei Hu hxmxiaomeihuhnueducn

Received 18 January 2016 Accepted 19 April 2016

Academic Editor Alicia Cordero

Copyright copy 2016 Jianmin Liu et al This is an open access article distributed under the Creative Commons Attribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

Based on panel data covering the period from2003 to 2012 inChinarsquos 281 prefecture-level cities we use superefficiency SBMmodel tomeasure regional financial efficiency and empirically test the spatial effects of fiscal decentralization on regional financial efficiencywith SDMThe estimated results indicate that there exist significant spatial spillover effects among regional financial efficiency withthe features of time inertia and spatial dependenceThe positive promoting effect of fiscal decentralization on financial efficiency inlocal region depends on the symmetry between fiscal expenditure decentralization and revenue decentralization Additionally thereexists inconsistency in the spatial effects of fiscal expenditure decentralization and revenue decentralization on financial efficiencyin neighboring regions The negative effect of fiscal revenue decentralization on financial efficiency in neighboring regions is moresignificant than that of fiscal expenditure decentralization

1 Introduction

Improving financial efficiency via financial innovation is animportant way to break the limitation of financial resourceWith the effects of market-oriented economy ldquovoting withtheir feetrdquo mechanism the flowing of financial resourcesand so on the regional financial competition is increasinglycharacterized as the financial efficiency competition drivenby financial innovation Unlike the dominant direct compe-tition of regional financial services and competitive growthof the number of regional financial institutions financialefficiency competition is mainly reflected in the implicitindirect competition of the financial development model andfinancial ideas This competition focuses on environmentaland institutional factors and pursues improving spatial allo-cation efficiency brought by the financial resources readjust-ment through removing the constraints of various factors Asone of the vital institutional factors fiscal decentralizationwill influence the adequacy of the local governmentsrsquo fiscalresources and the behaviors of local governments to someextent thus affecting the flowing of scarce financial resourcesand regional distribution of financial innovation On one

hand fiscal decentralization between governments is helpfulto induce technology diffusion spillover effect of innovationand competition effect among regions which is conducive toimproving regional financial efficiency while enhancing thepositive external effects of financial innovation On the otherhand local governmentsrsquo excessive or inappropriate interven-tion under the fiscal decentralization system will result in theloss of financial innovationrsquos leading role [1] which is detri-mental to the improvement of financial efficiency The topicsonwhat effects the existing fiscal decentralization systemmayhave on the regional financial efficiency among the regionalfinancial strategic competition and how to improve financialefficiency and achieve balancedfinancial development amongregions by optimizing the fiscal decentralized structure areworthy of our further research

From a general survey of relevant research it is notdifficult to find that people pay close attention to researchregarding the fiscal decentralization and regional financialdevelopment in terms of the relationship mechanism effectand so on Scholars have reached a consensus that the fiscaldecentralization brought by the launch of tax federalism

Hindawi Publishing CorporationDiscrete Dynamics in Nature and SocietyVolume 2016 Article ID 6597138 14 pageshttpdxdoiorg10115520166597138

2 Discrete Dynamics in Nature and Society

reform in 1994 has strengthened the local governmentsrsquo inter-vention in the financial sectors [2ndash7] Most scholars suggestthat the local governments will intervene the allocation anduse of financial capital directly or indirectly to obtain scarcefinancial resources among the regional competition in thecontext of fiscal decentralization system considering theirown fiscal pressures and the competition for regional GDP[8] The consequences of governmentsrsquo intervention in thefinancial sector are mainly reflected in the impact on theallocation efficiency of resource Chen et al [9] and Ding andFu [10] suggested that the impact is negative which resultsin a reduction of allocation efficiency of financial capitalYin et al [11] Yu and Pan [12] Wang et al [13] and Y XJiang and R B Jiang [14] also hold the same opinion Thefurther research of Yang et al [15] found that the impactof fiscal decentralization on allocation efficiency of financialresources stemmed from the degree of integration betweenfiscal decentralization reform and market autonomy regula-tion and also originated from the local governmentsrsquo regu-latory capacity However the strength of local governmentrsquosintervention in financial sectors varies from region to regiondue to the differences of economic ownership structures andfiscal pressures [16ndash18] Some scholars like Chen et al [9]Zhou et al [19] andWang et al [13] carried on the discussionabout the relationship amongfiscal decentralization financialdevelopment and economic growth They found that localgovernmentsrsquo intervention in financial sectors hindered theeconomic growth in the context of fiscal decentralizationsystem [20ndash22] Currently the difference of regional financialdevelopment caused by fiscal decentralization has become aresearch emphasis By constructing an endogenous growthmodel including financial sectors and fiscal decentraliza-tion arrangements Wang and Li [23] suggested that fiscaldecentralization plays positive influence on regional financialdevelopment in the eastern regions while the effects in thecentral and western regions are negative The research ofZhang and Qin [24] found that the economic growth effectof fiscal decentralization on financial development variesfrom region to region namely the effects in the westernand northeastern regions are positive while the effects inthe eastern and central regions are negative Ma et al [25]also suggested that there exist significant regional differencesin the long-term relationship among fiscal decentralizationfinancial development and economic growth To furtherreflect the mutual influence among regions Y X Jiangand R B Jiang [26] measured provincesrsquo regional financialinnovation with the use of efficiency method and then testedspatial external effects of regional financial innovation byspatial econometric models

From a general survey of relevant documents it isnot difficult to find that scholars pay close attention toresearch regarding how financial development and efficiencyare affected by fiscal decentralization Due to differencesin objectives thinking and methodology scholars havefailed to reach a consensus regarding the impact of fiscaldecentralization on financial efficiency We briefly discusssome shortcomings in the previous literature below

First scholars paymore attention to social economic andnoninstitutional factors than institutional factors including

fiscal decentralization in the research regarding the impactof fiscal decentralization on financial efficiency Second mostscholars focus on the fiscal decentralization between thecentral and provincial governments and the estimation offinancial efficiency is mostly concentrated in the provinciallevel which may overlook some of the more detailed micro-scopic features Third most scholars used the fiscal expen-diture decentralization to represent fiscal decentralizationbut overlooked the impact of fiscal revenue decentralizationFourth most scholars only paid more attention to desirableoutputs while choosing the input and output indexes tomeasure regional financial efficiency but ignored the negativeimpact of undesirable outputs

For the aforementioned reasonswe attempt to useChinarsquosprefecture-level city panel data covering the period from2003 to 2012 to measure Chinese financial efficiency mean-while placing fiscal decentralization and financial efficiencyinto a united analytical framework to explore the spatialexternal effects of fiscal decentralization on regional financialefficiency by using spatial econometric models that is thespatial correlations among regional competition of financialefficiency and the relationship between regional financialefficiency and agglomeration under fiscal decentralizationsystem

In this fashion we hope to make the following four keycontributions to the literature

(1) Unlike the former research that found that there aredifferences among regional financial efficiency we provide amore detailed perspective to comprehend spatial differencesof Chinese overall regional financial efficiency by explor-ing the institutional causes of interaction among differentregional financial efficiency meanwhile placing fiscal decen-tralization and financial efficiency into a united analyticalframework

(2) We choose prefecture-level cities as the researchsamples Moreover we will investigate the fiscal decentraliza-tion from two dimensions including fiscal expenditure andrevenue decentralization to make the research results morerealistic

(3) We use SBM model (Slack Based Measure theapproach of computational efficiency based on slack vari-ables) with the features of nonradial and no-angle to measureregional financial efficiency so as to avoid the problemsthat the deviation of measuring results caused by taking noaccount of undesirable output and impossibility of full per-mutation of the financial efficiency for all decision-makingunits (DMU)

(4) We use the SDM which synthesizes the advantagesof a Spatial Lag Model (SLM) and a Spatial Error Model(SEM) to study regional spillover effects that subnationalfiscal expenditure and revenue decentralization have onfinancial efficiency Then we further test for the presence ofintraregion spatial spillovers using the decomposition resultsof direct and indirect effects and we identify the cause offiscal decentralization on regional financial efficiency

The remainder of this paper is structured as followsSection 2 measures the regional financial efficiency andanalyzes the spatial clustering of financial efficiency Section 3builds a theoretical framework to illustrate the effects of fiscal

Discrete Dynamics in Nature and Society 3

decentralization on regional financial efficiency in additionto establishing econometric models Section 4 addresses theselection of variables Section 5 analyzes the empirical resultsand engages in further comparative study Section 6 presentsconclusions

2 Measurement and Spatial Clustering ofFinancial Efficiency

Financial efficiency refers to the input-output ratio of thefinancial sector In order to pursue the goal of economicdevelopment local governments try to maximize financialefficiency via adjusting the allocation of financial resourcesThemainmeasuremethods of financial efficiency in previousresearch include regression analysis method factor analysismethod and the traditional DEAmethodThis paper will usesuperefficiency SBM model considering undesirable outputto measure regional financial efficiency Different from thetraditional radial DEA (Data Envelopment Analysis) modelSBMmodel directly introduces slack variables into the objec-tive function to calculate the efficiency and the economicimplication of SBM model is the maximization of actualprofit instead of the maximization of benefit ratio [27] SBMmodel assumes that there are 119899 decision-making units andeach decision-making unit is constituted by three elementsincluding input desirable output and undesirable outputThe vector forms of the elements are represented as follows119909 isin 119877

119898 119910119889

isin 1198771199031 and 119910

119906

isin 1198771199032 119883 119884119889 and 119884

119906 are matrixes119883 = [119909

1 x

119899] isin 119877

119898times119899 119884119889

= [119910119889

1 119910

119889

119899] isin 119877

1199041times119899 and

119884119906

= [119910119906

1 119910

119906

119899] isin 1198771199042times119899 where 119883 gt 0 119884119889 gt 0 and 119884

119906

gt 0Production possibility set 119875 under the condition of constantreturns to scale is defined as

119875 = (119909 119910119889

119910119906

) | 119909 ge 119883120582 119910119889

le 119884119889

120582 119910119906

ge 119884119906

120582 120582

ge 0

(1)

The SBM model based on the slack variable is expressedas follows

120588lowast

= min1 minus (1119898)sum

119898

119894=1(119908minus

1198941199091198940)

1 + (1 (1199041+ 1199042)) (sum1199041

119903=1(119908119889

119903119910119889

1199030) + sum1199042

119903=1(119908119906

119903119910119906

1199030))

st 1199090= 119883120582 + 119908

minus

119910119889

0= 119884119889

120582 minus 119908119889

119910119906

0= 119884119906

120582 + 119908119906

119908minus

ge 0

119908119889

ge 0

119908119906

ge 0

(2)

In (1) and (2) 1199041and 1199042 respectively represent the slack

of input and output 120588 is the evaluation criteria of efficiency119908minus and 119908

119906 respectively represent the excessive input andundesirable output 119908

119889 indicates the insufficient output 119898

represents the type of input factors and 120582 represents theweight vector Objective function 120588 is strictly decreasing

based on 119908minus 119908119906 and 119908

119889 For a particular evaluation unitwhen 119908

minus

= 0 119908119906 = 0 and 119908119889

= 0 that is 120588 = 1 DMU isefficient for SBM model When 120588 lt 1 it shows that DMU isinefficient for SBM with redundancy of input or deficiency ofoutputMeanwhile Tone [27] also proposed a superefficiencySBM model and the basic principle of superefficiency SBMmodel is excluding the evaluated DMU from the referenceset so that the efficiency value may be greater than 1 whichmade up the defect that all decision-making efficiency valuescannot be calculated

The regional financial efficiency is measured on the basisof the productionmethod regarding the financial firms as thesuppliers of financial products and services [28] In consid-eration of attainability accuracy and consistency of data wetake labor capital and technology and other factors as inputsWhen it comes to the output indexes we take deposits andloans as the desirable outputs while the income gap betweenurban and rural areas is regarded as the undesirable outputThe selection and interpretation of indexes including inputvariables and desirable and undesirable output variables areas follows

(1)The labor input of financial industry (119897119886119887119900119903) as humancapital and intellectual resources of the financial industrythere is no doubt that employed persons engaged in financialindustry will affect the development performance of financialindustry This index is measured by the proportion of theemployees in the financial sectors account for the regionaltotal population

(2) The capital input of financial industry (119888119886119901119894119905119886119897) thefixed asset investment is the main fund for the infrastructureconstruction project in financial industry Since it is difficultto obtain the statistics of fixed asset investment for prefecture-level cities we use a proxy variable measured by the propor-tion that the fixed asset investment accounted for in regionalGDP to represent the capital input of financial industry

(3) Information technology level (119894119899119905119890119903119899119890119905) the improve-ment of financial efficiency is dependent on the innovationand development of electronic finance mobile finance inter-net finance and other emerging financial methods whichdepend on the development of information technology Thisindex is represented by a proxy variable measured by thenumber of international internet users in cities

(4)The desirable output of deposit (119889119890119901119900119904119894119905)The regionalfinancial efficiency ismeasured on the basis of the productionmethod regarding the financial firms as the suppliers offinancial products and services [29] The production methodtakes deposit and loan as output while the input is operatingcost including labor capital and technology The index isrepresented by per capita deposit of financial institutionswhere per capita deposit of financial institutions = totaldeposit of financial institutionsregional population

(5) The desirable output of loan (119897119900119886119899) the loan offinancial institutions especially the banks is the main sourceof profits for financial institutions The index is representedby per capita loan of financial institutions where per capitaloan of financial institutions = total loan of financial institu-tionsregional population

(6)The undesirable output of income gap between urbanand rural (119894119892119886119901) The development of the financial industry

4 Discrete Dynamics in Nature and Society

0

1

2

3

0 05 1 15 2 25

k density fe2003

k density fe2007k density fe2012

x

Figure 1 Kernel density segment curve of financial efficiency

results in money flowing from low income groups to highincome groups due to the threshold and cost to obtain finan-cial services and the difference between regional resourcesendowment which will exacerbate income inequality espe-cially in developing countries With reference to the researchresults of Yao [30] we take urban-rural income ratio as aproxy variable where urban-rural income ratio = per capitadisposable income of urban residentsper capita net incomeof rural residents

Based on the input-output data of 281 prefecture-levelcities covering the period from 2003 to 2012 this paper mea-sures the regional financial efficiency by using MAXDEA64software The calculation results show that there exists greatdifference in regional financial efficiency and the averagefinancial efficiency is 0576 while the standard deviationis 0624 From a regional perspective the average financialefficiency in the eastern central and western regions isrespectively 0628 0503 and 0604 and the financial effi-ciency of the central region is lower than that of eastern andwestern regions In addition this paper makes Kernel densityestimation on regional financial efficiency in the three typicalyears (2003 2007 and 2012) by using nonparametric Gaussnormal Kernel density (nuclear density) segment curve Asshown in Figure 1 the horizontal axis represents the regionalfinancial efficiency value of prefecture-level cities while thevertical axis represents the corresponding Kernel densityvalues

First there shows tendency toward left deviation atregional financial efficiency in Chinarsquos prefecture-level citiesin the view of the specific distribution form which indicatesthat the financial efficiency in most regions are below thenational average level Second themain peak level of financialefficiency gradually declines and moves to the left devia-tion during the three typical years (2003 2007 and 2012)which indicates that the level of financial efficiency graduallydecreases while the gap between the financial efficiency ofdifferent regions is widening In addition Chinarsquos regionalfinancial efficiency distribution form gradually evolves frommultipeak trend in 2003 to the double-peak trend in 2012

Table 1 Moranrsquos 119868 index of regional financial efficiency

Year Geographic distanceweighting matrix

Mixed geography-economyweighting matrix

2003 0152lowastlowastlowast 0130lowastlowastlowast

2004 0094lowastlowastlowast 0061lowastlowastlowast

2005 0094lowastlowastlowast 0028lowast

2006 0135lowastlowastlowast 0092lowastlowastlowast

2007 0137lowastlowastlowast 0100lowastlowastlowast

2008 0075lowastlowastlowast 0057lowastlowastlowast

2009 0019lowastlowastlowast 0024lowastlowastlowast

2010 0141lowastlowastlowast 0088lowastlowastlowast

2011 0140lowastlowastlowast 0071lowastlowastlowast

2012 0149lowastlowastlowast 0052lowastlowastlowast

Note lowastlowastlowast and lowast respectively represent the significant level of 1 and 10

which indicates that the development trend of regionalfinancial efficiency shifts frommultipolarity mode to double-polarity mode

In order to investigate the spatial clustering pattern andthe evolution trend of financial efficiency in 281 prefecture-level cities covering the period from 2003 to 2012 it is neededto further calculate Moranrsquos 119868 index of regional financialefficiency by using global spatial autocorrelation methodMoranrsquos 119868 index is the correlation coefficient of observedvalues and spatial lagged variables and the value of indexranges fromminus1 to 1 If the regional financial efficiency presentsspatial positive correlation the value of Moranrsquos 119868 indexranges from 0 to 1 which indicates that the regions withsimilar observed values are collectively distributed in thespatial range The closer Moranrsquos 119868 index to 1 the strongerthe spatial positive correlation On the contrary the regionswith similar observed value are discretely distributed in thegeographical space The specific measure formula of globalMoranrsquos 119868 index is as follows [31]

Moranrsquos 119868 =

119899sum119899

119894=1sum119899

119895=1119908119894119895(119909119894minus 119909) (119909

119895minus 119909)

sum119899

119894sum119899

119895=1119908119894119895sum119899

119894=1(119909119894minus 119909)2

(3)

In (3) 119909119894and 119909

119895 respectively represent region 119894 and

region 119895 and 119899 represents the number of regions while 119908119894119895

represents the spatial weightmatrixWewillmeasure regionalfinancial efficiencywith two spatial weightmatrixes includinggeographic distance weighting matrix and mixed geography-economy weighting matrix in the view of robustness ofresearch results The specific construction methods of thesetwo spatial weight matrixes are as in Section 4

Table 1 lists Moranrsquos 119868 indexes of regional financial effi-ciency calculatedwith two spatial weightmatrixes Accordingto the estimated results the values of Moranrsquos 119868 index calcu-lated with two spatial weight matrixes are positive and passthe 10 significance test which indicates that there exists sig-nificant positive autocorrelation (spatial dependence) amongregional financial efficiency in the geographical space Thespatial distribution of regional financial efficiency is notrandom which shows that the regions with similar observedvalues are collectively distributed in the geographical space

Discrete Dynamics in Nature and Society 5

fe2003

Wz

z

minus2 minus1 0 1 2 3 4 5 6minus1

0

1

2

3

76

61

75135138

17

113

199

143

20

9414277118

117

33

185

125

116115

1371227

8

55

226

1501395149177178

183

97

18012110

146

181

44911746

179

317124

259526413226136410128158

221

260

219

871585

211

184253100

276

16112913115180

212

167

42

188

186

213

41

89252

215436242

223

83

1283419

12

10286108105781659332

187

1531884

190

119144148152

56

182

25579

227215

2611689274156

176

96133

22

40

126279

269

52

25821

114163

25765

198

170262

229236

166164

243

171

253

3570

204

27215949

47155

50

110

63

8899

235

249

201

54

248

169

210

14029

9

145

206

127

46

244

216

173453873

12010490

57

277233

37

240172112

162157

111

202

124

200

72

60

30134147

24126398160

123256

195

237

11

196

228

64

4369

268

23103107109

230

278

192

13082

27231247

224

23259267

214

207

239

5366139

48

3968

265

238

250

141

194

14

271

189218

275

197

270

220

254

106

24527458

222

208225

175

193

203

205

273251

217

1

209

5162

281

266280

24616

234

67

81

191

Moran scatterplot (Moranrsquos I = 0130)

Figure 2 Moranrsquos 119868 index scatter diagram of financial efficiency in2003

fe2012

Wz

z

minus2 minus1 0 1 2 3 4 5minus1

0

1

2

3

4

258259

180

21

22

181166

173162261257

19155

11

23188

54

187186

218

1492646152

142

220

158

55

125185

126

10418413

212214

12910

1542632332721077

20

237103

9

1592815

51

8211260100

213

37256144

53

312261795150

190

49

239116245

50

15133106261826161121

38

117

241113231

112

177

271

219

16023013744

199

222

1713228122138

210

174

27097

115

221

240153

75248

232143163

179

362940

223244

165

178

242101

23456

102

195

253

471574518

193

229246235

268

108128109

4698

11876273127120

13513242

148

94

139

96

114

7116879

48

77236

167

9389276

169

41

99

74

57

145172

9192

140

12

274

34

27717065133

217

183

24327

198

80

164176

250

16123

255

130

13684175

254

110249

86

204

4514635

262

209

12478252

1526743

60

119

215

131134

203

196

147

88

25

83

72265

269

216

3966

85

247141227

8782

275

278156

192

280

90

224

194

202

68

206

32

207200

6930

205

21111

58

62

189

63

64

5981

208

251

197

14

2387370

225

201

281

24105

279

266

191

267

Moran scatterplot (Moranrsquos I = 0052)

Figure 3 Moranrsquos 119868 index scatter diagram of financial efficiency in2012

Namely it presents a spatial correlation structure that theregions with high financial efficiency are inclined to beadjacent to other high-efficiency regions while the regionswith low efficiency are inclined to be adjacent to other low-efficiency regions

In order to depict the clustering pattern of regionalfinancial efficiency more intuitively we will draw Moranrsquos119868 index scatter diagrams of regional financial efficiency intwo typical years namely 2003 and 2012 with the use of theeconomic-geographical mixed weighting matrix

Moranrsquos 119868 index scatter diagrams divide financial effi-ciency clustering distribution into four quadrants of whichthe first quadrant (HH high efficiency-high spatial lag) andthird quadrant (LL low efficiency-low spatial lag) reflectpositive spatial correlation and the second quadrant (LHlow efficiency-high spatial lag) and fourth quadrant (HL highefficiency-low spatial lag) reflect negative spatial correlationAs shown in Figures 2 and 3 most regions are in thefirst quadrant and the third quadrant which further con-firmed that there exists significant positive spatial correlationbetween regional financial efficiency

3 Model Specification

In order to investigate how the fiscal decentralization influ-ences the financial efficiency and strategic competitionamong regional financial development econometric modelis initially expressed as

fe119894119905

= 1205730+ 120573119883119894119905

+ 120583119894+ ]119905+ 120576119894119905 (4)

In (4) 119894 and 119905 respectively indicate the region 119894 andyear 119905 120583

119894is a regional disturbance ]

119905is timing disturbance

120576119894119905is the random error fe

119894119905represents financial efficiency

and 119909119894119905contains fiscal expenditure decentralization revenue

decentralization and other controlled variablesFlowing of capital and technology across sectors and

practitionersrsquo communication and cooperation lead to obvi-ous clustering and rivalry among regional financial devel-opment which indicates that the financial developmentand efficiency in geographically neighboring regions or theregions with similar economic structure are interdependentrather than independent to some degree Considering thefiscal decentralization system strategic competition causedby public policy spillover and ldquofree riderrdquo behavior furtherenhance the spatial correlation of financial efficiency andfinancial development among regions we find that it isnecessary to introduce the interactive dependencies amongregions into the spatial econometric analysis framework toreflect the spatial spillover effects of financial developmentunder the fiscal decentralization system by constructing aspatial econometric model

It has been proved that there exists spatial correlationin regional financial efficiency by using Moran statisticalanalysis so we can generate spatial econometric models byintroducing spatial autocorrelation factors into the generalregression models According to different forms of errorshock on an observationrsquos spatial autocorrelation spatialeconometric models can be divided into the Spatial ErrorModel (SEM) and the Spatial Lag Model (SLM) In ourresearch the SEM assumes that spatial autocorrelation stemsfrom the error shock of neighboring regions on the depen-dent variables and examines the effects of neighboringregions under observation while the SLM assumes thatspatial autocorrelation stems from dependent variables andexamines the effects of financial efficiency in neighboringregions on local financial efficiency According to the prin-ciples of model construction the SEM reflects the indirectbandwagon effects caused by error terms whereas the SLMcontrols the direct bandwagon effects LeSage and Pace [32]further establish the Spatial Durbin Model (SDM) on thebasis of the SEM and the SLM The SDM includes spatial lagterms fromdependent variables and independent variables tocapture the spillover effects deriving from different variables

Wewill construct static Spatial DurbinModel (SDM) anddynamic Spatial Durbin Model (DSDM) to investigate howfiscal decentralization affects regional financial efficiencyMeanwhile in order to reflect the financial efficiencyrsquos pos-sible inertia characteristics in reality the lag item (fe

119894119905-1) offinancial efficiency will be introduced to the independent

6 Discrete Dynamics in Nature and Society

variables to measure the influence of self-reinforcing mecha-nism The specific models can be expressed as

fe119894119905

= 120588119882119894119895fe119894119905

+ 120573119883119894119905

+ 120579119882119894119895119883119894119905

+ 120583119894+ ]119905+ 120576119894119905

fe119894119905

= 120593fe119894119905-1 + 120588119882

119894119895fe119894119905

+ 120573119883119894119905

+ 120579119882119894119895119883119894119905

+ 120583119894+ ]119905

+ 120576119894119905

(5)

In (5) 120588 is the spatial lag coefficient reflecting thedirection and extent of how financial efficiency in theneighboring regions affects local financial efficiency and thecoefficient directly reflects the degree of spatial competitionin the financial efficiency 119882

119894119895is the spatial weight matrix of

119873119879 times 119873119879 119873 is the number of cross-sectional samples (281prefecture-level cities) and 119879 is time (2003ndash2012)

4 Variables Selection and Data Sources

41 Variables Definition and Measurement Methods Accord-ing to research purpose we will focus on the relationshipbetween fiscal decentralization (fiscal expenditure decentral-ization and revenue decentralization) and regional financialefficiency in the subnational citiesOther economic and socialfactors affecting regional financial efficiency include regionaleconomic development urbanization industrial structureinformation technology development fixed asset investmentand the level of internal openingTherefore wewill take theseindexes as controlled variables and introduce them into anempirical analysis model the detailed variable setting andmeasuring methods are as follows

411 The Dependent Variable The dependent variable isfinancial efficiency (fe) which is measured by the aforemen-tioned SBMmodel

412 Main Independent Variables Fiscal expenditure decen-tralization and revenue decentralization are our main inde-pendent variable When it comes to the measure method offiscal decentralization we mainly use research results of Guoand Jia [33] and Gong and Lei [34] for reference The specificmeasure formula of fiscal expenditure decentralization (sd) isas follows

sd

=

PFEcity

PFEcity + PFEprovince + PFEnation[1 minus

GDPcity

GDPnation]

(6)

In (6) PFEcity PFEprovince and PFEnation respectivelyrepresent per capita fiscal expenditure of each city theprovincial per capita fiscal expenditure and the national percapita fiscal expenditure GDPcity and GDPnation respectivelyrepresent the GDP in each city and the national GDP[1 minus GDPcityGDPnation] is the reduction factor for economicscale The index has proposed the effects of population scale

and economic scale The specific measure formula of fiscalrevenue decentralization (rd) is as follows

rd

=

PFRcity

PFRcity + PFRprovince + PFRnation[1 minus

GDPcity

GDPnation]

(7)

In (7) PFRcity PFRprovince and PFRnation respectivelyrepresent per capita fiscal revenue of each city the provincialper capita fiscal revenue and the national per capita fiscalrevenue GDPcity and GDPnation respectively represent theGDP in each city and the national GDP

413 Controlled Variable Controlled variables are as follows(1) Regional economic development (119901119892119889119901) regional

economy provides the necessary material and nonmaterialsupport for the financial development and innovation thehigher the level of regional economic development is themore beneficial to financial efficiency the improvement isIn this paper per capita actual GDP is used as a proxyvariable for regional economic development To exclude theinfluence of price factor the foregoing data that involve valueare adjusted in accordance with constant 2003 prices

(2) Urbanization (119906119903119887119886119899) Urbanization of population(the proportion of nonagricultural population accounts fortotal population) is used as the proxy variable in which thedata of nonagricultural population covering the period from2003 to 2008 is from ldquoChina Urban Statistical Yearbookrdquo(2004ndash2009) and the data of nonagricultural populationcovering the period from 2009 to 2012 is from the annualldquoStatistics on Population of Counties and Cities in Chinardquo

(3) Industrial structure (119894119905119906) the upgrading and adjust-ment of regional industrial structure constitutes a soft envi-ronment for financial development and innovation Thehigher the proportion of the third industry is the morebeneficial to financial efficiency the improvement is In thispaper the index of industrial structure upgrading is usedto represent industrial structure and the specific measureformula is as follows

itu =

119899

sum

119895=1

119902 (119895) lowast 119895 (8)

In (8) 119902(119895) is the proportion of the third industry accountsfor local GDP in region 119895 119899 = 3 and the value of index rangesfrom 1 to 3

(4) Information technology development (119894119899119905119890119903119899119890119905) it ismeasured by the number of internet users in each city

(5) Fixed asset investment (119894119891119886) it is measured by theactual fixed asset investment in each city

(6)The level of internal opening (119901119903119894V119886119905119890) with referenceto the research result of Fan et al [35] we use the developmentof private enterprises to measure the level of internal openingand assume that the higher the level of the developmentof private enterprise is the stronger the flexibility of localfinancial innovation activities is which is more conduciveto financial efficiency improvement The proportion of theemployees of private enterprises and self-employed employ-ees account for the local total population is used as a proxy to

Discrete Dynamics in Nature and Society 7

measure the development of private enterprises in differentregions

414 Spatial Weight Matrixes Strategic competition tendsto occur in the regions with the similar level of economicdevelopment simultaneously the concentration effect willgradually decay with the geographical distance [36] In orderto comprehensively reflect the spatial effects of geographicaland economic characteristics and accurately measure thespatial correlation among regions we will construct themixed geography-economy weighting matrix The specificmeasure formula is as follows

119882eco-geo =

10038161003816100381610038161003816119876119894minus 119876119895

10038161003816100381610038161003816

119889119894119895

2119894 = 119895

0 119894 = 119895

(9)

In (9) 119882eco-geo is mixed geography-economy weightingmatrix119876

119894and119876

119895 respectively represent the per capita actual

GDP of city 119894 and city 119895 (the mean of the actual GDP coveringthe period from 2003 to 2012) and 119889

119894119895is the geographical

distance between city 119894 and city 119895 calculated with the latitudeand longitude data of prefecture-level cityrsquos administrativeunit The specific measure formula is as follows

119889119894119895

= 119877 arccos (sin120593119860sin120593119861

+ cos120593119860cos120593119861cos (120582

119861minus 120582119860))

(10)

In (10) 119877 is the earth circle radius (6378 km) and 120582 and120593 respectively represent administrative centerrsquos longitudeand latitude of city 119894 and city 119895 Longitude and latitude dataare from the national fundamental geographic informationsystem All of the diagonal elements of matrix 119882eco-geo are0 And the weight matrix is normalized so that the sum ofline elements is 1 In addition considering the reliability androbustness of statistical analysis we will also construct geo-graphical distance weight matrix tomake a spatial economet-ric analysis and then compare the corresponding estimatedresultsThe specificmeasure formula of geographical distanceweight matrix is as follows

119882geo =

1

119889119894119895

2119894 = 119895

0 119894 = 119895

(11)

In (11) 119882geo is a geographical distance weighting matrixwhich is also normalized so that the sum of line elements is 1and other indexes are consistent with that of (10)

42 Data Sources and Description Considering the avail-ability of data and consistency of research samples weeventually choose 281 prefecture-level cities (Hong KongMacao and Taiwan are excluded) as samples and excludeBazhong Ziyang Lhasa Jinchang Baiyin Zhongwei andother cities due to a serious lack of available dataThe researchperiod is covering the period from 2003 to 2012 and thedata is from ldquoChina City Statistical Yearbookrdquo ldquoRegional

Economic China Statistical Yearbookrdquo and provincial statis-tical yearbooks covering the period from 2004 to 2013 Themissing data was partially filled by interpolation In orderto weaken the influence of heteroskedasticity outliers andanomaly terms on the data stability all the data are processedby the logarithm An overview of the descriptive statistics ofthese data is provided in Table 2

In order to avoid the deviation of the model settingand spurious regression we carry on unit root testing onrelated variables by LLC and Fisher-ADF testing methodsbefore spatial econometric regression analysis The resultsof stationary testing of panel data indicate that all variablesreject the null hypothesis of unit root which passes the testat the 5 significance level Moreover the panel data arestationary and the regression results will be significantlyreliable

5 Empirical Methodology

51 Spatial Econometric Analysis of Fiscal Decentralization onRegional Financial Efficiency Since the SDMmodel containsthe spatial lag item of the dependent variable and the spatiallag items of independent variables at the same time so itcanmore fully reflect the impact of spatial autocorrelation onthe regression results than SEM model and SLM model [32]Therefore we construct SDMmodels to test the spatial effectsof the fiscal revenue decentralization and fiscal expendituredecentralization on the regional financial efficiency The 119875

values of Hausman test of the spatial econometric models areless than 005 which indicates that the estimated results offixed effect estimation are better than that of random effectestimation Therefore we will specifically carry out analysisaccording to the estimated results of spatially fixed effectmodels in the following parts Meanwhile a robust test iscarried by constructing DSDM models It is found that theestimated coefficients of SDMmodel and DSDMmodel havethe same direction while the values of estimated coefficientsand level of significance are also very close indicating that theresults of SDM model are significantly reliable Comparingthe values of 119877

2 and Log-likelihood of SDM model andDSDM model we find that the fitting degree of SDM modelis higher than that of DSDMmodel so the following analysisis mainly based on the econometric results of SDMmodel

As shown in Table 3 the overall spatial correlationcoefficients 120588 of four models are positive and pass thetests at the 1 significance level indicating that there arepositive spatial dependent relationships between the localfinancial efficiency and the neighboring regionsrsquo financialefficiency that is the economic activities of geographyneighborhood regions or regions with similar level of eco-nomic development have obvious spillover effect on localfinancial efficiency thereby causing imitation and strate-gic competition of financial development and innovationamong regions On one hand the geographical adjacencyand similarity of economic development in different regionsbring huge convenience to cooperation between financialenterprises sharing of infrastructure information exchangeand communication innovation and diffusion of knowledge

8 Discrete Dynamics in Nature and Society

Table 2 Statistical description of samplersquos variables

Index Variables Mean Standarddeviation Minimum 25 quantile 50 quantile 75 quantile Maximum

Finance efficiency fe 0576 0624 0133 0380 0500 0653 3958Fiscal expendituredecentralization sd 0368 0108 0138 0294 0351 0421 0888

Fiscal revenuedecentralization rd 0242 0146 0037 0133 0202 0318 0871

Per capita real GDP(yuanperson) pgdp 27119890 + 04 32119890 + 04 1891200 9837380 17119890 + 04 31119890 + 04 44119890 + 05

Urbanization () Urban 0345 0195 0013 0202 0296 0451 100000Industrial structure itu 2203 0136 1318 2110 2190 2283 2797Informationtechnologydevelopment (tenthousand families)

Internet 41066 128506 0024 8316 17139 36800 5174000

Fixed assetinvestment (tenthousand yuan)

ifa 15119890 + 04 15119890 + 04 654299 4606420 9841960 20119890 + 04 17119890 + 05

The level of internalopening () Private 0086 0106 0003 0036 0059 0096 100000

and technology flowing of financial resources among regionswhich causes regional financial development and innovationand imitation and strategic competition among regions [37]on the other hand the restriction of transportation cost onfinancial sector is not strong and the rapid development ofinternet technology leads to cross-regional financial transac-tions and innovation thus further expanding the clusteringeffect effect of economic scale and spatial spillover effectsof financial development The coefficients of lag item (fe

119894119905-1)of financial efficiency in DSDM models are positive andthey are respectively 0211 and 0212 both of them pass thetests at the 1 significance level indicating that the previousinputs and outputs of production will be present through thefinancial efficiency increasing in later stage (or later period)and financial efficiency increasing is a dynamic systemicprocess The specific impact of each variable on financialefficiency is analyzed as follows

The estimated coefficients of Lnsd and Lnrd are positiveand all of them pass the tests at the 1 significance levelwhich indicates that the fiscal expenditure and revenuedecentralization play a significant positive role in promotinglocal financial efficiency Under the dual pressures of makingup the fiscal gap and political promotion the local govern-ments regarding economic development as main missionhave instinctive strong interest in scarce flowing economicfactors [38] As the excellent combination of capital talenttechnology management and marketing financial resourceswill be the important target for battling among local govern-ments With the expansion of fiscal expenditure decentral-ization and revenue decentralization local government usesmany policy tools such as fiscal subsidies tax concessionsand government procurement to compete for getting thefinancial resources by attracting financial capital humancapital and technological innovation to flow into the localregion thus contributing to the financial innovation and

improvement of financial efficiencyThe influence coefficientof fiscal expenditure decentralization is 0357 greater thanthe influence coefficient (0153) of fiscal revenue decentral-ization indicating that the impacts of fiscal expendituredecentralization and revenue decentralization on regionalfinancial efficiency have unbalanced characteristics and thepossible reason is that local governments under provinceget very limited taxation autonomy resulting in the factthat revenue decentralization is often lower than expendi-ture decentralization In addition under the capital-driveneconomic growth model the flowing direction of capitaldepends on the level of infrastructure [39] thus the inflowof financial resources as well as financial innovation oftenoccurs in the regions with rapid economic developmentexcellent investment environment and great consumptionability so that local governments have to increase fiscalexpenditure to enhance the supply of the associated support-ing infrastructure Comparatively speaking tax preferenceof undeveloped regions is less attractive than the superiorityconditions of developed regions

We will further analyze the estimated coefficients andsignificance levels of spatial lags of fiscal expenditure andrevenue decentralization and investigate the strategic com-petitive effects of financial efficiency among neighboringregions The spatial lag of expenditure decentralization(WLnsd) does not pass the significance test suggestingthat spatial spillover effect of fiscal expenditure decentral-ization is uncertain and needs further examination whilethe spatial lag of fiscal revenue decentralization (WLnrd)is negative at the 5 significance level and the estimatedcoefficient is minus0228 indicating that with the increase offiscal revenue decentralization the negative externalities onfinancial development and innovation will be increasinglyprominent thus inhibiting the improvement of financialefficiency in neighboring regions The reason may be that

Discrete Dynamics in Nature and Society 9

Table 3 Estimated results of regional financial efficiency under theSpatial Durbin Model (economic-geographic weighting matrix)

Independentvariables

Dependent variable sd Dependent variable rd

SDMmodel DSDMmodel SDMmodel DSDM

model

feitminus10211lowastlowastlowast(1283)

0212lowastlowastlowast(1283)

Lnsd 0357lowastlowastlowast(626)

0334lowastlowastlowast(570)

Lnrd 0153lowastlowastlowast(470)

0143lowastlowastlowast(408)

Lnpgdp 0169lowastlowastlowast(462)

0120lowastlowastlowast(308)

0138lowastlowastlowast(360)

0099lowastlowast(244)

Lnurban minus0050lowastlowast(minus237)

minus0044lowastlowast(minus209)

minus0050lowastlowast(minus237)

minus0045lowastlowast(minus215)

Lnitu 0433lowastlowast(227)

0438lowastlowast(215)

0525lowastlowastlowast(273)

0531lowastlowastlowast(259)

Lninternet minus0295lowastlowastlowast(minus2970)

minus0299lowastlowastlowast(minus2716)

minus0292lowastlowastlowast(minus2945)

minus0297lowastlowastlowast(minus2690)

Lnifa minus0446lowastlowastlowast(minus2298)

minus0386lowastlowastlowast(minus1832)

minus0437lowastlowastlowast(minus2321)

minus0374lowastlowastlowast(minus1806)

Lnprivate 0038lowastlowastlowast(322)

0037lowastlowastlowast(290)

0039lowastlowastlowast(325)

0038lowastlowastlowast(302)

WLnsd 0179(099)

minus0041(minus022)

WLnrd minus0228lowastlowast(minus251)

minus0186lowast(minus179)

WLnpgdp 0016(027)

0064(101)

0058(096)

0097(150)

WLnurban 0097lowastlowast(236)

0082lowastlowast(198)

0107lowastlowastlowast(259)

0089lowastlowast(213)

WLnitu 1159lowast(193)

0516(082)

0615(100)

0228(035)

WLninternet 0177lowastlowastlowast(647)

0181lowastlowastlowast(622)

0180lowastlowastlowast(666)

0180lowastlowastlowast(619)

WLnifa 0237lowastlowastlowast(539)

0251lowastlowastlowast(532)

0275lowastlowastlowast(704)

0260lowastlowastlowast(600)

WLnprivate 0065lowast(185)

0045(115)

0083lowastlowast(232)

0060(153)

120588

0420lowastlowastlowast(1165)

0462lowastlowastlowast(1251)

0429lowastlowastlowast(1205)

0463lowastlowastlowast(1266)

1198772 0448 0443 0441 0438

log 119871 599812 595228 595228 587624Observations 2810 2529 2810 2529Note the values in parentheses represent the 119905-statistics values of thecorresponding estimated coefficientslowastlowastlowastlowastlowast andlowast respectively representsignificance level of 1 5 and 10

local governmentsrsquo motivation of pursuing revenues underthe fiscal revenue decentralization system is not conduciveto free flowing of financial resources among regions whichdirectly affects the financial efficiency in other regions

When it comes to the estimated coefficients of othercontrolled variables we found that not all the spatial spillovereffects of variables are significant Overall various economicactivities have direct and indirect influence on financial

efficiency and promote the strategic financial competitionamong regions The specific effects are as follows the levelof open-into-inside can not only improve the local financialefficiency significantly but also increase the financial effi-ciency correspondingly in neighboring regions The strongdemand for various financial products and services of privateenterprises promotes financial innovation and improves theefficiency of resource allocation in financial markets Onthe contrary urbanization information technology develop-ment and fixed assets investment suppress the improvementof local financial efficiency but play a positive role in promot-ing the financial efficiency in neighboring regionsThe accel-erating of urbanization process the application of internettechnology and investment in fixed assets promote economicexchanges and cooperation among regions so that the spatialspillover effects of financial development and innovation areincreasingly significant Furthermore although the regionaleconomic development and industrial structure significantlycontribute to the improvement of regional financial efficiencyin general the spillover effects among regions are not obvious

52 Decomposition of Fiscal Decentralizationrsquos Effects onRegional Financial Efficiency Whether there truly existspillover effects among the variables relying solely on the spa-tial lag coefficients of SDMmodel and DSDMmodel is likelyto lead to misinterpretation of the estimated results Spatialeffects can be further decomposed into direct effect indirecteffect (spillover effects) and total effect with reference to theresearch result of LeSage and Pace [32] The indirect effectsrepresent the independent variablesrsquo influence on the depen-dent variable in all other regions through spatial interactionsTherefore we will further verify the existence of spillovereffects by using partial differential matrix analysis Table 4illustrates the direct effect indirect effect and total effect ofthe fiscal expenditure and revenue decentralization in SDMmodel The results show that the estimated coefficients ofindependent variablesrsquo direct effect have the same directionwith the estimated coefficients of SDMmodel in Table 3 andthe estimated coefficients and significance levels are also veryclose

According to the estimated results of SDM model inTable 4 the direct effect indirect effect and total effectof fiscal expenditure decentralization on regional financialefficiency are positive and pass the 10 significance leveltest The regression coefficient of direct effect is 0368 theregression coefficient of indirect effect is 0560 and theregression coefficient of total effect is 0928 indicating thatthe degree of fiscal expenditure decentralization increases by1 while regional financial efficiency increases by 0928Fiscal expenditure decentralization plays a significant posi-tive impact on financial development and financial efficiencywhich is opposite to the research result of Wang and Li [23]The opposite conclusions are possibly due to the differencesof sample selection and indexesThe rise of fiscal expendituredecentralization in the local region leads to an increaseby 0368 of the financial efficiency while the increase ofexpenditure decentralization in the neighboring region leadsto financial efficiency increasing by 0560 indicating that

10 Discrete Dynamics in Nature and Society

Table 4 Decomposition of direct effect indirect effect and total effect under SDMmodel

Independent variablesDependent variable sd Dependent variable sd

Direct effect Indirect effect Total effect Direct effect Indirect effect Total effect

Lnsd 0368lowastlowastlowast(750)

0560lowast(195)

0928lowastlowastlowast(310)

Lnrd 0146lowastlowastlowast(529)

minus0283lowast(minus195)

minus0137lowast(minus192)

Lnpgdp 0175lowastlowastlowast(435)

0132(153)

0307lowastlowastlowast(359)

0145lowastlowastlowast(346)

0187lowastlowast(218)

0332lowastlowastlowast(392)

Lnurban minus0046lowastlowast(minus207)

0127lowastlowast(200)

0081(130)

minus0045lowastlowast(minus204)

0146lowastlowast(226)

0101(158)

Lnitu 0486lowastlowastlowast(275)

2321lowastlowast(237)

2807lowastlowastlowast(292)

0559lowastlowastlowast(315)

1488(151)

2047lowastlowast(211)

Lninternet minus0291lowastlowastlowast(minus3055)

0086lowast(190)

minus0205lowastlowastlowast(minus432)

minus0288lowastlowastlowast(minus3029)

0091lowastlowast(200)

minus0197lowastlowastlowast(minus410)

Lnifa minus0441lowastlowastlowast(minus2260)

0087(149)

minus0353lowastlowastlowast(minus594)

minus0430lowastlowastlowast(minus2254)

0154lowastlowastlowast(283)

minus0276lowastlowastlowast(minus494)

Lnprivate 0041lowastlowastlowast(319)

0134lowastlowast(235)

0175lowastlowastlowast(285)

0042lowastlowastlowast(327)

0166lowastlowastlowast(283)

0208lowastlowastlowast(330)

Note lowast lowast lowast lowastlowast and lowast respectively represent the significant level of 1 5 and 10

fiscal expenditure decentralization significantly promotesthe competition of financial development and innovationamong regions The expansion of the autonomy of localgovernment expenditure strengthens the regulatory capacityof local governments which is conducive to acceleratingmarket-oriented reform processing and achieves the organicintegration between fiscal system and market mechanism inthe allocation of financial resources which is beneficial tocreating a favorable institutional environment is also con-ducive to the flowing of financial resources among regionsand greatly promotes financial efficiency

The fiscal revenue decentralizationrsquos direct effect onregional financial efficiency is significantly positive whilethe spillover effect is significantly negative indicating thatthe impact of fiscal revenue decentralization on regionalfinancial efficiency presents the feature of ldquodouble-edged-swordrdquo Fiscal revenue decentralization has a significant posi-tive impact on the local financial efficiencyHowever there aresignificant negative spillover effects on the financial efficiencyin other regions The regression coefficient of the directeffect is 0146 the regression coefficient of the indirect effectis minus0283 and the regression coefficient of the total effectis minus0137 which indicates that the degree of fiscal revenuedecentralization increasing by 1 will lead to decrease by0137 of regional financial efficiency Furthermore whilethe increase of local fiscal revenue decentralization leads tofinancial efficiencyrsquos increase by 0146 the increase of thefiscal revenue decentralization in neighboring regions resultsin the financial efficiencyrsquos decrease by 0283 Comparingthe estimated coefficient of direct effect and that of indirecteffect it can be found that the negative spillover effectsof fiscal revenue decentralization are significantly greaterthan the direct scale effect so the negative externalitiesof fiscal revenue decentralization on financial efficiencyare greater than the positive externality On one hand inorder to obtain high-quality financial resources the local

governments ignore the differences like the level of economicdevelopment factor endowments and other aspects amongregions and seek to imitate policies and instruments inother regions thus leading to chain reaction of absorbinginvestment by other regions On the other hand whenthe neighboring regions or the regions with similar levelof economic development take some measures to attractliquidity factors the local government driven by yardstickcompetitionwill adopt similar policies to avoid the outflow ofelements which results in convergence of competitive tactics[40 41] Tax incentives reduction of fee and other wayswill become important means of competition and imitationbetween local governments in different regions on the basisof the independent revenue authority However such kind ofcompetitive mode between local governments for resourceswill consume huge fiscal resources and weaken the capacityof local governmentrsquos provision of public goods and servicesUltimately it is harmful to provide healthy stable andsustainable economic development environment for financialinnovation and increase of financial efficiency

53 Robustness Test To ensure the robustness and reliabilityof research results we will calculate the spatial lags withthe geographical distance weight matrix and do econometricregression by taking financial efficiency as the dependentvariable so that we can comparatively analyze the differenceof influence with different spatial weight setting based on theabove The robustness test results are as in Table 5

As shown in Table 5 the estimated results indicate thatthe sign of regression coefficients of the fiscal expendituredecentralization and revenue decentralization is consistentwith that of corresponding variable in the model of Table 3and the regression results in themodel are basically consistentwith the conclusions in Table 3 Comparing the estimatedresults based on two different weightmatrixes it is found thatthere are some differences in the influence coefficients of two

Discrete Dynamics in Nature and Society 11

Table 5 Robustness test on Spatial Durbin Models (geographicaldistance weight matrix)

Independentvariables

Dependent variable sd Dependent variable rdSDMmodel

DSDMmodel

SDMmodel

DSDMmodel

feitminus1 mdash 0202lowastlowastlowast(1276) mdash 0200lowastlowastlowast

(1255)

Lnsd or Lnrd 0364lowastlowastlowast(636)

0348lowastlowastlowast(592)

0173lowastlowastlowast(522)

0161lowastlowastlowast(458)

Lnpgdp 0268lowastlowastlowast(735)

0225lowastlowastlowast(578)

0242lowastlowastlowast(637)

0209lowastlowastlowast(522)

Lnurban minus0026(minus112)

minus0026(minus112)

minus0028(minus120)

minus0027(minus115)

Lnitu 0465lowastlowast(246)

0478lowastlowast(238)

0527lowastlowastlowast(278)

0542lowastlowastlowast(269)

Lninternet minus0295lowastlowastlowast(minus3038)

minus0300lowastlowastlowast(minus2799)

minus0292lowastlowastlowast(minus3005)

minus0298lowastlowastlowast(minus2771)

Lnifa minus0484lowastlowastlowast(minus2479)

minus0427lowastlowastlowast(minus2024)

minus0472lowastlowastlowast(minus2448)

minus0417lowastlowastlowast(minus1988)

Lnprivate 0036lowastlowastlowast(312)

0035lowastlowastlowast(284)

0035lowastlowastlowast(307)

0035lowastlowastlowast(288)

WLnsd orWLnrd

minus0264(minus149)

minus0418lowastlowast(minus225)

minus0277lowastlowastlowast(minus370)

minus0241lowastlowastlowast(minus276)

WLnpgdp minus0269lowastlowastlowast(minus427)

minus0286lowastlowastlowast(minus428)

minus0237lowastlowastlowast(minus369)

minus0267lowastlowastlowast(minus394)

WLnurban 0053(147)

0059lowast(167)

0059(163)

0062lowast(173)

WLnitu 0104(020)

minus0085(minus016)

minus0205(minus039)

minus0290(minus051)

WLninternet 0217lowastlowastlowast(838)

0239lowastlowastlowast(853)

0214lowastlowastlowast(831)

0235lowastlowastlowast(841)

WLnifa 0434lowastlowastlowast(1052)

0464lowastlowastlowast(1054)

0433lowastlowastlowast(1165)

0450lowastlowastlowast(1129)

WLnprivate 0062lowast(187)

0038(106)

0081lowastlowast(241)

0050(139)

120588

0547lowastlowastlowast(1858)

0580lowastlowastlowast(1938)

0541lowastlowastlowast(1839)

0570lowastlowastlowast(1898)

1198772 0444 0428 0443 0430

log 119871 659741 665804 665804 659509Observations 2810 2529 2810 2529Note lowastlowastlowast lowastlowast and lowast respectively represent the significant level of 1 5and 10

kinds of fiscal decentralizations indicators which show thatthe gap between the levels of economic developmentwill havean impact on role of fiscal decentralization acting on regionalfinancial efficiency On one hand under the fiscal decen-tralization system because of the heterogeneity of regionaleconomic development level the local governmentsrsquo ability ofmaking up the gap between fiscal revenue and expenditure isdifferent and the fiscal pressure is also different therefore thedegree of interference to the regional market environment bythe local government is different which leads to the regionaldifferences in financial efficiency in a certain degree [21]On the other hand the occurrence of financial innovationdepends on the support of the economic foundation and

the different economic and social environment will lead tothe fact that the financial innovation activities are more andmore concentrated in the regions with the advantages offinancial location resulting in nonequilibrium of financialefficiency in different regions [42] Under the performanceassessment method focusing on GDP local governments willnot only consider the strategic behaviors of the neighboringregions but also tend to think about the behaviors of theregions with the similar level of economy In additioncross-regional knowledge diffusion technology spillover andhuman capital flow are more prone to occur among theregions with similar level of economic development so thatthe financial exchanges and contacts among regions are morefrequent It also shows that besides considering the impact ofgeographical distance the economic distance factor shouldbe taken into consideration while we are researching thespatial effects of fiscal decentralization on regional financialinnovation in addition to the consideration of the impact ofgeographical distance the economic distance factor shouldbe considered

6 Conclusions and Enlightenments

Based on the panel data of 281 prefecture-level cities in Chinacovering the period of 2003ndash2012 the regional financialefficiency is measured by the superefficiency SBMmodel thegeographic distance weight matrix and economic-geographymixed weight matrix which can reflect the regional similarityare introduced into the SDM and the spatial effects of thefiscal expenditure and revenue decentralization on regionalfinancial efficiency under province are investigated as wellas the strategic financial competition arising from it Basedon the results of previous empirical analysis we can draw thefollowing basic conclusions

First the geographical distribution of Chinarsquos prefecture-level citiesrsquo financial efficiency presents significant features oftime inertia and spatial dependency Financial developmentand innovation among regions particularly neighboringregions is not independent but shows significant spatialspillover effects The financial efficiency of the local regionwill be significantly affected by the financial efficiency ofadjacent regions The spatial effect is increasingly obviousin the process of financial development and innovation andthe size of the spillover effects is closely related to spatialweight (which is used to measure regional similarities) andmainly shows that there is competition for financial efficiencyamong regions with similar economic attributes and adjacentregions

Second it cannot be ignored that the institutional factorsespecially fiscal decentralization have effect on the financialefficiency and the factors have a decisive role in the allocationof financial resource and the regional distribution of financialinnovation The improvement of fiscal decentralization levelwill help to attract more financial resources thereby promot-ing an increase in financial efficiencyThe improvement of thefinancial autonomy in neighboring jurisdictions will reducethe financial resources flowing into the local region andsuppresses the occurrence of the local financial innovationIn addition there exist significant spatial spillover effects of

12 Discrete Dynamics in Nature and Society

financial development and efficiency caused by fiscal decen-tralization The scope of spatial spillovers is mainly reflectedin two aspects Thus the local governmentsrsquo competition notonly strongly increases the financial efficiency of the regionbut also accelerates interregional competition for financialefficiency

Third fiscal expenditure decentralization and revenuedecentralizationrsquos impact on regional financial efficiencyhave nonequilibrium characteristics First of all there existsasymmetry between fiscal expenditure decentralization andrevenue decentralizationrsquos impact on regional financial effi-ciency Fiscal expenditure decentralization can promoteoptimization of the fiscal expenditure structure and fiscalrevenue decentralization can increase the local governmentsrsquorevenue share so as to provide positive incentives [43] bothof them have a positive and promoting effect on the improve-ment of local financial efficiency and such effect largelydepends on the symmetry between the fiscal expenditure andrevenue decentralization If it is asymmetric between theirproperty rights and the expenditure responsibilities underfiscal decentralization system the local governments maymake excessive intervention to regional financial resourcesallocation tomeet their own demands for financial resourcesNext there exists nonconformance between fiscal expendi-ture decentralization and revenue decentralizationrsquos impacton financial efficiency in neighboring regions Comparedwith the fiscal expenditure decentralization the fiscal revenuedecentralization has more apparent inhibitory effect on thefinancial efficiency in neighboring regions

Our research conclusions have important policy implica-tions that illuminate the relationship between fiscal decen-tralization and financial efficiency

(1) The regional financial efficiency is closely relatedto the location and the economic development of the sur-rounding cities The regional financial efficiency is affectednot only by the financial development and innovation inneighboring regions but also by the impact of the struc-tural differences among regions In this case breaking theadministrativemonopoly of local protectionism and carryingon the ldquorichening neighborrdquo policy of financial developmentare the best choices for the local government to developthe regional economy When making policies of financialdevelopment local governments need to consider their owneconomic development environment and economic featuresin neighboring regions and strengthen communication andcoordination with the surrounding regions so as to avoidthe possible contradictions and conflicts on financial devel-opment policies among regions

(2) The spatial spillover effects of the regional financialefficiency should be divided into two kinds of effects includ-ing global geographic spillover effects and local geographicspillover effects In order to promote the increase of financialefficiency in a better way it should also be based on thehierarchy of spatial effect of and the degree of spillover effectto define the relationship between the government and themarket and central and local governments and among thelocal governments and their role in regional financial devel-opment and innovation so as to better promote financialefficiency

(3) There is a two-way interactive mechanism betweenthe fiscal system and financial development such effectsmay affect the behaviors of the local governments and theallocation of resources and ultimately act on macroeconomicperformance [10] Under the fiscal decentralization systemthe strategic competitive behaviors among local governmentssignificantly affect the liquidity of financial resources and theoccurrence of financial innovation the local governmentsrsquoexpansion of fiscal expenditure and revenue shortfall becomethe main driving force of each region competing for financialresources In order to better play the role of incentivesand constraints of fiscal competition mechanism under thefiscal decentralization system and avoid the negative impactbrought by the vicious interregional strategic financial com-petitive behaviors on the regional economic development thelocal governmentsrsquo short-sighted behaviors should be strictlyregulated and then the interregional financial cooperationwith mutual benefit as the core should be actively promoted

(4) Constructing a scientific and rational framework forfiscal decentralization will be an important force to accelerateeffectively the improvement of regional financial efficiencyChinarsquos future subprovincial fiscal decentralization reformshould adapt to the new normal financial development andfurther rationalize the provincial and subprovincial fiscaldecentralization system to ensure a balanced and appropriatefiscal expenditure and revenue decentralization On onehand we should weaken local governmentsrsquo negative inter-vention in financial system by reducing local governmentsrsquofinancial pressure caused by the asymmetry between thefiscal resources and expenditure responsibilities On the otherhandwe should also actively play the function of fiscal systemto adjust the financial development and innovation and fullyplay public financersquos role in promoting the flowing of finan-cial resources among regions the balanced development ofregional finance and the improvement of financial efficiency

Competing Interests

There are no competing interests in regard to the contentdiscussed in this paper

Authorsrsquo Contributions

All authors conceived and designed the research XiaomeiHucollected and processed the data Jianmin Liu analyzed thedata and interpreted the results all authors were also involvedin writing the paper and have all approved the submittedform

Acknowledgments

The research of this paper is jointly funded by the NationalSocial Science Foundation of China under Project no14BJY159 and the Hunan Philosophy and Social ScienceFund Project under Grant no 13JD09 The authors gratefullyacknowledge the support of Jinguang Wu from HunanUniversity of Finance and Economics China

Discrete Dynamics in Nature and Society 13

References

[1] J Zhang and X Jin ldquoRe-examine of the relationship betweenChinarsquos financial deepening and productivity 1987ndash2001rdquo Eco-nomic Research Journal vol 11 pp 35ndash43 2005

[2] K Chen A L Hillman and Q Y Gu ldquoFiscal centralization andchanges of local governmentsrsquo behaviorsmdashfrom lsquohands for helprsquoto lsquohands for grabrsquordquoChina EconomicQuarterly vol 1 pp 111ndash1302002

[3] L Zhou and Z M Wang ldquoEmpirical analysis of regionalfinancial development of China and economic growth 1978ndash2000rdquo Financial Research vol 10 pp 1ndash13 2002

[4] L Zhou ldquolsquoSecond financersquo of Chinarsquos financial industry andfinancial division during the reforming periodrdquo The Journal ofWorld Economy vol 6 pp 72ndash79 2003

[5] L Zhou ldquoGradual transition state capacity and fiscal functionof financerdquo Journal of Finance and Economics vol 2 pp 26ndash372005

[6] E L He and W Liu ldquoFinancial capital local governmentsrsquointervention and industrial structuremdashan empirical analysisbased on the data of central regionsrdquo Economic Issues vol 5pp 37ndash40 2011

[7] X L Dou and W J Wang ldquoThe differences and trends of Chi-nese regional financial development and local governmentsrsquobehavior and analysis of tax sharing reform on the differencesand trends of Chinese regional financial developmentrdquo Shang-hai Finance vol 2 pp 9ndash15 2011

[8] J Zhang and K R Shen ldquoLocal government interventionregional financial development and the transformation ofChinarsquos economic growth modemdashan empirical study under thefiscal decentralization systemrdquo Nankai Economic Studies vol 6pp 122ndash141 2008

[9] G Chen X G Yin and Y Pan ldquoChinarsquos financial developmenttax sharing reform and economic growthrdquo Financial Researchvol 2 pp 99ndash109 2006

[10] P P Ding and Y Fu ldquoBehaviors of local governments therelationship of fiscal and finance and macroeconomic volatilityassociated with Chinamdashan analysis under Chinese-style fiscaldecentralization systemrdquo Comparative Economic amp Social Sys-tems vol 6 pp 87ndash97 2012

[11] X G Yin G Chen and Y Pan ldquoTax sharing reform local gov-ernmentsrsquo intervention and financial development efficiencyrdquoJournal of Finance and Economics vol 10 pp 92ndash101 2006

[12] M G Yu and H B Pan ldquoGovernment intervention the ruleof law financial development and bank loans of state-ownedenterprisesrdquo Financial Research vol 9 pp 1ndash23 2008

[13] D X Wang J Liu and L Li ldquoFiscal decentralization bankcredit and total factor productivityrdquo Journal of Finance andEconomics vol 4 pp 69ndash80 2011

[14] Y X Jiang and R B Jiang ldquoRegional financial innovationefficiency evaluation environmental impact and gap analysisrdquoJournal of Zhejiang University (Humanities and Social Sciences)vol 43 no 4 pp 52ndash65 2013

[15] H W Yang L W Cheng Y Xu and Y W Qi ldquoMeasure andinfluencing factors of financial resources allocation efficiencyunder fiscal decentralization system based on super-efficiencyDEA-TOBIT two-step methodrdquo Journal of Jiangxi NormalUniversity (Natural Sciences Edition) vol 6 pp 587ndash592 2014

[16] N Shen F Z Liu and J Q Zhao ldquoFiscal decentralizationfinancial deepening and regional international poverty prone todeveloprdquo Finance amp Trade Economics vol 1 pp 41ndash46 2006

[17] Q Y Du G P Li and N Shen ldquoFinancial efficiency andoptimization of the regional industrial structure under thecurrent fiscal systemrdquo Xinjiang Social Science vol 4 pp 16ndash212006

[18] T Feng Y W Song and Y Lu ldquoFiscal decentralization localgovernmentsrsquo behavior and regional financial developmentrdquoJournal of Xirsquoan Jiaotong University vol 5 pp 20ndash27 2007

[19] Y A Zhou X J Ma and J Y Zhao ldquoGovernment behaviorsfinancial development and economic growthrdquo Henan SocialSciences vol 1 pp 55ndash60 2007

[20] C Hao ldquoDevelopment of financial intermediation and eco-nomic growth the Chinese experiencerdquo China EconomicReview vol 17 no 4 pp 347ndash362 2006

[21] J Zhang Fiscal Decentralization Regional Financial Devel-opment and Chinarsquos Economic Growth Performance NanjingUniversity Press Nanjing China 2011

[22] Y J Yao ldquoGovernment intervention the development of finan-cial intermediation and economic growthrdquo Journal of Financeand Economics vol 8 pp 52ndash58 2010

[23] X Wang and L Li ldquoFiscal decentralization and regional finan-cial development the theoretical and empirical analysis fromthe perspective of central governmentrdquo Public Finance Researchvol 4 pp 31ndash35 2012

[24] W L Zhang and C L Qin ldquoFinancial development fiscaldecentralization and regional economic differencesrdquo EconomicGeography vol 1 pp 583ndash588 2009

[25] Y Ma J Li and B Chen ldquoProvincial differences in Chinesefiscal decentralization financial development industrializationand economic growthrdquo EconomicTheory and Business Manage-ment vol 2 pp 5ndash19 2015

[26] Y X Jiang and R B Jiang ldquoThe spatial external effects ofregional financial innovation financial competition and finan-cial agglomerationrdquo Social Science Front vol 3 pp 79ndash84 2014

[27] K Tone ldquoA slacks-based measure of super-efficiency indata envelopment analysisrdquo European Journal of OperationalResearch vol 143 no 1 pp 32ndash41 2002

[28] Y Q Lu and D G Zhang ldquoChinarsquos regional financial efficiencyand differences of efficiency measurementrdquo Economic Geogra-phy vol 1 pp 96ndash101 2012

[29] A N Berger and D B Humphrey ldquoEfficiency of financialinstitutions international survey and directions for futureresearchrdquo European Journal of Operational Research vol 98 no2 pp 175ndash212 1997

[30] Y J Yao ldquoEmpirical analysis of financial development andurban-rural income gaprdquo Journal of Finance and Economics vol2 pp 49ndash59 2005

[31] Q Li J Song E Wang H Hu J Zhang and Y WangldquoEconomic growth and pollutant emissions in China a spatialeconometric analysisrdquo Stochastic Environmental Research andRisk Assessment vol 28 no 2 pp 429ndash442 2014

[32] J P LeSage and R K Pace Introduction to Spatial EconometricsCRCPress Taylor amp Francis Group Boca Raton Fla USA 2009

[33] Q W Guo and J X Jia ldquoFiscal decentralization organizationalstructure of local governments and the size of governmentexpenditurerdquoEconomic Research Journal vol 11 pp 59ndash87 2010httpwwwcnkicomcnArticleCJFDTotal-JJYJ201011008htm

[34] F Gong and X Lei ldquoThe measurement of Chinese-style fiscaldecentralizationrdquo Statistical Research vol 10 pp 47ndash55 2010

[35] G Fan X L Wang and H P Zhu Chinese Marketing IndexRegional Marketing Process in 2011 Economic Science PressBeijing China 2011

14 Discrete Dynamics in Nature and Society

[36] S S Rosenthal and W C Strange ldquoGeography industrialorganization and agglomerationrdquoTheReview of Economics andStatistics vol 85 no 2 pp 377ndash393 2003

[37] S Park Yoon International Banking and Financial CentersKluwer Academic Boston Mass USA 1989

[38] B Y Qiao J Y Fan and X Y Feng ldquoChinarsquos fiscal decentral-ization and compulsory primary educationrdquo Social Sciences inChina vol 6 pp 37ndash46 2005

[39] W JWang and C L Qin ldquoRegional differences in the behaviorsof local governments and growth effect of fiscal decentralizationbased on the analysis of judgment hypothesis and testingrdquoManagement World vol 1 pp 9ndash21 2008

[40] J K Brueckner ldquoStrategic interaction among governments anoverview of empirical studiesrdquo International Regional ScienceReview vol 26 no 2 pp 175ndash188 2003

[41] F Revelli ldquoOn spatial public finance empiricsrdquo International Taxand Public Finance vol 12 no 4 pp 475ndash492 2005

[42] N Shen and T Y He ldquoFinancial efficiency and regional eco-nomic growth from the perspective of fiscal decentralizationrdquoReform vol 1 pp 61ndash65 2006

[43] J M Liu X M Hu and J G Wu ldquoThe threshold effect ofsub-national fiscal decentralization revenue and expendituredecentralization on provincial industrial transformation andupgrading based on the data of 14 cities in Hunan ProvincerdquoPublic Finance Research vol 8 pp 49ndash52 2014

Submit your manuscripts athttpwwwhindawicom

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

MathematicsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Mathematical Problems in Engineering

Hindawi Publishing Corporationhttpwwwhindawicom

Differential EquationsInternational Journal of

Volume 2014

Applied MathematicsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Probability and StatisticsHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Mathematical PhysicsAdvances in

Complex AnalysisJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

OptimizationJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

CombinatoricsHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Operations ResearchAdvances in

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Function Spaces

Abstract and Applied AnalysisHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of Mathematics and Mathematical Sciences

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Algebra

Discrete Dynamics in Nature and Society

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Decision SciencesAdvances in

Discrete MathematicsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom

Volume 2014 Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Stochastic AnalysisInternational Journal of

2 Discrete Dynamics in Nature and Society

reform in 1994 has strengthened the local governmentsrsquo inter-vention in the financial sectors [2ndash7] Most scholars suggestthat the local governments will intervene the allocation anduse of financial capital directly or indirectly to obtain scarcefinancial resources among the regional competition in thecontext of fiscal decentralization system considering theirown fiscal pressures and the competition for regional GDP[8] The consequences of governmentsrsquo intervention in thefinancial sector are mainly reflected in the impact on theallocation efficiency of resource Chen et al [9] and Ding andFu [10] suggested that the impact is negative which resultsin a reduction of allocation efficiency of financial capitalYin et al [11] Yu and Pan [12] Wang et al [13] and Y XJiang and R B Jiang [14] also hold the same opinion Thefurther research of Yang et al [15] found that the impactof fiscal decentralization on allocation efficiency of financialresources stemmed from the degree of integration betweenfiscal decentralization reform and market autonomy regula-tion and also originated from the local governmentsrsquo regu-latory capacity However the strength of local governmentrsquosintervention in financial sectors varies from region to regiondue to the differences of economic ownership structures andfiscal pressures [16ndash18] Some scholars like Chen et al [9]Zhou et al [19] andWang et al [13] carried on the discussionabout the relationship amongfiscal decentralization financialdevelopment and economic growth They found that localgovernmentsrsquo intervention in financial sectors hindered theeconomic growth in the context of fiscal decentralizationsystem [20ndash22] Currently the difference of regional financialdevelopment caused by fiscal decentralization has become aresearch emphasis By constructing an endogenous growthmodel including financial sectors and fiscal decentraliza-tion arrangements Wang and Li [23] suggested that fiscaldecentralization plays positive influence on regional financialdevelopment in the eastern regions while the effects in thecentral and western regions are negative The research ofZhang and Qin [24] found that the economic growth effectof fiscal decentralization on financial development variesfrom region to region namely the effects in the westernand northeastern regions are positive while the effects inthe eastern and central regions are negative Ma et al [25]also suggested that there exist significant regional differencesin the long-term relationship among fiscal decentralizationfinancial development and economic growth To furtherreflect the mutual influence among regions Y X Jiangand R B Jiang [26] measured provincesrsquo regional financialinnovation with the use of efficiency method and then testedspatial external effects of regional financial innovation byspatial econometric models

From a general survey of relevant documents it isnot difficult to find that scholars pay close attention toresearch regarding how financial development and efficiencyare affected by fiscal decentralization Due to differencesin objectives thinking and methodology scholars havefailed to reach a consensus regarding the impact of fiscaldecentralization on financial efficiency We briefly discusssome shortcomings in the previous literature below

First scholars paymore attention to social economic andnoninstitutional factors than institutional factors including

fiscal decentralization in the research regarding the impactof fiscal decentralization on financial efficiency Second mostscholars focus on the fiscal decentralization between thecentral and provincial governments and the estimation offinancial efficiency is mostly concentrated in the provinciallevel which may overlook some of the more detailed micro-scopic features Third most scholars used the fiscal expen-diture decentralization to represent fiscal decentralizationbut overlooked the impact of fiscal revenue decentralizationFourth most scholars only paid more attention to desirableoutputs while choosing the input and output indexes tomeasure regional financial efficiency but ignored the negativeimpact of undesirable outputs

For the aforementioned reasonswe attempt to useChinarsquosprefecture-level city panel data covering the period from2003 to 2012 to measure Chinese financial efficiency mean-while placing fiscal decentralization and financial efficiencyinto a united analytical framework to explore the spatialexternal effects of fiscal decentralization on regional financialefficiency by using spatial econometric models that is thespatial correlations among regional competition of financialefficiency and the relationship between regional financialefficiency and agglomeration under fiscal decentralizationsystem

In this fashion we hope to make the following four keycontributions to the literature

(1) Unlike the former research that found that there aredifferences among regional financial efficiency we provide amore detailed perspective to comprehend spatial differencesof Chinese overall regional financial efficiency by explor-ing the institutional causes of interaction among differentregional financial efficiency meanwhile placing fiscal decen-tralization and financial efficiency into a united analyticalframework

(2) We choose prefecture-level cities as the researchsamples Moreover we will investigate the fiscal decentraliza-tion from two dimensions including fiscal expenditure andrevenue decentralization to make the research results morerealistic

(3) We use SBM model (Slack Based Measure theapproach of computational efficiency based on slack vari-ables) with the features of nonradial and no-angle to measureregional financial efficiency so as to avoid the problemsthat the deviation of measuring results caused by taking noaccount of undesirable output and impossibility of full per-mutation of the financial efficiency for all decision-makingunits (DMU)

(4) We use the SDM which synthesizes the advantagesof a Spatial Lag Model (SLM) and a Spatial Error Model(SEM) to study regional spillover effects that subnationalfiscal expenditure and revenue decentralization have onfinancial efficiency Then we further test for the presence ofintraregion spatial spillovers using the decomposition resultsof direct and indirect effects and we identify the cause offiscal decentralization on regional financial efficiency

The remainder of this paper is structured as followsSection 2 measures the regional financial efficiency andanalyzes the spatial clustering of financial efficiency Section 3builds a theoretical framework to illustrate the effects of fiscal

Discrete Dynamics in Nature and Society 3

decentralization on regional financial efficiency in additionto establishing econometric models Section 4 addresses theselection of variables Section 5 analyzes the empirical resultsand engages in further comparative study Section 6 presentsconclusions

2 Measurement and Spatial Clustering ofFinancial Efficiency

Financial efficiency refers to the input-output ratio of thefinancial sector In order to pursue the goal of economicdevelopment local governments try to maximize financialefficiency via adjusting the allocation of financial resourcesThemainmeasuremethods of financial efficiency in previousresearch include regression analysis method factor analysismethod and the traditional DEAmethodThis paper will usesuperefficiency SBM model considering undesirable outputto measure regional financial efficiency Different from thetraditional radial DEA (Data Envelopment Analysis) modelSBMmodel directly introduces slack variables into the objec-tive function to calculate the efficiency and the economicimplication of SBM model is the maximization of actualprofit instead of the maximization of benefit ratio [27] SBMmodel assumes that there are 119899 decision-making units andeach decision-making unit is constituted by three elementsincluding input desirable output and undesirable outputThe vector forms of the elements are represented as follows119909 isin 119877

119898 119910119889

isin 1198771199031 and 119910

119906

isin 1198771199032 119883 119884119889 and 119884

119906 are matrixes119883 = [119909

1 x

119899] isin 119877

119898times119899 119884119889

= [119910119889

1 119910

119889

119899] isin 119877

1199041times119899 and

119884119906

= [119910119906

1 119910

119906

119899] isin 1198771199042times119899 where 119883 gt 0 119884119889 gt 0 and 119884

119906

gt 0Production possibility set 119875 under the condition of constantreturns to scale is defined as

119875 = (119909 119910119889

119910119906

) | 119909 ge 119883120582 119910119889

le 119884119889

120582 119910119906

ge 119884119906

120582 120582

ge 0

(1)

The SBM model based on the slack variable is expressedas follows

120588lowast

= min1 minus (1119898)sum

119898

119894=1(119908minus

1198941199091198940)

1 + (1 (1199041+ 1199042)) (sum1199041

119903=1(119908119889

119903119910119889

1199030) + sum1199042

119903=1(119908119906

119903119910119906

1199030))

st 1199090= 119883120582 + 119908

minus

119910119889

0= 119884119889

120582 minus 119908119889

119910119906

0= 119884119906

120582 + 119908119906

119908minus

ge 0

119908119889

ge 0

119908119906

ge 0

(2)

In (1) and (2) 1199041and 1199042 respectively represent the slack

of input and output 120588 is the evaluation criteria of efficiency119908minus and 119908

119906 respectively represent the excessive input andundesirable output 119908

119889 indicates the insufficient output 119898

represents the type of input factors and 120582 represents theweight vector Objective function 120588 is strictly decreasing

based on 119908minus 119908119906 and 119908

119889 For a particular evaluation unitwhen 119908

minus

= 0 119908119906 = 0 and 119908119889

= 0 that is 120588 = 1 DMU isefficient for SBM model When 120588 lt 1 it shows that DMU isinefficient for SBM with redundancy of input or deficiency ofoutputMeanwhile Tone [27] also proposed a superefficiencySBM model and the basic principle of superefficiency SBMmodel is excluding the evaluated DMU from the referenceset so that the efficiency value may be greater than 1 whichmade up the defect that all decision-making efficiency valuescannot be calculated

The regional financial efficiency is measured on the basisof the productionmethod regarding the financial firms as thesuppliers of financial products and services [28] In consid-eration of attainability accuracy and consistency of data wetake labor capital and technology and other factors as inputsWhen it comes to the output indexes we take deposits andloans as the desirable outputs while the income gap betweenurban and rural areas is regarded as the undesirable outputThe selection and interpretation of indexes including inputvariables and desirable and undesirable output variables areas follows

(1)The labor input of financial industry (119897119886119887119900119903) as humancapital and intellectual resources of the financial industrythere is no doubt that employed persons engaged in financialindustry will affect the development performance of financialindustry This index is measured by the proportion of theemployees in the financial sectors account for the regionaltotal population

(2) The capital input of financial industry (119888119886119901119894119905119886119897) thefixed asset investment is the main fund for the infrastructureconstruction project in financial industry Since it is difficultto obtain the statistics of fixed asset investment for prefecture-level cities we use a proxy variable measured by the propor-tion that the fixed asset investment accounted for in regionalGDP to represent the capital input of financial industry

(3) Information technology level (119894119899119905119890119903119899119890119905) the improve-ment of financial efficiency is dependent on the innovationand development of electronic finance mobile finance inter-net finance and other emerging financial methods whichdepend on the development of information technology Thisindex is represented by a proxy variable measured by thenumber of international internet users in cities

(4)The desirable output of deposit (119889119890119901119900119904119894119905)The regionalfinancial efficiency ismeasured on the basis of the productionmethod regarding the financial firms as the suppliers offinancial products and services [29] The production methodtakes deposit and loan as output while the input is operatingcost including labor capital and technology The index isrepresented by per capita deposit of financial institutionswhere per capita deposit of financial institutions = totaldeposit of financial institutionsregional population

(5) The desirable output of loan (119897119900119886119899) the loan offinancial institutions especially the banks is the main sourceof profits for financial institutions The index is representedby per capita loan of financial institutions where per capitaloan of financial institutions = total loan of financial institu-tionsregional population

(6)The undesirable output of income gap between urbanand rural (119894119892119886119901) The development of the financial industry

4 Discrete Dynamics in Nature and Society

0

1

2

3

0 05 1 15 2 25

k density fe2003

k density fe2007k density fe2012

x

Figure 1 Kernel density segment curve of financial efficiency

results in money flowing from low income groups to highincome groups due to the threshold and cost to obtain finan-cial services and the difference between regional resourcesendowment which will exacerbate income inequality espe-cially in developing countries With reference to the researchresults of Yao [30] we take urban-rural income ratio as aproxy variable where urban-rural income ratio = per capitadisposable income of urban residentsper capita net incomeof rural residents

Based on the input-output data of 281 prefecture-levelcities covering the period from 2003 to 2012 this paper mea-sures the regional financial efficiency by using MAXDEA64software The calculation results show that there exists greatdifference in regional financial efficiency and the averagefinancial efficiency is 0576 while the standard deviationis 0624 From a regional perspective the average financialefficiency in the eastern central and western regions isrespectively 0628 0503 and 0604 and the financial effi-ciency of the central region is lower than that of eastern andwestern regions In addition this paper makes Kernel densityestimation on regional financial efficiency in the three typicalyears (2003 2007 and 2012) by using nonparametric Gaussnormal Kernel density (nuclear density) segment curve Asshown in Figure 1 the horizontal axis represents the regionalfinancial efficiency value of prefecture-level cities while thevertical axis represents the corresponding Kernel densityvalues

First there shows tendency toward left deviation atregional financial efficiency in Chinarsquos prefecture-level citiesin the view of the specific distribution form which indicatesthat the financial efficiency in most regions are below thenational average level Second themain peak level of financialefficiency gradually declines and moves to the left devia-tion during the three typical years (2003 2007 and 2012)which indicates that the level of financial efficiency graduallydecreases while the gap between the financial efficiency ofdifferent regions is widening In addition Chinarsquos regionalfinancial efficiency distribution form gradually evolves frommultipeak trend in 2003 to the double-peak trend in 2012

Table 1 Moranrsquos 119868 index of regional financial efficiency

Year Geographic distanceweighting matrix

Mixed geography-economyweighting matrix

2003 0152lowastlowastlowast 0130lowastlowastlowast

2004 0094lowastlowastlowast 0061lowastlowastlowast

2005 0094lowastlowastlowast 0028lowast

2006 0135lowastlowastlowast 0092lowastlowastlowast

2007 0137lowastlowastlowast 0100lowastlowastlowast

2008 0075lowastlowastlowast 0057lowastlowastlowast

2009 0019lowastlowastlowast 0024lowastlowastlowast

2010 0141lowastlowastlowast 0088lowastlowastlowast

2011 0140lowastlowastlowast 0071lowastlowastlowast

2012 0149lowastlowastlowast 0052lowastlowastlowast

Note lowastlowastlowast and lowast respectively represent the significant level of 1 and 10

which indicates that the development trend of regionalfinancial efficiency shifts frommultipolarity mode to double-polarity mode

In order to investigate the spatial clustering pattern andthe evolution trend of financial efficiency in 281 prefecture-level cities covering the period from 2003 to 2012 it is neededto further calculate Moranrsquos 119868 index of regional financialefficiency by using global spatial autocorrelation methodMoranrsquos 119868 index is the correlation coefficient of observedvalues and spatial lagged variables and the value of indexranges fromminus1 to 1 If the regional financial efficiency presentsspatial positive correlation the value of Moranrsquos 119868 indexranges from 0 to 1 which indicates that the regions withsimilar observed values are collectively distributed in thespatial range The closer Moranrsquos 119868 index to 1 the strongerthe spatial positive correlation On the contrary the regionswith similar observed value are discretely distributed in thegeographical space The specific measure formula of globalMoranrsquos 119868 index is as follows [31]

Moranrsquos 119868 =

119899sum119899

119894=1sum119899

119895=1119908119894119895(119909119894minus 119909) (119909

119895minus 119909)

sum119899

119894sum119899

119895=1119908119894119895sum119899

119894=1(119909119894minus 119909)2

(3)

In (3) 119909119894and 119909

119895 respectively represent region 119894 and

region 119895 and 119899 represents the number of regions while 119908119894119895

represents the spatial weightmatrixWewillmeasure regionalfinancial efficiencywith two spatial weightmatrixes includinggeographic distance weighting matrix and mixed geography-economy weighting matrix in the view of robustness ofresearch results The specific construction methods of thesetwo spatial weight matrixes are as in Section 4

Table 1 lists Moranrsquos 119868 indexes of regional financial effi-ciency calculatedwith two spatial weightmatrixes Accordingto the estimated results the values of Moranrsquos 119868 index calcu-lated with two spatial weight matrixes are positive and passthe 10 significance test which indicates that there exists sig-nificant positive autocorrelation (spatial dependence) amongregional financial efficiency in the geographical space Thespatial distribution of regional financial efficiency is notrandom which shows that the regions with similar observedvalues are collectively distributed in the geographical space

Discrete Dynamics in Nature and Society 5

fe2003

Wz

z

minus2 minus1 0 1 2 3 4 5 6minus1

0

1

2

3

76

61

75135138

17

113

199

143

20

9414277118

117

33

185

125

116115

1371227

8

55

226

1501395149177178

183

97

18012110

146

181

44911746

179

317124

259526413226136410128158

221

260

219

871585

211

184253100

276

16112913115180

212

167

42

188

186

213

41

89252

215436242

223

83

1283419

12

10286108105781659332

187

1531884

190

119144148152

56

182

25579

227215

2611689274156

176

96133

22

40

126279

269

52

25821

114163

25765

198

170262

229236

166164

243

171

253

3570

204

27215949

47155

50

110

63

8899

235

249

201

54

248

169

210

14029

9

145

206

127

46

244

216

173453873

12010490

57

277233

37

240172112

162157

111

202

124

200

72

60

30134147

24126398160

123256

195

237

11

196

228

64

4369

268

23103107109

230

278

192

13082

27231247

224

23259267

214

207

239

5366139

48

3968

265

238

250

141

194

14

271

189218

275

197

270

220

254

106

24527458

222

208225

175

193

203

205

273251

217

1

209

5162

281

266280

24616

234

67

81

191

Moran scatterplot (Moranrsquos I = 0130)

Figure 2 Moranrsquos 119868 index scatter diagram of financial efficiency in2003

fe2012

Wz

z

minus2 minus1 0 1 2 3 4 5minus1

0

1

2

3

4

258259

180

21

22

181166

173162261257

19155

11

23188

54

187186

218

1492646152

142

220

158

55

125185

126

10418413

212214

12910

1542632332721077

20

237103

9

1592815

51

8211260100

213

37256144

53

312261795150

190

49

239116245

50

15133106261826161121

38

117

241113231

112

177

271

219

16023013744

199

222

1713228122138

210

174

27097

115

221

240153

75248

232143163

179

362940

223244

165

178

242101

23456

102

195

253

471574518

193

229246235

268

108128109

4698

11876273127120

13513242

148

94

139

96

114

7116879

48

77236

167

9389276

169

41

99

74

57

145172

9192

140

12

274

34

27717065133

217

183

24327

198

80

164176

250

16123

255

130

13684175

254

110249

86

204

4514635

262

209

12478252

1526743

60

119

215

131134

203

196

147

88

25

83

72265

269

216

3966

85

247141227

8782

275

278156

192

280

90

224

194

202

68

206

32

207200

6930

205

21111

58

62

189

63

64

5981

208

251

197

14

2387370

225

201

281

24105

279

266

191

267

Moran scatterplot (Moranrsquos I = 0052)

Figure 3 Moranrsquos 119868 index scatter diagram of financial efficiency in2012

Namely it presents a spatial correlation structure that theregions with high financial efficiency are inclined to beadjacent to other high-efficiency regions while the regionswith low efficiency are inclined to be adjacent to other low-efficiency regions

In order to depict the clustering pattern of regionalfinancial efficiency more intuitively we will draw Moranrsquos119868 index scatter diagrams of regional financial efficiency intwo typical years namely 2003 and 2012 with the use of theeconomic-geographical mixed weighting matrix

Moranrsquos 119868 index scatter diagrams divide financial effi-ciency clustering distribution into four quadrants of whichthe first quadrant (HH high efficiency-high spatial lag) andthird quadrant (LL low efficiency-low spatial lag) reflectpositive spatial correlation and the second quadrant (LHlow efficiency-high spatial lag) and fourth quadrant (HL highefficiency-low spatial lag) reflect negative spatial correlationAs shown in Figures 2 and 3 most regions are in thefirst quadrant and the third quadrant which further con-firmed that there exists significant positive spatial correlationbetween regional financial efficiency

3 Model Specification

In order to investigate how the fiscal decentralization influ-ences the financial efficiency and strategic competitionamong regional financial development econometric modelis initially expressed as

fe119894119905

= 1205730+ 120573119883119894119905

+ 120583119894+ ]119905+ 120576119894119905 (4)

In (4) 119894 and 119905 respectively indicate the region 119894 andyear 119905 120583

119894is a regional disturbance ]

119905is timing disturbance

120576119894119905is the random error fe

119894119905represents financial efficiency

and 119909119894119905contains fiscal expenditure decentralization revenue

decentralization and other controlled variablesFlowing of capital and technology across sectors and

practitionersrsquo communication and cooperation lead to obvi-ous clustering and rivalry among regional financial devel-opment which indicates that the financial developmentand efficiency in geographically neighboring regions or theregions with similar economic structure are interdependentrather than independent to some degree Considering thefiscal decentralization system strategic competition causedby public policy spillover and ldquofree riderrdquo behavior furtherenhance the spatial correlation of financial efficiency andfinancial development among regions we find that it isnecessary to introduce the interactive dependencies amongregions into the spatial econometric analysis framework toreflect the spatial spillover effects of financial developmentunder the fiscal decentralization system by constructing aspatial econometric model

It has been proved that there exists spatial correlationin regional financial efficiency by using Moran statisticalanalysis so we can generate spatial econometric models byintroducing spatial autocorrelation factors into the generalregression models According to different forms of errorshock on an observationrsquos spatial autocorrelation spatialeconometric models can be divided into the Spatial ErrorModel (SEM) and the Spatial Lag Model (SLM) In ourresearch the SEM assumes that spatial autocorrelation stemsfrom the error shock of neighboring regions on the depen-dent variables and examines the effects of neighboringregions under observation while the SLM assumes thatspatial autocorrelation stems from dependent variables andexamines the effects of financial efficiency in neighboringregions on local financial efficiency According to the prin-ciples of model construction the SEM reflects the indirectbandwagon effects caused by error terms whereas the SLMcontrols the direct bandwagon effects LeSage and Pace [32]further establish the Spatial Durbin Model (SDM) on thebasis of the SEM and the SLM The SDM includes spatial lagterms fromdependent variables and independent variables tocapture the spillover effects deriving from different variables

Wewill construct static Spatial DurbinModel (SDM) anddynamic Spatial Durbin Model (DSDM) to investigate howfiscal decentralization affects regional financial efficiencyMeanwhile in order to reflect the financial efficiencyrsquos pos-sible inertia characteristics in reality the lag item (fe

119894119905-1) offinancial efficiency will be introduced to the independent

6 Discrete Dynamics in Nature and Society

variables to measure the influence of self-reinforcing mecha-nism The specific models can be expressed as

fe119894119905

= 120588119882119894119895fe119894119905

+ 120573119883119894119905

+ 120579119882119894119895119883119894119905

+ 120583119894+ ]119905+ 120576119894119905

fe119894119905

= 120593fe119894119905-1 + 120588119882

119894119895fe119894119905

+ 120573119883119894119905

+ 120579119882119894119895119883119894119905

+ 120583119894+ ]119905

+ 120576119894119905

(5)

In (5) 120588 is the spatial lag coefficient reflecting thedirection and extent of how financial efficiency in theneighboring regions affects local financial efficiency and thecoefficient directly reflects the degree of spatial competitionin the financial efficiency 119882

119894119895is the spatial weight matrix of

119873119879 times 119873119879 119873 is the number of cross-sectional samples (281prefecture-level cities) and 119879 is time (2003ndash2012)

4 Variables Selection and Data Sources

41 Variables Definition and Measurement Methods Accord-ing to research purpose we will focus on the relationshipbetween fiscal decentralization (fiscal expenditure decentral-ization and revenue decentralization) and regional financialefficiency in the subnational citiesOther economic and socialfactors affecting regional financial efficiency include regionaleconomic development urbanization industrial structureinformation technology development fixed asset investmentand the level of internal openingTherefore wewill take theseindexes as controlled variables and introduce them into anempirical analysis model the detailed variable setting andmeasuring methods are as follows

411 The Dependent Variable The dependent variable isfinancial efficiency (fe) which is measured by the aforemen-tioned SBMmodel

412 Main Independent Variables Fiscal expenditure decen-tralization and revenue decentralization are our main inde-pendent variable When it comes to the measure method offiscal decentralization we mainly use research results of Guoand Jia [33] and Gong and Lei [34] for reference The specificmeasure formula of fiscal expenditure decentralization (sd) isas follows

sd

=

PFEcity

PFEcity + PFEprovince + PFEnation[1 minus

GDPcity

GDPnation]

(6)

In (6) PFEcity PFEprovince and PFEnation respectivelyrepresent per capita fiscal expenditure of each city theprovincial per capita fiscal expenditure and the national percapita fiscal expenditure GDPcity and GDPnation respectivelyrepresent the GDP in each city and the national GDP[1 minus GDPcityGDPnation] is the reduction factor for economicscale The index has proposed the effects of population scale

and economic scale The specific measure formula of fiscalrevenue decentralization (rd) is as follows

rd

=

PFRcity

PFRcity + PFRprovince + PFRnation[1 minus

GDPcity

GDPnation]

(7)

In (7) PFRcity PFRprovince and PFRnation respectivelyrepresent per capita fiscal revenue of each city the provincialper capita fiscal revenue and the national per capita fiscalrevenue GDPcity and GDPnation respectively represent theGDP in each city and the national GDP

413 Controlled Variable Controlled variables are as follows(1) Regional economic development (119901119892119889119901) regional

economy provides the necessary material and nonmaterialsupport for the financial development and innovation thehigher the level of regional economic development is themore beneficial to financial efficiency the improvement isIn this paper per capita actual GDP is used as a proxyvariable for regional economic development To exclude theinfluence of price factor the foregoing data that involve valueare adjusted in accordance with constant 2003 prices

(2) Urbanization (119906119903119887119886119899) Urbanization of population(the proportion of nonagricultural population accounts fortotal population) is used as the proxy variable in which thedata of nonagricultural population covering the period from2003 to 2008 is from ldquoChina Urban Statistical Yearbookrdquo(2004ndash2009) and the data of nonagricultural populationcovering the period from 2009 to 2012 is from the annualldquoStatistics on Population of Counties and Cities in Chinardquo

(3) Industrial structure (119894119905119906) the upgrading and adjust-ment of regional industrial structure constitutes a soft envi-ronment for financial development and innovation Thehigher the proportion of the third industry is the morebeneficial to financial efficiency the improvement is In thispaper the index of industrial structure upgrading is usedto represent industrial structure and the specific measureformula is as follows

itu =

119899

sum

119895=1

119902 (119895) lowast 119895 (8)

In (8) 119902(119895) is the proportion of the third industry accountsfor local GDP in region 119895 119899 = 3 and the value of index rangesfrom 1 to 3

(4) Information technology development (119894119899119905119890119903119899119890119905) it ismeasured by the number of internet users in each city

(5) Fixed asset investment (119894119891119886) it is measured by theactual fixed asset investment in each city

(6)The level of internal opening (119901119903119894V119886119905119890) with referenceto the research result of Fan et al [35] we use the developmentof private enterprises to measure the level of internal openingand assume that the higher the level of the developmentof private enterprise is the stronger the flexibility of localfinancial innovation activities is which is more conduciveto financial efficiency improvement The proportion of theemployees of private enterprises and self-employed employ-ees account for the local total population is used as a proxy to

Discrete Dynamics in Nature and Society 7

measure the development of private enterprises in differentregions

414 Spatial Weight Matrixes Strategic competition tendsto occur in the regions with the similar level of economicdevelopment simultaneously the concentration effect willgradually decay with the geographical distance [36] In orderto comprehensively reflect the spatial effects of geographicaland economic characteristics and accurately measure thespatial correlation among regions we will construct themixed geography-economy weighting matrix The specificmeasure formula is as follows

119882eco-geo =

10038161003816100381610038161003816119876119894minus 119876119895

10038161003816100381610038161003816

119889119894119895

2119894 = 119895

0 119894 = 119895

(9)

In (9) 119882eco-geo is mixed geography-economy weightingmatrix119876

119894and119876

119895 respectively represent the per capita actual

GDP of city 119894 and city 119895 (the mean of the actual GDP coveringthe period from 2003 to 2012) and 119889

119894119895is the geographical

distance between city 119894 and city 119895 calculated with the latitudeand longitude data of prefecture-level cityrsquos administrativeunit The specific measure formula is as follows

119889119894119895

= 119877 arccos (sin120593119860sin120593119861

+ cos120593119860cos120593119861cos (120582

119861minus 120582119860))

(10)

In (10) 119877 is the earth circle radius (6378 km) and 120582 and120593 respectively represent administrative centerrsquos longitudeand latitude of city 119894 and city 119895 Longitude and latitude dataare from the national fundamental geographic informationsystem All of the diagonal elements of matrix 119882eco-geo are0 And the weight matrix is normalized so that the sum ofline elements is 1 In addition considering the reliability androbustness of statistical analysis we will also construct geo-graphical distance weight matrix tomake a spatial economet-ric analysis and then compare the corresponding estimatedresultsThe specificmeasure formula of geographical distanceweight matrix is as follows

119882geo =

1

119889119894119895

2119894 = 119895

0 119894 = 119895

(11)

In (11) 119882geo is a geographical distance weighting matrixwhich is also normalized so that the sum of line elements is 1and other indexes are consistent with that of (10)

42 Data Sources and Description Considering the avail-ability of data and consistency of research samples weeventually choose 281 prefecture-level cities (Hong KongMacao and Taiwan are excluded) as samples and excludeBazhong Ziyang Lhasa Jinchang Baiyin Zhongwei andother cities due to a serious lack of available dataThe researchperiod is covering the period from 2003 to 2012 and thedata is from ldquoChina City Statistical Yearbookrdquo ldquoRegional

Economic China Statistical Yearbookrdquo and provincial statis-tical yearbooks covering the period from 2004 to 2013 Themissing data was partially filled by interpolation In orderto weaken the influence of heteroskedasticity outliers andanomaly terms on the data stability all the data are processedby the logarithm An overview of the descriptive statistics ofthese data is provided in Table 2

In order to avoid the deviation of the model settingand spurious regression we carry on unit root testing onrelated variables by LLC and Fisher-ADF testing methodsbefore spatial econometric regression analysis The resultsof stationary testing of panel data indicate that all variablesreject the null hypothesis of unit root which passes the testat the 5 significance level Moreover the panel data arestationary and the regression results will be significantlyreliable

5 Empirical Methodology

51 Spatial Econometric Analysis of Fiscal Decentralization onRegional Financial Efficiency Since the SDMmodel containsthe spatial lag item of the dependent variable and the spatiallag items of independent variables at the same time so itcanmore fully reflect the impact of spatial autocorrelation onthe regression results than SEM model and SLM model [32]Therefore we construct SDMmodels to test the spatial effectsof the fiscal revenue decentralization and fiscal expendituredecentralization on the regional financial efficiency The 119875

values of Hausman test of the spatial econometric models areless than 005 which indicates that the estimated results offixed effect estimation are better than that of random effectestimation Therefore we will specifically carry out analysisaccording to the estimated results of spatially fixed effectmodels in the following parts Meanwhile a robust test iscarried by constructing DSDM models It is found that theestimated coefficients of SDMmodel and DSDMmodel havethe same direction while the values of estimated coefficientsand level of significance are also very close indicating that theresults of SDM model are significantly reliable Comparingthe values of 119877

2 and Log-likelihood of SDM model andDSDM model we find that the fitting degree of SDM modelis higher than that of DSDMmodel so the following analysisis mainly based on the econometric results of SDMmodel

As shown in Table 3 the overall spatial correlationcoefficients 120588 of four models are positive and pass thetests at the 1 significance level indicating that there arepositive spatial dependent relationships between the localfinancial efficiency and the neighboring regionsrsquo financialefficiency that is the economic activities of geographyneighborhood regions or regions with similar level of eco-nomic development have obvious spillover effect on localfinancial efficiency thereby causing imitation and strate-gic competition of financial development and innovationamong regions On one hand the geographical adjacencyand similarity of economic development in different regionsbring huge convenience to cooperation between financialenterprises sharing of infrastructure information exchangeand communication innovation and diffusion of knowledge

8 Discrete Dynamics in Nature and Society

Table 2 Statistical description of samplersquos variables

Index Variables Mean Standarddeviation Minimum 25 quantile 50 quantile 75 quantile Maximum

Finance efficiency fe 0576 0624 0133 0380 0500 0653 3958Fiscal expendituredecentralization sd 0368 0108 0138 0294 0351 0421 0888

Fiscal revenuedecentralization rd 0242 0146 0037 0133 0202 0318 0871

Per capita real GDP(yuanperson) pgdp 27119890 + 04 32119890 + 04 1891200 9837380 17119890 + 04 31119890 + 04 44119890 + 05

Urbanization () Urban 0345 0195 0013 0202 0296 0451 100000Industrial structure itu 2203 0136 1318 2110 2190 2283 2797Informationtechnologydevelopment (tenthousand families)

Internet 41066 128506 0024 8316 17139 36800 5174000

Fixed assetinvestment (tenthousand yuan)

ifa 15119890 + 04 15119890 + 04 654299 4606420 9841960 20119890 + 04 17119890 + 05

The level of internalopening () Private 0086 0106 0003 0036 0059 0096 100000

and technology flowing of financial resources among regionswhich causes regional financial development and innovationand imitation and strategic competition among regions [37]on the other hand the restriction of transportation cost onfinancial sector is not strong and the rapid development ofinternet technology leads to cross-regional financial transac-tions and innovation thus further expanding the clusteringeffect effect of economic scale and spatial spillover effectsof financial development The coefficients of lag item (fe

119894119905-1)of financial efficiency in DSDM models are positive andthey are respectively 0211 and 0212 both of them pass thetests at the 1 significance level indicating that the previousinputs and outputs of production will be present through thefinancial efficiency increasing in later stage (or later period)and financial efficiency increasing is a dynamic systemicprocess The specific impact of each variable on financialefficiency is analyzed as follows

The estimated coefficients of Lnsd and Lnrd are positiveand all of them pass the tests at the 1 significance levelwhich indicates that the fiscal expenditure and revenuedecentralization play a significant positive role in promotinglocal financial efficiency Under the dual pressures of makingup the fiscal gap and political promotion the local govern-ments regarding economic development as main missionhave instinctive strong interest in scarce flowing economicfactors [38] As the excellent combination of capital talenttechnology management and marketing financial resourceswill be the important target for battling among local govern-ments With the expansion of fiscal expenditure decentral-ization and revenue decentralization local government usesmany policy tools such as fiscal subsidies tax concessionsand government procurement to compete for getting thefinancial resources by attracting financial capital humancapital and technological innovation to flow into the localregion thus contributing to the financial innovation and

improvement of financial efficiencyThe influence coefficientof fiscal expenditure decentralization is 0357 greater thanthe influence coefficient (0153) of fiscal revenue decentral-ization indicating that the impacts of fiscal expendituredecentralization and revenue decentralization on regionalfinancial efficiency have unbalanced characteristics and thepossible reason is that local governments under provinceget very limited taxation autonomy resulting in the factthat revenue decentralization is often lower than expendi-ture decentralization In addition under the capital-driveneconomic growth model the flowing direction of capitaldepends on the level of infrastructure [39] thus the inflowof financial resources as well as financial innovation oftenoccurs in the regions with rapid economic developmentexcellent investment environment and great consumptionability so that local governments have to increase fiscalexpenditure to enhance the supply of the associated support-ing infrastructure Comparatively speaking tax preferenceof undeveloped regions is less attractive than the superiorityconditions of developed regions

We will further analyze the estimated coefficients andsignificance levels of spatial lags of fiscal expenditure andrevenue decentralization and investigate the strategic com-petitive effects of financial efficiency among neighboringregions The spatial lag of expenditure decentralization(WLnsd) does not pass the significance test suggestingthat spatial spillover effect of fiscal expenditure decentral-ization is uncertain and needs further examination whilethe spatial lag of fiscal revenue decentralization (WLnrd)is negative at the 5 significance level and the estimatedcoefficient is minus0228 indicating that with the increase offiscal revenue decentralization the negative externalities onfinancial development and innovation will be increasinglyprominent thus inhibiting the improvement of financialefficiency in neighboring regions The reason may be that

Discrete Dynamics in Nature and Society 9

Table 3 Estimated results of regional financial efficiency under theSpatial Durbin Model (economic-geographic weighting matrix)

Independentvariables

Dependent variable sd Dependent variable rd

SDMmodel DSDMmodel SDMmodel DSDM

model

feitminus10211lowastlowastlowast(1283)

0212lowastlowastlowast(1283)

Lnsd 0357lowastlowastlowast(626)

0334lowastlowastlowast(570)

Lnrd 0153lowastlowastlowast(470)

0143lowastlowastlowast(408)

Lnpgdp 0169lowastlowastlowast(462)

0120lowastlowastlowast(308)

0138lowastlowastlowast(360)

0099lowastlowast(244)

Lnurban minus0050lowastlowast(minus237)

minus0044lowastlowast(minus209)

minus0050lowastlowast(minus237)

minus0045lowastlowast(minus215)

Lnitu 0433lowastlowast(227)

0438lowastlowast(215)

0525lowastlowastlowast(273)

0531lowastlowastlowast(259)

Lninternet minus0295lowastlowastlowast(minus2970)

minus0299lowastlowastlowast(minus2716)

minus0292lowastlowastlowast(minus2945)

minus0297lowastlowastlowast(minus2690)

Lnifa minus0446lowastlowastlowast(minus2298)

minus0386lowastlowastlowast(minus1832)

minus0437lowastlowastlowast(minus2321)

minus0374lowastlowastlowast(minus1806)

Lnprivate 0038lowastlowastlowast(322)

0037lowastlowastlowast(290)

0039lowastlowastlowast(325)

0038lowastlowastlowast(302)

WLnsd 0179(099)

minus0041(minus022)

WLnrd minus0228lowastlowast(minus251)

minus0186lowast(minus179)

WLnpgdp 0016(027)

0064(101)

0058(096)

0097(150)

WLnurban 0097lowastlowast(236)

0082lowastlowast(198)

0107lowastlowastlowast(259)

0089lowastlowast(213)

WLnitu 1159lowast(193)

0516(082)

0615(100)

0228(035)

WLninternet 0177lowastlowastlowast(647)

0181lowastlowastlowast(622)

0180lowastlowastlowast(666)

0180lowastlowastlowast(619)

WLnifa 0237lowastlowastlowast(539)

0251lowastlowastlowast(532)

0275lowastlowastlowast(704)

0260lowastlowastlowast(600)

WLnprivate 0065lowast(185)

0045(115)

0083lowastlowast(232)

0060(153)

120588

0420lowastlowastlowast(1165)

0462lowastlowastlowast(1251)

0429lowastlowastlowast(1205)

0463lowastlowastlowast(1266)

1198772 0448 0443 0441 0438

log 119871 599812 595228 595228 587624Observations 2810 2529 2810 2529Note the values in parentheses represent the 119905-statistics values of thecorresponding estimated coefficientslowastlowastlowastlowastlowast andlowast respectively representsignificance level of 1 5 and 10

local governmentsrsquo motivation of pursuing revenues underthe fiscal revenue decentralization system is not conduciveto free flowing of financial resources among regions whichdirectly affects the financial efficiency in other regions

When it comes to the estimated coefficients of othercontrolled variables we found that not all the spatial spillovereffects of variables are significant Overall various economicactivities have direct and indirect influence on financial

efficiency and promote the strategic financial competitionamong regions The specific effects are as follows the levelof open-into-inside can not only improve the local financialefficiency significantly but also increase the financial effi-ciency correspondingly in neighboring regions The strongdemand for various financial products and services of privateenterprises promotes financial innovation and improves theefficiency of resource allocation in financial markets Onthe contrary urbanization information technology develop-ment and fixed assets investment suppress the improvementof local financial efficiency but play a positive role in promot-ing the financial efficiency in neighboring regionsThe accel-erating of urbanization process the application of internettechnology and investment in fixed assets promote economicexchanges and cooperation among regions so that the spatialspillover effects of financial development and innovation areincreasingly significant Furthermore although the regionaleconomic development and industrial structure significantlycontribute to the improvement of regional financial efficiencyin general the spillover effects among regions are not obvious

52 Decomposition of Fiscal Decentralizationrsquos Effects onRegional Financial Efficiency Whether there truly existspillover effects among the variables relying solely on the spa-tial lag coefficients of SDMmodel and DSDMmodel is likelyto lead to misinterpretation of the estimated results Spatialeffects can be further decomposed into direct effect indirecteffect (spillover effects) and total effect with reference to theresearch result of LeSage and Pace [32] The indirect effectsrepresent the independent variablesrsquo influence on the depen-dent variable in all other regions through spatial interactionsTherefore we will further verify the existence of spillovereffects by using partial differential matrix analysis Table 4illustrates the direct effect indirect effect and total effect ofthe fiscal expenditure and revenue decentralization in SDMmodel The results show that the estimated coefficients ofindependent variablesrsquo direct effect have the same directionwith the estimated coefficients of SDMmodel in Table 3 andthe estimated coefficients and significance levels are also veryclose

According to the estimated results of SDM model inTable 4 the direct effect indirect effect and total effectof fiscal expenditure decentralization on regional financialefficiency are positive and pass the 10 significance leveltest The regression coefficient of direct effect is 0368 theregression coefficient of indirect effect is 0560 and theregression coefficient of total effect is 0928 indicating thatthe degree of fiscal expenditure decentralization increases by1 while regional financial efficiency increases by 0928Fiscal expenditure decentralization plays a significant posi-tive impact on financial development and financial efficiencywhich is opposite to the research result of Wang and Li [23]The opposite conclusions are possibly due to the differencesof sample selection and indexesThe rise of fiscal expendituredecentralization in the local region leads to an increaseby 0368 of the financial efficiency while the increase ofexpenditure decentralization in the neighboring region leadsto financial efficiency increasing by 0560 indicating that

10 Discrete Dynamics in Nature and Society

Table 4 Decomposition of direct effect indirect effect and total effect under SDMmodel

Independent variablesDependent variable sd Dependent variable sd

Direct effect Indirect effect Total effect Direct effect Indirect effect Total effect

Lnsd 0368lowastlowastlowast(750)

0560lowast(195)

0928lowastlowastlowast(310)

Lnrd 0146lowastlowastlowast(529)

minus0283lowast(minus195)

minus0137lowast(minus192)

Lnpgdp 0175lowastlowastlowast(435)

0132(153)

0307lowastlowastlowast(359)

0145lowastlowastlowast(346)

0187lowastlowast(218)

0332lowastlowastlowast(392)

Lnurban minus0046lowastlowast(minus207)

0127lowastlowast(200)

0081(130)

minus0045lowastlowast(minus204)

0146lowastlowast(226)

0101(158)

Lnitu 0486lowastlowastlowast(275)

2321lowastlowast(237)

2807lowastlowastlowast(292)

0559lowastlowastlowast(315)

1488(151)

2047lowastlowast(211)

Lninternet minus0291lowastlowastlowast(minus3055)

0086lowast(190)

minus0205lowastlowastlowast(minus432)

minus0288lowastlowastlowast(minus3029)

0091lowastlowast(200)

minus0197lowastlowastlowast(minus410)

Lnifa minus0441lowastlowastlowast(minus2260)

0087(149)

minus0353lowastlowastlowast(minus594)

minus0430lowastlowastlowast(minus2254)

0154lowastlowastlowast(283)

minus0276lowastlowastlowast(minus494)

Lnprivate 0041lowastlowastlowast(319)

0134lowastlowast(235)

0175lowastlowastlowast(285)

0042lowastlowastlowast(327)

0166lowastlowastlowast(283)

0208lowastlowastlowast(330)

Note lowast lowast lowast lowastlowast and lowast respectively represent the significant level of 1 5 and 10

fiscal expenditure decentralization significantly promotesthe competition of financial development and innovationamong regions The expansion of the autonomy of localgovernment expenditure strengthens the regulatory capacityof local governments which is conducive to acceleratingmarket-oriented reform processing and achieves the organicintegration between fiscal system and market mechanism inthe allocation of financial resources which is beneficial tocreating a favorable institutional environment is also con-ducive to the flowing of financial resources among regionsand greatly promotes financial efficiency

The fiscal revenue decentralizationrsquos direct effect onregional financial efficiency is significantly positive whilethe spillover effect is significantly negative indicating thatthe impact of fiscal revenue decentralization on regionalfinancial efficiency presents the feature of ldquodouble-edged-swordrdquo Fiscal revenue decentralization has a significant posi-tive impact on the local financial efficiencyHowever there aresignificant negative spillover effects on the financial efficiencyin other regions The regression coefficient of the directeffect is 0146 the regression coefficient of the indirect effectis minus0283 and the regression coefficient of the total effectis minus0137 which indicates that the degree of fiscal revenuedecentralization increasing by 1 will lead to decrease by0137 of regional financial efficiency Furthermore whilethe increase of local fiscal revenue decentralization leads tofinancial efficiencyrsquos increase by 0146 the increase of thefiscal revenue decentralization in neighboring regions resultsin the financial efficiencyrsquos decrease by 0283 Comparingthe estimated coefficient of direct effect and that of indirecteffect it can be found that the negative spillover effectsof fiscal revenue decentralization are significantly greaterthan the direct scale effect so the negative externalitiesof fiscal revenue decentralization on financial efficiencyare greater than the positive externality On one hand inorder to obtain high-quality financial resources the local

governments ignore the differences like the level of economicdevelopment factor endowments and other aspects amongregions and seek to imitate policies and instruments inother regions thus leading to chain reaction of absorbinginvestment by other regions On the other hand whenthe neighboring regions or the regions with similar levelof economic development take some measures to attractliquidity factors the local government driven by yardstickcompetitionwill adopt similar policies to avoid the outflow ofelements which results in convergence of competitive tactics[40 41] Tax incentives reduction of fee and other wayswill become important means of competition and imitationbetween local governments in different regions on the basisof the independent revenue authority However such kind ofcompetitive mode between local governments for resourceswill consume huge fiscal resources and weaken the capacityof local governmentrsquos provision of public goods and servicesUltimately it is harmful to provide healthy stable andsustainable economic development environment for financialinnovation and increase of financial efficiency

53 Robustness Test To ensure the robustness and reliabilityof research results we will calculate the spatial lags withthe geographical distance weight matrix and do econometricregression by taking financial efficiency as the dependentvariable so that we can comparatively analyze the differenceof influence with different spatial weight setting based on theabove The robustness test results are as in Table 5

As shown in Table 5 the estimated results indicate thatthe sign of regression coefficients of the fiscal expendituredecentralization and revenue decentralization is consistentwith that of corresponding variable in the model of Table 3and the regression results in themodel are basically consistentwith the conclusions in Table 3 Comparing the estimatedresults based on two different weightmatrixes it is found thatthere are some differences in the influence coefficients of two

Discrete Dynamics in Nature and Society 11

Table 5 Robustness test on Spatial Durbin Models (geographicaldistance weight matrix)

Independentvariables

Dependent variable sd Dependent variable rdSDMmodel

DSDMmodel

SDMmodel

DSDMmodel

feitminus1 mdash 0202lowastlowastlowast(1276) mdash 0200lowastlowastlowast

(1255)

Lnsd or Lnrd 0364lowastlowastlowast(636)

0348lowastlowastlowast(592)

0173lowastlowastlowast(522)

0161lowastlowastlowast(458)

Lnpgdp 0268lowastlowastlowast(735)

0225lowastlowastlowast(578)

0242lowastlowastlowast(637)

0209lowastlowastlowast(522)

Lnurban minus0026(minus112)

minus0026(minus112)

minus0028(minus120)

minus0027(minus115)

Lnitu 0465lowastlowast(246)

0478lowastlowast(238)

0527lowastlowastlowast(278)

0542lowastlowastlowast(269)

Lninternet minus0295lowastlowastlowast(minus3038)

minus0300lowastlowastlowast(minus2799)

minus0292lowastlowastlowast(minus3005)

minus0298lowastlowastlowast(minus2771)

Lnifa minus0484lowastlowastlowast(minus2479)

minus0427lowastlowastlowast(minus2024)

minus0472lowastlowastlowast(minus2448)

minus0417lowastlowastlowast(minus1988)

Lnprivate 0036lowastlowastlowast(312)

0035lowastlowastlowast(284)

0035lowastlowastlowast(307)

0035lowastlowastlowast(288)

WLnsd orWLnrd

minus0264(minus149)

minus0418lowastlowast(minus225)

minus0277lowastlowastlowast(minus370)

minus0241lowastlowastlowast(minus276)

WLnpgdp minus0269lowastlowastlowast(minus427)

minus0286lowastlowastlowast(minus428)

minus0237lowastlowastlowast(minus369)

minus0267lowastlowastlowast(minus394)

WLnurban 0053(147)

0059lowast(167)

0059(163)

0062lowast(173)

WLnitu 0104(020)

minus0085(minus016)

minus0205(minus039)

minus0290(minus051)

WLninternet 0217lowastlowastlowast(838)

0239lowastlowastlowast(853)

0214lowastlowastlowast(831)

0235lowastlowastlowast(841)

WLnifa 0434lowastlowastlowast(1052)

0464lowastlowastlowast(1054)

0433lowastlowastlowast(1165)

0450lowastlowastlowast(1129)

WLnprivate 0062lowast(187)

0038(106)

0081lowastlowast(241)

0050(139)

120588

0547lowastlowastlowast(1858)

0580lowastlowastlowast(1938)

0541lowastlowastlowast(1839)

0570lowastlowastlowast(1898)

1198772 0444 0428 0443 0430

log 119871 659741 665804 665804 659509Observations 2810 2529 2810 2529Note lowastlowastlowast lowastlowast and lowast respectively represent the significant level of 1 5and 10

kinds of fiscal decentralizations indicators which show thatthe gap between the levels of economic developmentwill havean impact on role of fiscal decentralization acting on regionalfinancial efficiency On one hand under the fiscal decen-tralization system because of the heterogeneity of regionaleconomic development level the local governmentsrsquo ability ofmaking up the gap between fiscal revenue and expenditure isdifferent and the fiscal pressure is also different therefore thedegree of interference to the regional market environment bythe local government is different which leads to the regionaldifferences in financial efficiency in a certain degree [21]On the other hand the occurrence of financial innovationdepends on the support of the economic foundation and

the different economic and social environment will lead tothe fact that the financial innovation activities are more andmore concentrated in the regions with the advantages offinancial location resulting in nonequilibrium of financialefficiency in different regions [42] Under the performanceassessment method focusing on GDP local governments willnot only consider the strategic behaviors of the neighboringregions but also tend to think about the behaviors of theregions with the similar level of economy In additioncross-regional knowledge diffusion technology spillover andhuman capital flow are more prone to occur among theregions with similar level of economic development so thatthe financial exchanges and contacts among regions are morefrequent It also shows that besides considering the impact ofgeographical distance the economic distance factor shouldbe taken into consideration while we are researching thespatial effects of fiscal decentralization on regional financialinnovation in addition to the consideration of the impact ofgeographical distance the economic distance factor shouldbe considered

6 Conclusions and Enlightenments

Based on the panel data of 281 prefecture-level cities in Chinacovering the period of 2003ndash2012 the regional financialefficiency is measured by the superefficiency SBMmodel thegeographic distance weight matrix and economic-geographymixed weight matrix which can reflect the regional similarityare introduced into the SDM and the spatial effects of thefiscal expenditure and revenue decentralization on regionalfinancial efficiency under province are investigated as wellas the strategic financial competition arising from it Basedon the results of previous empirical analysis we can draw thefollowing basic conclusions

First the geographical distribution of Chinarsquos prefecture-level citiesrsquo financial efficiency presents significant features oftime inertia and spatial dependency Financial developmentand innovation among regions particularly neighboringregions is not independent but shows significant spatialspillover effects The financial efficiency of the local regionwill be significantly affected by the financial efficiency ofadjacent regions The spatial effect is increasingly obviousin the process of financial development and innovation andthe size of the spillover effects is closely related to spatialweight (which is used to measure regional similarities) andmainly shows that there is competition for financial efficiencyamong regions with similar economic attributes and adjacentregions

Second it cannot be ignored that the institutional factorsespecially fiscal decentralization have effect on the financialefficiency and the factors have a decisive role in the allocationof financial resource and the regional distribution of financialinnovation The improvement of fiscal decentralization levelwill help to attract more financial resources thereby promot-ing an increase in financial efficiencyThe improvement of thefinancial autonomy in neighboring jurisdictions will reducethe financial resources flowing into the local region andsuppresses the occurrence of the local financial innovationIn addition there exist significant spatial spillover effects of

12 Discrete Dynamics in Nature and Society

financial development and efficiency caused by fiscal decen-tralization The scope of spatial spillovers is mainly reflectedin two aspects Thus the local governmentsrsquo competition notonly strongly increases the financial efficiency of the regionbut also accelerates interregional competition for financialefficiency

Third fiscal expenditure decentralization and revenuedecentralizationrsquos impact on regional financial efficiencyhave nonequilibrium characteristics First of all there existsasymmetry between fiscal expenditure decentralization andrevenue decentralizationrsquos impact on regional financial effi-ciency Fiscal expenditure decentralization can promoteoptimization of the fiscal expenditure structure and fiscalrevenue decentralization can increase the local governmentsrsquorevenue share so as to provide positive incentives [43] bothof them have a positive and promoting effect on the improve-ment of local financial efficiency and such effect largelydepends on the symmetry between the fiscal expenditure andrevenue decentralization If it is asymmetric between theirproperty rights and the expenditure responsibilities underfiscal decentralization system the local governments maymake excessive intervention to regional financial resourcesallocation tomeet their own demands for financial resourcesNext there exists nonconformance between fiscal expendi-ture decentralization and revenue decentralizationrsquos impacton financial efficiency in neighboring regions Comparedwith the fiscal expenditure decentralization the fiscal revenuedecentralization has more apparent inhibitory effect on thefinancial efficiency in neighboring regions

Our research conclusions have important policy implica-tions that illuminate the relationship between fiscal decen-tralization and financial efficiency

(1) The regional financial efficiency is closely relatedto the location and the economic development of the sur-rounding cities The regional financial efficiency is affectednot only by the financial development and innovation inneighboring regions but also by the impact of the struc-tural differences among regions In this case breaking theadministrativemonopoly of local protectionism and carryingon the ldquorichening neighborrdquo policy of financial developmentare the best choices for the local government to developthe regional economy When making policies of financialdevelopment local governments need to consider their owneconomic development environment and economic featuresin neighboring regions and strengthen communication andcoordination with the surrounding regions so as to avoidthe possible contradictions and conflicts on financial devel-opment policies among regions

(2) The spatial spillover effects of the regional financialefficiency should be divided into two kinds of effects includ-ing global geographic spillover effects and local geographicspillover effects In order to promote the increase of financialefficiency in a better way it should also be based on thehierarchy of spatial effect of and the degree of spillover effectto define the relationship between the government and themarket and central and local governments and among thelocal governments and their role in regional financial devel-opment and innovation so as to better promote financialefficiency

(3) There is a two-way interactive mechanism betweenthe fiscal system and financial development such effectsmay affect the behaviors of the local governments and theallocation of resources and ultimately act on macroeconomicperformance [10] Under the fiscal decentralization systemthe strategic competitive behaviors among local governmentssignificantly affect the liquidity of financial resources and theoccurrence of financial innovation the local governmentsrsquoexpansion of fiscal expenditure and revenue shortfall becomethe main driving force of each region competing for financialresources In order to better play the role of incentivesand constraints of fiscal competition mechanism under thefiscal decentralization system and avoid the negative impactbrought by the vicious interregional strategic financial com-petitive behaviors on the regional economic development thelocal governmentsrsquo short-sighted behaviors should be strictlyregulated and then the interregional financial cooperationwith mutual benefit as the core should be actively promoted

(4) Constructing a scientific and rational framework forfiscal decentralization will be an important force to accelerateeffectively the improvement of regional financial efficiencyChinarsquos future subprovincial fiscal decentralization reformshould adapt to the new normal financial development andfurther rationalize the provincial and subprovincial fiscaldecentralization system to ensure a balanced and appropriatefiscal expenditure and revenue decentralization On onehand we should weaken local governmentsrsquo negative inter-vention in financial system by reducing local governmentsrsquofinancial pressure caused by the asymmetry between thefiscal resources and expenditure responsibilities On the otherhandwe should also actively play the function of fiscal systemto adjust the financial development and innovation and fullyplay public financersquos role in promoting the flowing of finan-cial resources among regions the balanced development ofregional finance and the improvement of financial efficiency

Competing Interests

There are no competing interests in regard to the contentdiscussed in this paper

Authorsrsquo Contributions

All authors conceived and designed the research XiaomeiHucollected and processed the data Jianmin Liu analyzed thedata and interpreted the results all authors were also involvedin writing the paper and have all approved the submittedform

Acknowledgments

The research of this paper is jointly funded by the NationalSocial Science Foundation of China under Project no14BJY159 and the Hunan Philosophy and Social ScienceFund Project under Grant no 13JD09 The authors gratefullyacknowledge the support of Jinguang Wu from HunanUniversity of Finance and Economics China

Discrete Dynamics in Nature and Society 13

References

[1] J Zhang and X Jin ldquoRe-examine of the relationship betweenChinarsquos financial deepening and productivity 1987ndash2001rdquo Eco-nomic Research Journal vol 11 pp 35ndash43 2005

[2] K Chen A L Hillman and Q Y Gu ldquoFiscal centralization andchanges of local governmentsrsquo behaviorsmdashfrom lsquohands for helprsquoto lsquohands for grabrsquordquoChina EconomicQuarterly vol 1 pp 111ndash1302002

[3] L Zhou and Z M Wang ldquoEmpirical analysis of regionalfinancial development of China and economic growth 1978ndash2000rdquo Financial Research vol 10 pp 1ndash13 2002

[4] L Zhou ldquolsquoSecond financersquo of Chinarsquos financial industry andfinancial division during the reforming periodrdquo The Journal ofWorld Economy vol 6 pp 72ndash79 2003

[5] L Zhou ldquoGradual transition state capacity and fiscal functionof financerdquo Journal of Finance and Economics vol 2 pp 26ndash372005

[6] E L He and W Liu ldquoFinancial capital local governmentsrsquointervention and industrial structuremdashan empirical analysisbased on the data of central regionsrdquo Economic Issues vol 5pp 37ndash40 2011

[7] X L Dou and W J Wang ldquoThe differences and trends of Chi-nese regional financial development and local governmentsrsquobehavior and analysis of tax sharing reform on the differencesand trends of Chinese regional financial developmentrdquo Shang-hai Finance vol 2 pp 9ndash15 2011

[8] J Zhang and K R Shen ldquoLocal government interventionregional financial development and the transformation ofChinarsquos economic growth modemdashan empirical study under thefiscal decentralization systemrdquo Nankai Economic Studies vol 6pp 122ndash141 2008

[9] G Chen X G Yin and Y Pan ldquoChinarsquos financial developmenttax sharing reform and economic growthrdquo Financial Researchvol 2 pp 99ndash109 2006

[10] P P Ding and Y Fu ldquoBehaviors of local governments therelationship of fiscal and finance and macroeconomic volatilityassociated with Chinamdashan analysis under Chinese-style fiscaldecentralization systemrdquo Comparative Economic amp Social Sys-tems vol 6 pp 87ndash97 2012

[11] X G Yin G Chen and Y Pan ldquoTax sharing reform local gov-ernmentsrsquo intervention and financial development efficiencyrdquoJournal of Finance and Economics vol 10 pp 92ndash101 2006

[12] M G Yu and H B Pan ldquoGovernment intervention the ruleof law financial development and bank loans of state-ownedenterprisesrdquo Financial Research vol 9 pp 1ndash23 2008

[13] D X Wang J Liu and L Li ldquoFiscal decentralization bankcredit and total factor productivityrdquo Journal of Finance andEconomics vol 4 pp 69ndash80 2011

[14] Y X Jiang and R B Jiang ldquoRegional financial innovationefficiency evaluation environmental impact and gap analysisrdquoJournal of Zhejiang University (Humanities and Social Sciences)vol 43 no 4 pp 52ndash65 2013

[15] H W Yang L W Cheng Y Xu and Y W Qi ldquoMeasure andinfluencing factors of financial resources allocation efficiencyunder fiscal decentralization system based on super-efficiencyDEA-TOBIT two-step methodrdquo Journal of Jiangxi NormalUniversity (Natural Sciences Edition) vol 6 pp 587ndash592 2014

[16] N Shen F Z Liu and J Q Zhao ldquoFiscal decentralizationfinancial deepening and regional international poverty prone todeveloprdquo Finance amp Trade Economics vol 1 pp 41ndash46 2006

[17] Q Y Du G P Li and N Shen ldquoFinancial efficiency andoptimization of the regional industrial structure under thecurrent fiscal systemrdquo Xinjiang Social Science vol 4 pp 16ndash212006

[18] T Feng Y W Song and Y Lu ldquoFiscal decentralization localgovernmentsrsquo behavior and regional financial developmentrdquoJournal of Xirsquoan Jiaotong University vol 5 pp 20ndash27 2007

[19] Y A Zhou X J Ma and J Y Zhao ldquoGovernment behaviorsfinancial development and economic growthrdquo Henan SocialSciences vol 1 pp 55ndash60 2007

[20] C Hao ldquoDevelopment of financial intermediation and eco-nomic growth the Chinese experiencerdquo China EconomicReview vol 17 no 4 pp 347ndash362 2006

[21] J Zhang Fiscal Decentralization Regional Financial Devel-opment and Chinarsquos Economic Growth Performance NanjingUniversity Press Nanjing China 2011

[22] Y J Yao ldquoGovernment intervention the development of finan-cial intermediation and economic growthrdquo Journal of Financeand Economics vol 8 pp 52ndash58 2010

[23] X Wang and L Li ldquoFiscal decentralization and regional finan-cial development the theoretical and empirical analysis fromthe perspective of central governmentrdquo Public Finance Researchvol 4 pp 31ndash35 2012

[24] W L Zhang and C L Qin ldquoFinancial development fiscaldecentralization and regional economic differencesrdquo EconomicGeography vol 1 pp 583ndash588 2009

[25] Y Ma J Li and B Chen ldquoProvincial differences in Chinesefiscal decentralization financial development industrializationand economic growthrdquo EconomicTheory and Business Manage-ment vol 2 pp 5ndash19 2015

[26] Y X Jiang and R B Jiang ldquoThe spatial external effects ofregional financial innovation financial competition and finan-cial agglomerationrdquo Social Science Front vol 3 pp 79ndash84 2014

[27] K Tone ldquoA slacks-based measure of super-efficiency indata envelopment analysisrdquo European Journal of OperationalResearch vol 143 no 1 pp 32ndash41 2002

[28] Y Q Lu and D G Zhang ldquoChinarsquos regional financial efficiencyand differences of efficiency measurementrdquo Economic Geogra-phy vol 1 pp 96ndash101 2012

[29] A N Berger and D B Humphrey ldquoEfficiency of financialinstitutions international survey and directions for futureresearchrdquo European Journal of Operational Research vol 98 no2 pp 175ndash212 1997

[30] Y J Yao ldquoEmpirical analysis of financial development andurban-rural income gaprdquo Journal of Finance and Economics vol2 pp 49ndash59 2005

[31] Q Li J Song E Wang H Hu J Zhang and Y WangldquoEconomic growth and pollutant emissions in China a spatialeconometric analysisrdquo Stochastic Environmental Research andRisk Assessment vol 28 no 2 pp 429ndash442 2014

[32] J P LeSage and R K Pace Introduction to Spatial EconometricsCRCPress Taylor amp Francis Group Boca Raton Fla USA 2009

[33] Q W Guo and J X Jia ldquoFiscal decentralization organizationalstructure of local governments and the size of governmentexpenditurerdquoEconomic Research Journal vol 11 pp 59ndash87 2010httpwwwcnkicomcnArticleCJFDTotal-JJYJ201011008htm

[34] F Gong and X Lei ldquoThe measurement of Chinese-style fiscaldecentralizationrdquo Statistical Research vol 10 pp 47ndash55 2010

[35] G Fan X L Wang and H P Zhu Chinese Marketing IndexRegional Marketing Process in 2011 Economic Science PressBeijing China 2011

14 Discrete Dynamics in Nature and Society

[36] S S Rosenthal and W C Strange ldquoGeography industrialorganization and agglomerationrdquoTheReview of Economics andStatistics vol 85 no 2 pp 377ndash393 2003

[37] S Park Yoon International Banking and Financial CentersKluwer Academic Boston Mass USA 1989

[38] B Y Qiao J Y Fan and X Y Feng ldquoChinarsquos fiscal decentral-ization and compulsory primary educationrdquo Social Sciences inChina vol 6 pp 37ndash46 2005

[39] W JWang and C L Qin ldquoRegional differences in the behaviorsof local governments and growth effect of fiscal decentralizationbased on the analysis of judgment hypothesis and testingrdquoManagement World vol 1 pp 9ndash21 2008

[40] J K Brueckner ldquoStrategic interaction among governments anoverview of empirical studiesrdquo International Regional ScienceReview vol 26 no 2 pp 175ndash188 2003

[41] F Revelli ldquoOn spatial public finance empiricsrdquo International Taxand Public Finance vol 12 no 4 pp 475ndash492 2005

[42] N Shen and T Y He ldquoFinancial efficiency and regional eco-nomic growth from the perspective of fiscal decentralizationrdquoReform vol 1 pp 61ndash65 2006

[43] J M Liu X M Hu and J G Wu ldquoThe threshold effect ofsub-national fiscal decentralization revenue and expendituredecentralization on provincial industrial transformation andupgrading based on the data of 14 cities in Hunan ProvincerdquoPublic Finance Research vol 8 pp 49ndash52 2014

Submit your manuscripts athttpwwwhindawicom

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

MathematicsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Mathematical Problems in Engineering

Hindawi Publishing Corporationhttpwwwhindawicom

Differential EquationsInternational Journal of

Volume 2014

Applied MathematicsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Probability and StatisticsHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Mathematical PhysicsAdvances in

Complex AnalysisJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

OptimizationJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

CombinatoricsHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Operations ResearchAdvances in

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Function Spaces

Abstract and Applied AnalysisHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of Mathematics and Mathematical Sciences

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Algebra

Discrete Dynamics in Nature and Society

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Decision SciencesAdvances in

Discrete MathematicsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom

Volume 2014 Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Stochastic AnalysisInternational Journal of

Discrete Dynamics in Nature and Society 3

decentralization on regional financial efficiency in additionto establishing econometric models Section 4 addresses theselection of variables Section 5 analyzes the empirical resultsand engages in further comparative study Section 6 presentsconclusions

2 Measurement and Spatial Clustering ofFinancial Efficiency

Financial efficiency refers to the input-output ratio of thefinancial sector In order to pursue the goal of economicdevelopment local governments try to maximize financialefficiency via adjusting the allocation of financial resourcesThemainmeasuremethods of financial efficiency in previousresearch include regression analysis method factor analysismethod and the traditional DEAmethodThis paper will usesuperefficiency SBM model considering undesirable outputto measure regional financial efficiency Different from thetraditional radial DEA (Data Envelopment Analysis) modelSBMmodel directly introduces slack variables into the objec-tive function to calculate the efficiency and the economicimplication of SBM model is the maximization of actualprofit instead of the maximization of benefit ratio [27] SBMmodel assumes that there are 119899 decision-making units andeach decision-making unit is constituted by three elementsincluding input desirable output and undesirable outputThe vector forms of the elements are represented as follows119909 isin 119877

119898 119910119889

isin 1198771199031 and 119910

119906

isin 1198771199032 119883 119884119889 and 119884

119906 are matrixes119883 = [119909

1 x

119899] isin 119877

119898times119899 119884119889

= [119910119889

1 119910

119889

119899] isin 119877

1199041times119899 and

119884119906

= [119910119906

1 119910

119906

119899] isin 1198771199042times119899 where 119883 gt 0 119884119889 gt 0 and 119884

119906

gt 0Production possibility set 119875 under the condition of constantreturns to scale is defined as

119875 = (119909 119910119889

119910119906

) | 119909 ge 119883120582 119910119889

le 119884119889

120582 119910119906

ge 119884119906

120582 120582

ge 0

(1)

The SBM model based on the slack variable is expressedas follows

120588lowast

= min1 minus (1119898)sum

119898

119894=1(119908minus

1198941199091198940)

1 + (1 (1199041+ 1199042)) (sum1199041

119903=1(119908119889

119903119910119889

1199030) + sum1199042

119903=1(119908119906

119903119910119906

1199030))

st 1199090= 119883120582 + 119908

minus

119910119889

0= 119884119889

120582 minus 119908119889

119910119906

0= 119884119906

120582 + 119908119906

119908minus

ge 0

119908119889

ge 0

119908119906

ge 0

(2)

In (1) and (2) 1199041and 1199042 respectively represent the slack

of input and output 120588 is the evaluation criteria of efficiency119908minus and 119908

119906 respectively represent the excessive input andundesirable output 119908

119889 indicates the insufficient output 119898

represents the type of input factors and 120582 represents theweight vector Objective function 120588 is strictly decreasing

based on 119908minus 119908119906 and 119908

119889 For a particular evaluation unitwhen 119908

minus

= 0 119908119906 = 0 and 119908119889

= 0 that is 120588 = 1 DMU isefficient for SBM model When 120588 lt 1 it shows that DMU isinefficient for SBM with redundancy of input or deficiency ofoutputMeanwhile Tone [27] also proposed a superefficiencySBM model and the basic principle of superefficiency SBMmodel is excluding the evaluated DMU from the referenceset so that the efficiency value may be greater than 1 whichmade up the defect that all decision-making efficiency valuescannot be calculated

The regional financial efficiency is measured on the basisof the productionmethod regarding the financial firms as thesuppliers of financial products and services [28] In consid-eration of attainability accuracy and consistency of data wetake labor capital and technology and other factors as inputsWhen it comes to the output indexes we take deposits andloans as the desirable outputs while the income gap betweenurban and rural areas is regarded as the undesirable outputThe selection and interpretation of indexes including inputvariables and desirable and undesirable output variables areas follows

(1)The labor input of financial industry (119897119886119887119900119903) as humancapital and intellectual resources of the financial industrythere is no doubt that employed persons engaged in financialindustry will affect the development performance of financialindustry This index is measured by the proportion of theemployees in the financial sectors account for the regionaltotal population

(2) The capital input of financial industry (119888119886119901119894119905119886119897) thefixed asset investment is the main fund for the infrastructureconstruction project in financial industry Since it is difficultto obtain the statistics of fixed asset investment for prefecture-level cities we use a proxy variable measured by the propor-tion that the fixed asset investment accounted for in regionalGDP to represent the capital input of financial industry

(3) Information technology level (119894119899119905119890119903119899119890119905) the improve-ment of financial efficiency is dependent on the innovationand development of electronic finance mobile finance inter-net finance and other emerging financial methods whichdepend on the development of information technology Thisindex is represented by a proxy variable measured by thenumber of international internet users in cities

(4)The desirable output of deposit (119889119890119901119900119904119894119905)The regionalfinancial efficiency ismeasured on the basis of the productionmethod regarding the financial firms as the suppliers offinancial products and services [29] The production methodtakes deposit and loan as output while the input is operatingcost including labor capital and technology The index isrepresented by per capita deposit of financial institutionswhere per capita deposit of financial institutions = totaldeposit of financial institutionsregional population

(5) The desirable output of loan (119897119900119886119899) the loan offinancial institutions especially the banks is the main sourceof profits for financial institutions The index is representedby per capita loan of financial institutions where per capitaloan of financial institutions = total loan of financial institu-tionsregional population

(6)The undesirable output of income gap between urbanand rural (119894119892119886119901) The development of the financial industry

4 Discrete Dynamics in Nature and Society

0

1

2

3

0 05 1 15 2 25

k density fe2003

k density fe2007k density fe2012

x

Figure 1 Kernel density segment curve of financial efficiency

results in money flowing from low income groups to highincome groups due to the threshold and cost to obtain finan-cial services and the difference between regional resourcesendowment which will exacerbate income inequality espe-cially in developing countries With reference to the researchresults of Yao [30] we take urban-rural income ratio as aproxy variable where urban-rural income ratio = per capitadisposable income of urban residentsper capita net incomeof rural residents

Based on the input-output data of 281 prefecture-levelcities covering the period from 2003 to 2012 this paper mea-sures the regional financial efficiency by using MAXDEA64software The calculation results show that there exists greatdifference in regional financial efficiency and the averagefinancial efficiency is 0576 while the standard deviationis 0624 From a regional perspective the average financialefficiency in the eastern central and western regions isrespectively 0628 0503 and 0604 and the financial effi-ciency of the central region is lower than that of eastern andwestern regions In addition this paper makes Kernel densityestimation on regional financial efficiency in the three typicalyears (2003 2007 and 2012) by using nonparametric Gaussnormal Kernel density (nuclear density) segment curve Asshown in Figure 1 the horizontal axis represents the regionalfinancial efficiency value of prefecture-level cities while thevertical axis represents the corresponding Kernel densityvalues

First there shows tendency toward left deviation atregional financial efficiency in Chinarsquos prefecture-level citiesin the view of the specific distribution form which indicatesthat the financial efficiency in most regions are below thenational average level Second themain peak level of financialefficiency gradually declines and moves to the left devia-tion during the three typical years (2003 2007 and 2012)which indicates that the level of financial efficiency graduallydecreases while the gap between the financial efficiency ofdifferent regions is widening In addition Chinarsquos regionalfinancial efficiency distribution form gradually evolves frommultipeak trend in 2003 to the double-peak trend in 2012

Table 1 Moranrsquos 119868 index of regional financial efficiency

Year Geographic distanceweighting matrix

Mixed geography-economyweighting matrix

2003 0152lowastlowastlowast 0130lowastlowastlowast

2004 0094lowastlowastlowast 0061lowastlowastlowast

2005 0094lowastlowastlowast 0028lowast

2006 0135lowastlowastlowast 0092lowastlowastlowast

2007 0137lowastlowastlowast 0100lowastlowastlowast

2008 0075lowastlowastlowast 0057lowastlowastlowast

2009 0019lowastlowastlowast 0024lowastlowastlowast

2010 0141lowastlowastlowast 0088lowastlowastlowast

2011 0140lowastlowastlowast 0071lowastlowastlowast

2012 0149lowastlowastlowast 0052lowastlowastlowast

Note lowastlowastlowast and lowast respectively represent the significant level of 1 and 10

which indicates that the development trend of regionalfinancial efficiency shifts frommultipolarity mode to double-polarity mode

In order to investigate the spatial clustering pattern andthe evolution trend of financial efficiency in 281 prefecture-level cities covering the period from 2003 to 2012 it is neededto further calculate Moranrsquos 119868 index of regional financialefficiency by using global spatial autocorrelation methodMoranrsquos 119868 index is the correlation coefficient of observedvalues and spatial lagged variables and the value of indexranges fromminus1 to 1 If the regional financial efficiency presentsspatial positive correlation the value of Moranrsquos 119868 indexranges from 0 to 1 which indicates that the regions withsimilar observed values are collectively distributed in thespatial range The closer Moranrsquos 119868 index to 1 the strongerthe spatial positive correlation On the contrary the regionswith similar observed value are discretely distributed in thegeographical space The specific measure formula of globalMoranrsquos 119868 index is as follows [31]

Moranrsquos 119868 =

119899sum119899

119894=1sum119899

119895=1119908119894119895(119909119894minus 119909) (119909

119895minus 119909)

sum119899

119894sum119899

119895=1119908119894119895sum119899

119894=1(119909119894minus 119909)2

(3)

In (3) 119909119894and 119909

119895 respectively represent region 119894 and

region 119895 and 119899 represents the number of regions while 119908119894119895

represents the spatial weightmatrixWewillmeasure regionalfinancial efficiencywith two spatial weightmatrixes includinggeographic distance weighting matrix and mixed geography-economy weighting matrix in the view of robustness ofresearch results The specific construction methods of thesetwo spatial weight matrixes are as in Section 4

Table 1 lists Moranrsquos 119868 indexes of regional financial effi-ciency calculatedwith two spatial weightmatrixes Accordingto the estimated results the values of Moranrsquos 119868 index calcu-lated with two spatial weight matrixes are positive and passthe 10 significance test which indicates that there exists sig-nificant positive autocorrelation (spatial dependence) amongregional financial efficiency in the geographical space Thespatial distribution of regional financial efficiency is notrandom which shows that the regions with similar observedvalues are collectively distributed in the geographical space

Discrete Dynamics in Nature and Society 5

fe2003

Wz

z

minus2 minus1 0 1 2 3 4 5 6minus1

0

1

2

3

76

61

75135138

17

113

199

143

20

9414277118

117

33

185

125

116115

1371227

8

55

226

1501395149177178

183

97

18012110

146

181

44911746

179

317124

259526413226136410128158

221

260

219

871585

211

184253100

276

16112913115180

212

167

42

188

186

213

41

89252

215436242

223

83

1283419

12

10286108105781659332

187

1531884

190

119144148152

56

182

25579

227215

2611689274156

176

96133

22

40

126279

269

52

25821

114163

25765

198

170262

229236

166164

243

171

253

3570

204

27215949

47155

50

110

63

8899

235

249

201

54

248

169

210

14029

9

145

206

127

46

244

216

173453873

12010490

57

277233

37

240172112

162157

111

202

124

200

72

60

30134147

24126398160

123256

195

237

11

196

228

64

4369

268

23103107109

230

278

192

13082

27231247

224

23259267

214

207

239

5366139

48

3968

265

238

250

141

194

14

271

189218

275

197

270

220

254

106

24527458

222

208225

175

193

203

205

273251

217

1

209

5162

281

266280

24616

234

67

81

191

Moran scatterplot (Moranrsquos I = 0130)

Figure 2 Moranrsquos 119868 index scatter diagram of financial efficiency in2003

fe2012

Wz

z

minus2 minus1 0 1 2 3 4 5minus1

0

1

2

3

4

258259

180

21

22

181166

173162261257

19155

11

23188

54

187186

218

1492646152

142

220

158

55

125185

126

10418413

212214

12910

1542632332721077

20

237103

9

1592815

51

8211260100

213

37256144

53

312261795150

190

49

239116245

50

15133106261826161121

38

117

241113231

112

177

271

219

16023013744

199

222

1713228122138

210

174

27097

115

221

240153

75248

232143163

179

362940

223244

165

178

242101

23456

102

195

253

471574518

193

229246235

268

108128109

4698

11876273127120

13513242

148

94

139

96

114

7116879

48

77236

167

9389276

169

41

99

74

57

145172

9192

140

12

274

34

27717065133

217

183

24327

198

80

164176

250

16123

255

130

13684175

254

110249

86

204

4514635

262

209

12478252

1526743

60

119

215

131134

203

196

147

88

25

83

72265

269

216

3966

85

247141227

8782

275

278156

192

280

90

224

194

202

68

206

32

207200

6930

205

21111

58

62

189

63

64

5981

208

251

197

14

2387370

225

201

281

24105

279

266

191

267

Moran scatterplot (Moranrsquos I = 0052)

Figure 3 Moranrsquos 119868 index scatter diagram of financial efficiency in2012

Namely it presents a spatial correlation structure that theregions with high financial efficiency are inclined to beadjacent to other high-efficiency regions while the regionswith low efficiency are inclined to be adjacent to other low-efficiency regions

In order to depict the clustering pattern of regionalfinancial efficiency more intuitively we will draw Moranrsquos119868 index scatter diagrams of regional financial efficiency intwo typical years namely 2003 and 2012 with the use of theeconomic-geographical mixed weighting matrix

Moranrsquos 119868 index scatter diagrams divide financial effi-ciency clustering distribution into four quadrants of whichthe first quadrant (HH high efficiency-high spatial lag) andthird quadrant (LL low efficiency-low spatial lag) reflectpositive spatial correlation and the second quadrant (LHlow efficiency-high spatial lag) and fourth quadrant (HL highefficiency-low spatial lag) reflect negative spatial correlationAs shown in Figures 2 and 3 most regions are in thefirst quadrant and the third quadrant which further con-firmed that there exists significant positive spatial correlationbetween regional financial efficiency

3 Model Specification

In order to investigate how the fiscal decentralization influ-ences the financial efficiency and strategic competitionamong regional financial development econometric modelis initially expressed as

fe119894119905

= 1205730+ 120573119883119894119905

+ 120583119894+ ]119905+ 120576119894119905 (4)

In (4) 119894 and 119905 respectively indicate the region 119894 andyear 119905 120583

119894is a regional disturbance ]

119905is timing disturbance

120576119894119905is the random error fe

119894119905represents financial efficiency

and 119909119894119905contains fiscal expenditure decentralization revenue

decentralization and other controlled variablesFlowing of capital and technology across sectors and

practitionersrsquo communication and cooperation lead to obvi-ous clustering and rivalry among regional financial devel-opment which indicates that the financial developmentand efficiency in geographically neighboring regions or theregions with similar economic structure are interdependentrather than independent to some degree Considering thefiscal decentralization system strategic competition causedby public policy spillover and ldquofree riderrdquo behavior furtherenhance the spatial correlation of financial efficiency andfinancial development among regions we find that it isnecessary to introduce the interactive dependencies amongregions into the spatial econometric analysis framework toreflect the spatial spillover effects of financial developmentunder the fiscal decentralization system by constructing aspatial econometric model

It has been proved that there exists spatial correlationin regional financial efficiency by using Moran statisticalanalysis so we can generate spatial econometric models byintroducing spatial autocorrelation factors into the generalregression models According to different forms of errorshock on an observationrsquos spatial autocorrelation spatialeconometric models can be divided into the Spatial ErrorModel (SEM) and the Spatial Lag Model (SLM) In ourresearch the SEM assumes that spatial autocorrelation stemsfrom the error shock of neighboring regions on the depen-dent variables and examines the effects of neighboringregions under observation while the SLM assumes thatspatial autocorrelation stems from dependent variables andexamines the effects of financial efficiency in neighboringregions on local financial efficiency According to the prin-ciples of model construction the SEM reflects the indirectbandwagon effects caused by error terms whereas the SLMcontrols the direct bandwagon effects LeSage and Pace [32]further establish the Spatial Durbin Model (SDM) on thebasis of the SEM and the SLM The SDM includes spatial lagterms fromdependent variables and independent variables tocapture the spillover effects deriving from different variables

Wewill construct static Spatial DurbinModel (SDM) anddynamic Spatial Durbin Model (DSDM) to investigate howfiscal decentralization affects regional financial efficiencyMeanwhile in order to reflect the financial efficiencyrsquos pos-sible inertia characteristics in reality the lag item (fe

119894119905-1) offinancial efficiency will be introduced to the independent

6 Discrete Dynamics in Nature and Society

variables to measure the influence of self-reinforcing mecha-nism The specific models can be expressed as

fe119894119905

= 120588119882119894119895fe119894119905

+ 120573119883119894119905

+ 120579119882119894119895119883119894119905

+ 120583119894+ ]119905+ 120576119894119905

fe119894119905

= 120593fe119894119905-1 + 120588119882

119894119895fe119894119905

+ 120573119883119894119905

+ 120579119882119894119895119883119894119905

+ 120583119894+ ]119905

+ 120576119894119905

(5)

In (5) 120588 is the spatial lag coefficient reflecting thedirection and extent of how financial efficiency in theneighboring regions affects local financial efficiency and thecoefficient directly reflects the degree of spatial competitionin the financial efficiency 119882

119894119895is the spatial weight matrix of

119873119879 times 119873119879 119873 is the number of cross-sectional samples (281prefecture-level cities) and 119879 is time (2003ndash2012)

4 Variables Selection and Data Sources

41 Variables Definition and Measurement Methods Accord-ing to research purpose we will focus on the relationshipbetween fiscal decentralization (fiscal expenditure decentral-ization and revenue decentralization) and regional financialefficiency in the subnational citiesOther economic and socialfactors affecting regional financial efficiency include regionaleconomic development urbanization industrial structureinformation technology development fixed asset investmentand the level of internal openingTherefore wewill take theseindexes as controlled variables and introduce them into anempirical analysis model the detailed variable setting andmeasuring methods are as follows

411 The Dependent Variable The dependent variable isfinancial efficiency (fe) which is measured by the aforemen-tioned SBMmodel

412 Main Independent Variables Fiscal expenditure decen-tralization and revenue decentralization are our main inde-pendent variable When it comes to the measure method offiscal decentralization we mainly use research results of Guoand Jia [33] and Gong and Lei [34] for reference The specificmeasure formula of fiscal expenditure decentralization (sd) isas follows

sd

=

PFEcity

PFEcity + PFEprovince + PFEnation[1 minus

GDPcity

GDPnation]

(6)

In (6) PFEcity PFEprovince and PFEnation respectivelyrepresent per capita fiscal expenditure of each city theprovincial per capita fiscal expenditure and the national percapita fiscal expenditure GDPcity and GDPnation respectivelyrepresent the GDP in each city and the national GDP[1 minus GDPcityGDPnation] is the reduction factor for economicscale The index has proposed the effects of population scale

and economic scale The specific measure formula of fiscalrevenue decentralization (rd) is as follows

rd

=

PFRcity

PFRcity + PFRprovince + PFRnation[1 minus

GDPcity

GDPnation]

(7)

In (7) PFRcity PFRprovince and PFRnation respectivelyrepresent per capita fiscal revenue of each city the provincialper capita fiscal revenue and the national per capita fiscalrevenue GDPcity and GDPnation respectively represent theGDP in each city and the national GDP

413 Controlled Variable Controlled variables are as follows(1) Regional economic development (119901119892119889119901) regional

economy provides the necessary material and nonmaterialsupport for the financial development and innovation thehigher the level of regional economic development is themore beneficial to financial efficiency the improvement isIn this paper per capita actual GDP is used as a proxyvariable for regional economic development To exclude theinfluence of price factor the foregoing data that involve valueare adjusted in accordance with constant 2003 prices

(2) Urbanization (119906119903119887119886119899) Urbanization of population(the proportion of nonagricultural population accounts fortotal population) is used as the proxy variable in which thedata of nonagricultural population covering the period from2003 to 2008 is from ldquoChina Urban Statistical Yearbookrdquo(2004ndash2009) and the data of nonagricultural populationcovering the period from 2009 to 2012 is from the annualldquoStatistics on Population of Counties and Cities in Chinardquo

(3) Industrial structure (119894119905119906) the upgrading and adjust-ment of regional industrial structure constitutes a soft envi-ronment for financial development and innovation Thehigher the proportion of the third industry is the morebeneficial to financial efficiency the improvement is In thispaper the index of industrial structure upgrading is usedto represent industrial structure and the specific measureformula is as follows

itu =

119899

sum

119895=1

119902 (119895) lowast 119895 (8)

In (8) 119902(119895) is the proportion of the third industry accountsfor local GDP in region 119895 119899 = 3 and the value of index rangesfrom 1 to 3

(4) Information technology development (119894119899119905119890119903119899119890119905) it ismeasured by the number of internet users in each city

(5) Fixed asset investment (119894119891119886) it is measured by theactual fixed asset investment in each city

(6)The level of internal opening (119901119903119894V119886119905119890) with referenceto the research result of Fan et al [35] we use the developmentof private enterprises to measure the level of internal openingand assume that the higher the level of the developmentof private enterprise is the stronger the flexibility of localfinancial innovation activities is which is more conduciveto financial efficiency improvement The proportion of theemployees of private enterprises and self-employed employ-ees account for the local total population is used as a proxy to

Discrete Dynamics in Nature and Society 7

measure the development of private enterprises in differentregions

414 Spatial Weight Matrixes Strategic competition tendsto occur in the regions with the similar level of economicdevelopment simultaneously the concentration effect willgradually decay with the geographical distance [36] In orderto comprehensively reflect the spatial effects of geographicaland economic characteristics and accurately measure thespatial correlation among regions we will construct themixed geography-economy weighting matrix The specificmeasure formula is as follows

119882eco-geo =

10038161003816100381610038161003816119876119894minus 119876119895

10038161003816100381610038161003816

119889119894119895

2119894 = 119895

0 119894 = 119895

(9)

In (9) 119882eco-geo is mixed geography-economy weightingmatrix119876

119894and119876

119895 respectively represent the per capita actual

GDP of city 119894 and city 119895 (the mean of the actual GDP coveringthe period from 2003 to 2012) and 119889

119894119895is the geographical

distance between city 119894 and city 119895 calculated with the latitudeand longitude data of prefecture-level cityrsquos administrativeunit The specific measure formula is as follows

119889119894119895

= 119877 arccos (sin120593119860sin120593119861

+ cos120593119860cos120593119861cos (120582

119861minus 120582119860))

(10)

In (10) 119877 is the earth circle radius (6378 km) and 120582 and120593 respectively represent administrative centerrsquos longitudeand latitude of city 119894 and city 119895 Longitude and latitude dataare from the national fundamental geographic informationsystem All of the diagonal elements of matrix 119882eco-geo are0 And the weight matrix is normalized so that the sum ofline elements is 1 In addition considering the reliability androbustness of statistical analysis we will also construct geo-graphical distance weight matrix tomake a spatial economet-ric analysis and then compare the corresponding estimatedresultsThe specificmeasure formula of geographical distanceweight matrix is as follows

119882geo =

1

119889119894119895

2119894 = 119895

0 119894 = 119895

(11)

In (11) 119882geo is a geographical distance weighting matrixwhich is also normalized so that the sum of line elements is 1and other indexes are consistent with that of (10)

42 Data Sources and Description Considering the avail-ability of data and consistency of research samples weeventually choose 281 prefecture-level cities (Hong KongMacao and Taiwan are excluded) as samples and excludeBazhong Ziyang Lhasa Jinchang Baiyin Zhongwei andother cities due to a serious lack of available dataThe researchperiod is covering the period from 2003 to 2012 and thedata is from ldquoChina City Statistical Yearbookrdquo ldquoRegional

Economic China Statistical Yearbookrdquo and provincial statis-tical yearbooks covering the period from 2004 to 2013 Themissing data was partially filled by interpolation In orderto weaken the influence of heteroskedasticity outliers andanomaly terms on the data stability all the data are processedby the logarithm An overview of the descriptive statistics ofthese data is provided in Table 2

In order to avoid the deviation of the model settingand spurious regression we carry on unit root testing onrelated variables by LLC and Fisher-ADF testing methodsbefore spatial econometric regression analysis The resultsof stationary testing of panel data indicate that all variablesreject the null hypothesis of unit root which passes the testat the 5 significance level Moreover the panel data arestationary and the regression results will be significantlyreliable

5 Empirical Methodology

51 Spatial Econometric Analysis of Fiscal Decentralization onRegional Financial Efficiency Since the SDMmodel containsthe spatial lag item of the dependent variable and the spatiallag items of independent variables at the same time so itcanmore fully reflect the impact of spatial autocorrelation onthe regression results than SEM model and SLM model [32]Therefore we construct SDMmodels to test the spatial effectsof the fiscal revenue decentralization and fiscal expendituredecentralization on the regional financial efficiency The 119875

values of Hausman test of the spatial econometric models areless than 005 which indicates that the estimated results offixed effect estimation are better than that of random effectestimation Therefore we will specifically carry out analysisaccording to the estimated results of spatially fixed effectmodels in the following parts Meanwhile a robust test iscarried by constructing DSDM models It is found that theestimated coefficients of SDMmodel and DSDMmodel havethe same direction while the values of estimated coefficientsand level of significance are also very close indicating that theresults of SDM model are significantly reliable Comparingthe values of 119877

2 and Log-likelihood of SDM model andDSDM model we find that the fitting degree of SDM modelis higher than that of DSDMmodel so the following analysisis mainly based on the econometric results of SDMmodel

As shown in Table 3 the overall spatial correlationcoefficients 120588 of four models are positive and pass thetests at the 1 significance level indicating that there arepositive spatial dependent relationships between the localfinancial efficiency and the neighboring regionsrsquo financialefficiency that is the economic activities of geographyneighborhood regions or regions with similar level of eco-nomic development have obvious spillover effect on localfinancial efficiency thereby causing imitation and strate-gic competition of financial development and innovationamong regions On one hand the geographical adjacencyand similarity of economic development in different regionsbring huge convenience to cooperation between financialenterprises sharing of infrastructure information exchangeand communication innovation and diffusion of knowledge

8 Discrete Dynamics in Nature and Society

Table 2 Statistical description of samplersquos variables

Index Variables Mean Standarddeviation Minimum 25 quantile 50 quantile 75 quantile Maximum

Finance efficiency fe 0576 0624 0133 0380 0500 0653 3958Fiscal expendituredecentralization sd 0368 0108 0138 0294 0351 0421 0888

Fiscal revenuedecentralization rd 0242 0146 0037 0133 0202 0318 0871

Per capita real GDP(yuanperson) pgdp 27119890 + 04 32119890 + 04 1891200 9837380 17119890 + 04 31119890 + 04 44119890 + 05

Urbanization () Urban 0345 0195 0013 0202 0296 0451 100000Industrial structure itu 2203 0136 1318 2110 2190 2283 2797Informationtechnologydevelopment (tenthousand families)

Internet 41066 128506 0024 8316 17139 36800 5174000

Fixed assetinvestment (tenthousand yuan)

ifa 15119890 + 04 15119890 + 04 654299 4606420 9841960 20119890 + 04 17119890 + 05

The level of internalopening () Private 0086 0106 0003 0036 0059 0096 100000

and technology flowing of financial resources among regionswhich causes regional financial development and innovationand imitation and strategic competition among regions [37]on the other hand the restriction of transportation cost onfinancial sector is not strong and the rapid development ofinternet technology leads to cross-regional financial transac-tions and innovation thus further expanding the clusteringeffect effect of economic scale and spatial spillover effectsof financial development The coefficients of lag item (fe

119894119905-1)of financial efficiency in DSDM models are positive andthey are respectively 0211 and 0212 both of them pass thetests at the 1 significance level indicating that the previousinputs and outputs of production will be present through thefinancial efficiency increasing in later stage (or later period)and financial efficiency increasing is a dynamic systemicprocess The specific impact of each variable on financialefficiency is analyzed as follows

The estimated coefficients of Lnsd and Lnrd are positiveand all of them pass the tests at the 1 significance levelwhich indicates that the fiscal expenditure and revenuedecentralization play a significant positive role in promotinglocal financial efficiency Under the dual pressures of makingup the fiscal gap and political promotion the local govern-ments regarding economic development as main missionhave instinctive strong interest in scarce flowing economicfactors [38] As the excellent combination of capital talenttechnology management and marketing financial resourceswill be the important target for battling among local govern-ments With the expansion of fiscal expenditure decentral-ization and revenue decentralization local government usesmany policy tools such as fiscal subsidies tax concessionsand government procurement to compete for getting thefinancial resources by attracting financial capital humancapital and technological innovation to flow into the localregion thus contributing to the financial innovation and

improvement of financial efficiencyThe influence coefficientof fiscal expenditure decentralization is 0357 greater thanthe influence coefficient (0153) of fiscal revenue decentral-ization indicating that the impacts of fiscal expendituredecentralization and revenue decentralization on regionalfinancial efficiency have unbalanced characteristics and thepossible reason is that local governments under provinceget very limited taxation autonomy resulting in the factthat revenue decentralization is often lower than expendi-ture decentralization In addition under the capital-driveneconomic growth model the flowing direction of capitaldepends on the level of infrastructure [39] thus the inflowof financial resources as well as financial innovation oftenoccurs in the regions with rapid economic developmentexcellent investment environment and great consumptionability so that local governments have to increase fiscalexpenditure to enhance the supply of the associated support-ing infrastructure Comparatively speaking tax preferenceof undeveloped regions is less attractive than the superiorityconditions of developed regions

We will further analyze the estimated coefficients andsignificance levels of spatial lags of fiscal expenditure andrevenue decentralization and investigate the strategic com-petitive effects of financial efficiency among neighboringregions The spatial lag of expenditure decentralization(WLnsd) does not pass the significance test suggestingthat spatial spillover effect of fiscal expenditure decentral-ization is uncertain and needs further examination whilethe spatial lag of fiscal revenue decentralization (WLnrd)is negative at the 5 significance level and the estimatedcoefficient is minus0228 indicating that with the increase offiscal revenue decentralization the negative externalities onfinancial development and innovation will be increasinglyprominent thus inhibiting the improvement of financialefficiency in neighboring regions The reason may be that

Discrete Dynamics in Nature and Society 9

Table 3 Estimated results of regional financial efficiency under theSpatial Durbin Model (economic-geographic weighting matrix)

Independentvariables

Dependent variable sd Dependent variable rd

SDMmodel DSDMmodel SDMmodel DSDM

model

feitminus10211lowastlowastlowast(1283)

0212lowastlowastlowast(1283)

Lnsd 0357lowastlowastlowast(626)

0334lowastlowastlowast(570)

Lnrd 0153lowastlowastlowast(470)

0143lowastlowastlowast(408)

Lnpgdp 0169lowastlowastlowast(462)

0120lowastlowastlowast(308)

0138lowastlowastlowast(360)

0099lowastlowast(244)

Lnurban minus0050lowastlowast(minus237)

minus0044lowastlowast(minus209)

minus0050lowastlowast(minus237)

minus0045lowastlowast(minus215)

Lnitu 0433lowastlowast(227)

0438lowastlowast(215)

0525lowastlowastlowast(273)

0531lowastlowastlowast(259)

Lninternet minus0295lowastlowastlowast(minus2970)

minus0299lowastlowastlowast(minus2716)

minus0292lowastlowastlowast(minus2945)

minus0297lowastlowastlowast(minus2690)

Lnifa minus0446lowastlowastlowast(minus2298)

minus0386lowastlowastlowast(minus1832)

minus0437lowastlowastlowast(minus2321)

minus0374lowastlowastlowast(minus1806)

Lnprivate 0038lowastlowastlowast(322)

0037lowastlowastlowast(290)

0039lowastlowastlowast(325)

0038lowastlowastlowast(302)

WLnsd 0179(099)

minus0041(minus022)

WLnrd minus0228lowastlowast(minus251)

minus0186lowast(minus179)

WLnpgdp 0016(027)

0064(101)

0058(096)

0097(150)

WLnurban 0097lowastlowast(236)

0082lowastlowast(198)

0107lowastlowastlowast(259)

0089lowastlowast(213)

WLnitu 1159lowast(193)

0516(082)

0615(100)

0228(035)

WLninternet 0177lowastlowastlowast(647)

0181lowastlowastlowast(622)

0180lowastlowastlowast(666)

0180lowastlowastlowast(619)

WLnifa 0237lowastlowastlowast(539)

0251lowastlowastlowast(532)

0275lowastlowastlowast(704)

0260lowastlowastlowast(600)

WLnprivate 0065lowast(185)

0045(115)

0083lowastlowast(232)

0060(153)

120588

0420lowastlowastlowast(1165)

0462lowastlowastlowast(1251)

0429lowastlowastlowast(1205)

0463lowastlowastlowast(1266)

1198772 0448 0443 0441 0438

log 119871 599812 595228 595228 587624Observations 2810 2529 2810 2529Note the values in parentheses represent the 119905-statistics values of thecorresponding estimated coefficientslowastlowastlowastlowastlowast andlowast respectively representsignificance level of 1 5 and 10

local governmentsrsquo motivation of pursuing revenues underthe fiscal revenue decentralization system is not conduciveto free flowing of financial resources among regions whichdirectly affects the financial efficiency in other regions

When it comes to the estimated coefficients of othercontrolled variables we found that not all the spatial spillovereffects of variables are significant Overall various economicactivities have direct and indirect influence on financial

efficiency and promote the strategic financial competitionamong regions The specific effects are as follows the levelof open-into-inside can not only improve the local financialefficiency significantly but also increase the financial effi-ciency correspondingly in neighboring regions The strongdemand for various financial products and services of privateenterprises promotes financial innovation and improves theefficiency of resource allocation in financial markets Onthe contrary urbanization information technology develop-ment and fixed assets investment suppress the improvementof local financial efficiency but play a positive role in promot-ing the financial efficiency in neighboring regionsThe accel-erating of urbanization process the application of internettechnology and investment in fixed assets promote economicexchanges and cooperation among regions so that the spatialspillover effects of financial development and innovation areincreasingly significant Furthermore although the regionaleconomic development and industrial structure significantlycontribute to the improvement of regional financial efficiencyin general the spillover effects among regions are not obvious

52 Decomposition of Fiscal Decentralizationrsquos Effects onRegional Financial Efficiency Whether there truly existspillover effects among the variables relying solely on the spa-tial lag coefficients of SDMmodel and DSDMmodel is likelyto lead to misinterpretation of the estimated results Spatialeffects can be further decomposed into direct effect indirecteffect (spillover effects) and total effect with reference to theresearch result of LeSage and Pace [32] The indirect effectsrepresent the independent variablesrsquo influence on the depen-dent variable in all other regions through spatial interactionsTherefore we will further verify the existence of spillovereffects by using partial differential matrix analysis Table 4illustrates the direct effect indirect effect and total effect ofthe fiscal expenditure and revenue decentralization in SDMmodel The results show that the estimated coefficients ofindependent variablesrsquo direct effect have the same directionwith the estimated coefficients of SDMmodel in Table 3 andthe estimated coefficients and significance levels are also veryclose

According to the estimated results of SDM model inTable 4 the direct effect indirect effect and total effectof fiscal expenditure decentralization on regional financialefficiency are positive and pass the 10 significance leveltest The regression coefficient of direct effect is 0368 theregression coefficient of indirect effect is 0560 and theregression coefficient of total effect is 0928 indicating thatthe degree of fiscal expenditure decentralization increases by1 while regional financial efficiency increases by 0928Fiscal expenditure decentralization plays a significant posi-tive impact on financial development and financial efficiencywhich is opposite to the research result of Wang and Li [23]The opposite conclusions are possibly due to the differencesof sample selection and indexesThe rise of fiscal expendituredecentralization in the local region leads to an increaseby 0368 of the financial efficiency while the increase ofexpenditure decentralization in the neighboring region leadsto financial efficiency increasing by 0560 indicating that

10 Discrete Dynamics in Nature and Society

Table 4 Decomposition of direct effect indirect effect and total effect under SDMmodel

Independent variablesDependent variable sd Dependent variable sd

Direct effect Indirect effect Total effect Direct effect Indirect effect Total effect

Lnsd 0368lowastlowastlowast(750)

0560lowast(195)

0928lowastlowastlowast(310)

Lnrd 0146lowastlowastlowast(529)

minus0283lowast(minus195)

minus0137lowast(minus192)

Lnpgdp 0175lowastlowastlowast(435)

0132(153)

0307lowastlowastlowast(359)

0145lowastlowastlowast(346)

0187lowastlowast(218)

0332lowastlowastlowast(392)

Lnurban minus0046lowastlowast(minus207)

0127lowastlowast(200)

0081(130)

minus0045lowastlowast(minus204)

0146lowastlowast(226)

0101(158)

Lnitu 0486lowastlowastlowast(275)

2321lowastlowast(237)

2807lowastlowastlowast(292)

0559lowastlowastlowast(315)

1488(151)

2047lowastlowast(211)

Lninternet minus0291lowastlowastlowast(minus3055)

0086lowast(190)

minus0205lowastlowastlowast(minus432)

minus0288lowastlowastlowast(minus3029)

0091lowastlowast(200)

minus0197lowastlowastlowast(minus410)

Lnifa minus0441lowastlowastlowast(minus2260)

0087(149)

minus0353lowastlowastlowast(minus594)

minus0430lowastlowastlowast(minus2254)

0154lowastlowastlowast(283)

minus0276lowastlowastlowast(minus494)

Lnprivate 0041lowastlowastlowast(319)

0134lowastlowast(235)

0175lowastlowastlowast(285)

0042lowastlowastlowast(327)

0166lowastlowastlowast(283)

0208lowastlowastlowast(330)

Note lowast lowast lowast lowastlowast and lowast respectively represent the significant level of 1 5 and 10

fiscal expenditure decentralization significantly promotesthe competition of financial development and innovationamong regions The expansion of the autonomy of localgovernment expenditure strengthens the regulatory capacityof local governments which is conducive to acceleratingmarket-oriented reform processing and achieves the organicintegration between fiscal system and market mechanism inthe allocation of financial resources which is beneficial tocreating a favorable institutional environment is also con-ducive to the flowing of financial resources among regionsand greatly promotes financial efficiency

The fiscal revenue decentralizationrsquos direct effect onregional financial efficiency is significantly positive whilethe spillover effect is significantly negative indicating thatthe impact of fiscal revenue decentralization on regionalfinancial efficiency presents the feature of ldquodouble-edged-swordrdquo Fiscal revenue decentralization has a significant posi-tive impact on the local financial efficiencyHowever there aresignificant negative spillover effects on the financial efficiencyin other regions The regression coefficient of the directeffect is 0146 the regression coefficient of the indirect effectis minus0283 and the regression coefficient of the total effectis minus0137 which indicates that the degree of fiscal revenuedecentralization increasing by 1 will lead to decrease by0137 of regional financial efficiency Furthermore whilethe increase of local fiscal revenue decentralization leads tofinancial efficiencyrsquos increase by 0146 the increase of thefiscal revenue decentralization in neighboring regions resultsin the financial efficiencyrsquos decrease by 0283 Comparingthe estimated coefficient of direct effect and that of indirecteffect it can be found that the negative spillover effectsof fiscal revenue decentralization are significantly greaterthan the direct scale effect so the negative externalitiesof fiscal revenue decentralization on financial efficiencyare greater than the positive externality On one hand inorder to obtain high-quality financial resources the local

governments ignore the differences like the level of economicdevelopment factor endowments and other aspects amongregions and seek to imitate policies and instruments inother regions thus leading to chain reaction of absorbinginvestment by other regions On the other hand whenthe neighboring regions or the regions with similar levelof economic development take some measures to attractliquidity factors the local government driven by yardstickcompetitionwill adopt similar policies to avoid the outflow ofelements which results in convergence of competitive tactics[40 41] Tax incentives reduction of fee and other wayswill become important means of competition and imitationbetween local governments in different regions on the basisof the independent revenue authority However such kind ofcompetitive mode between local governments for resourceswill consume huge fiscal resources and weaken the capacityof local governmentrsquos provision of public goods and servicesUltimately it is harmful to provide healthy stable andsustainable economic development environment for financialinnovation and increase of financial efficiency

53 Robustness Test To ensure the robustness and reliabilityof research results we will calculate the spatial lags withthe geographical distance weight matrix and do econometricregression by taking financial efficiency as the dependentvariable so that we can comparatively analyze the differenceof influence with different spatial weight setting based on theabove The robustness test results are as in Table 5

As shown in Table 5 the estimated results indicate thatthe sign of regression coefficients of the fiscal expendituredecentralization and revenue decentralization is consistentwith that of corresponding variable in the model of Table 3and the regression results in themodel are basically consistentwith the conclusions in Table 3 Comparing the estimatedresults based on two different weightmatrixes it is found thatthere are some differences in the influence coefficients of two

Discrete Dynamics in Nature and Society 11

Table 5 Robustness test on Spatial Durbin Models (geographicaldistance weight matrix)

Independentvariables

Dependent variable sd Dependent variable rdSDMmodel

DSDMmodel

SDMmodel

DSDMmodel

feitminus1 mdash 0202lowastlowastlowast(1276) mdash 0200lowastlowastlowast

(1255)

Lnsd or Lnrd 0364lowastlowastlowast(636)

0348lowastlowastlowast(592)

0173lowastlowastlowast(522)

0161lowastlowastlowast(458)

Lnpgdp 0268lowastlowastlowast(735)

0225lowastlowastlowast(578)

0242lowastlowastlowast(637)

0209lowastlowastlowast(522)

Lnurban minus0026(minus112)

minus0026(minus112)

minus0028(minus120)

minus0027(minus115)

Lnitu 0465lowastlowast(246)

0478lowastlowast(238)

0527lowastlowastlowast(278)

0542lowastlowastlowast(269)

Lninternet minus0295lowastlowastlowast(minus3038)

minus0300lowastlowastlowast(minus2799)

minus0292lowastlowastlowast(minus3005)

minus0298lowastlowastlowast(minus2771)

Lnifa minus0484lowastlowastlowast(minus2479)

minus0427lowastlowastlowast(minus2024)

minus0472lowastlowastlowast(minus2448)

minus0417lowastlowastlowast(minus1988)

Lnprivate 0036lowastlowastlowast(312)

0035lowastlowastlowast(284)

0035lowastlowastlowast(307)

0035lowastlowastlowast(288)

WLnsd orWLnrd

minus0264(minus149)

minus0418lowastlowast(minus225)

minus0277lowastlowastlowast(minus370)

minus0241lowastlowastlowast(minus276)

WLnpgdp minus0269lowastlowastlowast(minus427)

minus0286lowastlowastlowast(minus428)

minus0237lowastlowastlowast(minus369)

minus0267lowastlowastlowast(minus394)

WLnurban 0053(147)

0059lowast(167)

0059(163)

0062lowast(173)

WLnitu 0104(020)

minus0085(minus016)

minus0205(minus039)

minus0290(minus051)

WLninternet 0217lowastlowastlowast(838)

0239lowastlowastlowast(853)

0214lowastlowastlowast(831)

0235lowastlowastlowast(841)

WLnifa 0434lowastlowastlowast(1052)

0464lowastlowastlowast(1054)

0433lowastlowastlowast(1165)

0450lowastlowastlowast(1129)

WLnprivate 0062lowast(187)

0038(106)

0081lowastlowast(241)

0050(139)

120588

0547lowastlowastlowast(1858)

0580lowastlowastlowast(1938)

0541lowastlowastlowast(1839)

0570lowastlowastlowast(1898)

1198772 0444 0428 0443 0430

log 119871 659741 665804 665804 659509Observations 2810 2529 2810 2529Note lowastlowastlowast lowastlowast and lowast respectively represent the significant level of 1 5and 10

kinds of fiscal decentralizations indicators which show thatthe gap between the levels of economic developmentwill havean impact on role of fiscal decentralization acting on regionalfinancial efficiency On one hand under the fiscal decen-tralization system because of the heterogeneity of regionaleconomic development level the local governmentsrsquo ability ofmaking up the gap between fiscal revenue and expenditure isdifferent and the fiscal pressure is also different therefore thedegree of interference to the regional market environment bythe local government is different which leads to the regionaldifferences in financial efficiency in a certain degree [21]On the other hand the occurrence of financial innovationdepends on the support of the economic foundation and

the different economic and social environment will lead tothe fact that the financial innovation activities are more andmore concentrated in the regions with the advantages offinancial location resulting in nonequilibrium of financialefficiency in different regions [42] Under the performanceassessment method focusing on GDP local governments willnot only consider the strategic behaviors of the neighboringregions but also tend to think about the behaviors of theregions with the similar level of economy In additioncross-regional knowledge diffusion technology spillover andhuman capital flow are more prone to occur among theregions with similar level of economic development so thatthe financial exchanges and contacts among regions are morefrequent It also shows that besides considering the impact ofgeographical distance the economic distance factor shouldbe taken into consideration while we are researching thespatial effects of fiscal decentralization on regional financialinnovation in addition to the consideration of the impact ofgeographical distance the economic distance factor shouldbe considered

6 Conclusions and Enlightenments

Based on the panel data of 281 prefecture-level cities in Chinacovering the period of 2003ndash2012 the regional financialefficiency is measured by the superefficiency SBMmodel thegeographic distance weight matrix and economic-geographymixed weight matrix which can reflect the regional similarityare introduced into the SDM and the spatial effects of thefiscal expenditure and revenue decentralization on regionalfinancial efficiency under province are investigated as wellas the strategic financial competition arising from it Basedon the results of previous empirical analysis we can draw thefollowing basic conclusions

First the geographical distribution of Chinarsquos prefecture-level citiesrsquo financial efficiency presents significant features oftime inertia and spatial dependency Financial developmentand innovation among regions particularly neighboringregions is not independent but shows significant spatialspillover effects The financial efficiency of the local regionwill be significantly affected by the financial efficiency ofadjacent regions The spatial effect is increasingly obviousin the process of financial development and innovation andthe size of the spillover effects is closely related to spatialweight (which is used to measure regional similarities) andmainly shows that there is competition for financial efficiencyamong regions with similar economic attributes and adjacentregions

Second it cannot be ignored that the institutional factorsespecially fiscal decentralization have effect on the financialefficiency and the factors have a decisive role in the allocationof financial resource and the regional distribution of financialinnovation The improvement of fiscal decentralization levelwill help to attract more financial resources thereby promot-ing an increase in financial efficiencyThe improvement of thefinancial autonomy in neighboring jurisdictions will reducethe financial resources flowing into the local region andsuppresses the occurrence of the local financial innovationIn addition there exist significant spatial spillover effects of

12 Discrete Dynamics in Nature and Society

financial development and efficiency caused by fiscal decen-tralization The scope of spatial spillovers is mainly reflectedin two aspects Thus the local governmentsrsquo competition notonly strongly increases the financial efficiency of the regionbut also accelerates interregional competition for financialefficiency

Third fiscal expenditure decentralization and revenuedecentralizationrsquos impact on regional financial efficiencyhave nonequilibrium characteristics First of all there existsasymmetry between fiscal expenditure decentralization andrevenue decentralizationrsquos impact on regional financial effi-ciency Fiscal expenditure decentralization can promoteoptimization of the fiscal expenditure structure and fiscalrevenue decentralization can increase the local governmentsrsquorevenue share so as to provide positive incentives [43] bothof them have a positive and promoting effect on the improve-ment of local financial efficiency and such effect largelydepends on the symmetry between the fiscal expenditure andrevenue decentralization If it is asymmetric between theirproperty rights and the expenditure responsibilities underfiscal decentralization system the local governments maymake excessive intervention to regional financial resourcesallocation tomeet their own demands for financial resourcesNext there exists nonconformance between fiscal expendi-ture decentralization and revenue decentralizationrsquos impacton financial efficiency in neighboring regions Comparedwith the fiscal expenditure decentralization the fiscal revenuedecentralization has more apparent inhibitory effect on thefinancial efficiency in neighboring regions

Our research conclusions have important policy implica-tions that illuminate the relationship between fiscal decen-tralization and financial efficiency

(1) The regional financial efficiency is closely relatedto the location and the economic development of the sur-rounding cities The regional financial efficiency is affectednot only by the financial development and innovation inneighboring regions but also by the impact of the struc-tural differences among regions In this case breaking theadministrativemonopoly of local protectionism and carryingon the ldquorichening neighborrdquo policy of financial developmentare the best choices for the local government to developthe regional economy When making policies of financialdevelopment local governments need to consider their owneconomic development environment and economic featuresin neighboring regions and strengthen communication andcoordination with the surrounding regions so as to avoidthe possible contradictions and conflicts on financial devel-opment policies among regions

(2) The spatial spillover effects of the regional financialefficiency should be divided into two kinds of effects includ-ing global geographic spillover effects and local geographicspillover effects In order to promote the increase of financialefficiency in a better way it should also be based on thehierarchy of spatial effect of and the degree of spillover effectto define the relationship between the government and themarket and central and local governments and among thelocal governments and their role in regional financial devel-opment and innovation so as to better promote financialefficiency

(3) There is a two-way interactive mechanism betweenthe fiscal system and financial development such effectsmay affect the behaviors of the local governments and theallocation of resources and ultimately act on macroeconomicperformance [10] Under the fiscal decentralization systemthe strategic competitive behaviors among local governmentssignificantly affect the liquidity of financial resources and theoccurrence of financial innovation the local governmentsrsquoexpansion of fiscal expenditure and revenue shortfall becomethe main driving force of each region competing for financialresources In order to better play the role of incentivesand constraints of fiscal competition mechanism under thefiscal decentralization system and avoid the negative impactbrought by the vicious interregional strategic financial com-petitive behaviors on the regional economic development thelocal governmentsrsquo short-sighted behaviors should be strictlyregulated and then the interregional financial cooperationwith mutual benefit as the core should be actively promoted

(4) Constructing a scientific and rational framework forfiscal decentralization will be an important force to accelerateeffectively the improvement of regional financial efficiencyChinarsquos future subprovincial fiscal decentralization reformshould adapt to the new normal financial development andfurther rationalize the provincial and subprovincial fiscaldecentralization system to ensure a balanced and appropriatefiscal expenditure and revenue decentralization On onehand we should weaken local governmentsrsquo negative inter-vention in financial system by reducing local governmentsrsquofinancial pressure caused by the asymmetry between thefiscal resources and expenditure responsibilities On the otherhandwe should also actively play the function of fiscal systemto adjust the financial development and innovation and fullyplay public financersquos role in promoting the flowing of finan-cial resources among regions the balanced development ofregional finance and the improvement of financial efficiency

Competing Interests

There are no competing interests in regard to the contentdiscussed in this paper

Authorsrsquo Contributions

All authors conceived and designed the research XiaomeiHucollected and processed the data Jianmin Liu analyzed thedata and interpreted the results all authors were also involvedin writing the paper and have all approved the submittedform

Acknowledgments

The research of this paper is jointly funded by the NationalSocial Science Foundation of China under Project no14BJY159 and the Hunan Philosophy and Social ScienceFund Project under Grant no 13JD09 The authors gratefullyacknowledge the support of Jinguang Wu from HunanUniversity of Finance and Economics China

Discrete Dynamics in Nature and Society 13

References

[1] J Zhang and X Jin ldquoRe-examine of the relationship betweenChinarsquos financial deepening and productivity 1987ndash2001rdquo Eco-nomic Research Journal vol 11 pp 35ndash43 2005

[2] K Chen A L Hillman and Q Y Gu ldquoFiscal centralization andchanges of local governmentsrsquo behaviorsmdashfrom lsquohands for helprsquoto lsquohands for grabrsquordquoChina EconomicQuarterly vol 1 pp 111ndash1302002

[3] L Zhou and Z M Wang ldquoEmpirical analysis of regionalfinancial development of China and economic growth 1978ndash2000rdquo Financial Research vol 10 pp 1ndash13 2002

[4] L Zhou ldquolsquoSecond financersquo of Chinarsquos financial industry andfinancial division during the reforming periodrdquo The Journal ofWorld Economy vol 6 pp 72ndash79 2003

[5] L Zhou ldquoGradual transition state capacity and fiscal functionof financerdquo Journal of Finance and Economics vol 2 pp 26ndash372005

[6] E L He and W Liu ldquoFinancial capital local governmentsrsquointervention and industrial structuremdashan empirical analysisbased on the data of central regionsrdquo Economic Issues vol 5pp 37ndash40 2011

[7] X L Dou and W J Wang ldquoThe differences and trends of Chi-nese regional financial development and local governmentsrsquobehavior and analysis of tax sharing reform on the differencesand trends of Chinese regional financial developmentrdquo Shang-hai Finance vol 2 pp 9ndash15 2011

[8] J Zhang and K R Shen ldquoLocal government interventionregional financial development and the transformation ofChinarsquos economic growth modemdashan empirical study under thefiscal decentralization systemrdquo Nankai Economic Studies vol 6pp 122ndash141 2008

[9] G Chen X G Yin and Y Pan ldquoChinarsquos financial developmenttax sharing reform and economic growthrdquo Financial Researchvol 2 pp 99ndash109 2006

[10] P P Ding and Y Fu ldquoBehaviors of local governments therelationship of fiscal and finance and macroeconomic volatilityassociated with Chinamdashan analysis under Chinese-style fiscaldecentralization systemrdquo Comparative Economic amp Social Sys-tems vol 6 pp 87ndash97 2012

[11] X G Yin G Chen and Y Pan ldquoTax sharing reform local gov-ernmentsrsquo intervention and financial development efficiencyrdquoJournal of Finance and Economics vol 10 pp 92ndash101 2006

[12] M G Yu and H B Pan ldquoGovernment intervention the ruleof law financial development and bank loans of state-ownedenterprisesrdquo Financial Research vol 9 pp 1ndash23 2008

[13] D X Wang J Liu and L Li ldquoFiscal decentralization bankcredit and total factor productivityrdquo Journal of Finance andEconomics vol 4 pp 69ndash80 2011

[14] Y X Jiang and R B Jiang ldquoRegional financial innovationefficiency evaluation environmental impact and gap analysisrdquoJournal of Zhejiang University (Humanities and Social Sciences)vol 43 no 4 pp 52ndash65 2013

[15] H W Yang L W Cheng Y Xu and Y W Qi ldquoMeasure andinfluencing factors of financial resources allocation efficiencyunder fiscal decentralization system based on super-efficiencyDEA-TOBIT two-step methodrdquo Journal of Jiangxi NormalUniversity (Natural Sciences Edition) vol 6 pp 587ndash592 2014

[16] N Shen F Z Liu and J Q Zhao ldquoFiscal decentralizationfinancial deepening and regional international poverty prone todeveloprdquo Finance amp Trade Economics vol 1 pp 41ndash46 2006

[17] Q Y Du G P Li and N Shen ldquoFinancial efficiency andoptimization of the regional industrial structure under thecurrent fiscal systemrdquo Xinjiang Social Science vol 4 pp 16ndash212006

[18] T Feng Y W Song and Y Lu ldquoFiscal decentralization localgovernmentsrsquo behavior and regional financial developmentrdquoJournal of Xirsquoan Jiaotong University vol 5 pp 20ndash27 2007

[19] Y A Zhou X J Ma and J Y Zhao ldquoGovernment behaviorsfinancial development and economic growthrdquo Henan SocialSciences vol 1 pp 55ndash60 2007

[20] C Hao ldquoDevelopment of financial intermediation and eco-nomic growth the Chinese experiencerdquo China EconomicReview vol 17 no 4 pp 347ndash362 2006

[21] J Zhang Fiscal Decentralization Regional Financial Devel-opment and Chinarsquos Economic Growth Performance NanjingUniversity Press Nanjing China 2011

[22] Y J Yao ldquoGovernment intervention the development of finan-cial intermediation and economic growthrdquo Journal of Financeand Economics vol 8 pp 52ndash58 2010

[23] X Wang and L Li ldquoFiscal decentralization and regional finan-cial development the theoretical and empirical analysis fromthe perspective of central governmentrdquo Public Finance Researchvol 4 pp 31ndash35 2012

[24] W L Zhang and C L Qin ldquoFinancial development fiscaldecentralization and regional economic differencesrdquo EconomicGeography vol 1 pp 583ndash588 2009

[25] Y Ma J Li and B Chen ldquoProvincial differences in Chinesefiscal decentralization financial development industrializationand economic growthrdquo EconomicTheory and Business Manage-ment vol 2 pp 5ndash19 2015

[26] Y X Jiang and R B Jiang ldquoThe spatial external effects ofregional financial innovation financial competition and finan-cial agglomerationrdquo Social Science Front vol 3 pp 79ndash84 2014

[27] K Tone ldquoA slacks-based measure of super-efficiency indata envelopment analysisrdquo European Journal of OperationalResearch vol 143 no 1 pp 32ndash41 2002

[28] Y Q Lu and D G Zhang ldquoChinarsquos regional financial efficiencyand differences of efficiency measurementrdquo Economic Geogra-phy vol 1 pp 96ndash101 2012

[29] A N Berger and D B Humphrey ldquoEfficiency of financialinstitutions international survey and directions for futureresearchrdquo European Journal of Operational Research vol 98 no2 pp 175ndash212 1997

[30] Y J Yao ldquoEmpirical analysis of financial development andurban-rural income gaprdquo Journal of Finance and Economics vol2 pp 49ndash59 2005

[31] Q Li J Song E Wang H Hu J Zhang and Y WangldquoEconomic growth and pollutant emissions in China a spatialeconometric analysisrdquo Stochastic Environmental Research andRisk Assessment vol 28 no 2 pp 429ndash442 2014

[32] J P LeSage and R K Pace Introduction to Spatial EconometricsCRCPress Taylor amp Francis Group Boca Raton Fla USA 2009

[33] Q W Guo and J X Jia ldquoFiscal decentralization organizationalstructure of local governments and the size of governmentexpenditurerdquoEconomic Research Journal vol 11 pp 59ndash87 2010httpwwwcnkicomcnArticleCJFDTotal-JJYJ201011008htm

[34] F Gong and X Lei ldquoThe measurement of Chinese-style fiscaldecentralizationrdquo Statistical Research vol 10 pp 47ndash55 2010

[35] G Fan X L Wang and H P Zhu Chinese Marketing IndexRegional Marketing Process in 2011 Economic Science PressBeijing China 2011

14 Discrete Dynamics in Nature and Society

[36] S S Rosenthal and W C Strange ldquoGeography industrialorganization and agglomerationrdquoTheReview of Economics andStatistics vol 85 no 2 pp 377ndash393 2003

[37] S Park Yoon International Banking and Financial CentersKluwer Academic Boston Mass USA 1989

[38] B Y Qiao J Y Fan and X Y Feng ldquoChinarsquos fiscal decentral-ization and compulsory primary educationrdquo Social Sciences inChina vol 6 pp 37ndash46 2005

[39] W JWang and C L Qin ldquoRegional differences in the behaviorsof local governments and growth effect of fiscal decentralizationbased on the analysis of judgment hypothesis and testingrdquoManagement World vol 1 pp 9ndash21 2008

[40] J K Brueckner ldquoStrategic interaction among governments anoverview of empirical studiesrdquo International Regional ScienceReview vol 26 no 2 pp 175ndash188 2003

[41] F Revelli ldquoOn spatial public finance empiricsrdquo International Taxand Public Finance vol 12 no 4 pp 475ndash492 2005

[42] N Shen and T Y He ldquoFinancial efficiency and regional eco-nomic growth from the perspective of fiscal decentralizationrdquoReform vol 1 pp 61ndash65 2006

[43] J M Liu X M Hu and J G Wu ldquoThe threshold effect ofsub-national fiscal decentralization revenue and expendituredecentralization on provincial industrial transformation andupgrading based on the data of 14 cities in Hunan ProvincerdquoPublic Finance Research vol 8 pp 49ndash52 2014

Submit your manuscripts athttpwwwhindawicom

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

MathematicsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Mathematical Problems in Engineering

Hindawi Publishing Corporationhttpwwwhindawicom

Differential EquationsInternational Journal of

Volume 2014

Applied MathematicsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Probability and StatisticsHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Mathematical PhysicsAdvances in

Complex AnalysisJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

OptimizationJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

CombinatoricsHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Operations ResearchAdvances in

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Function Spaces

Abstract and Applied AnalysisHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of Mathematics and Mathematical Sciences

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Algebra

Discrete Dynamics in Nature and Society

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Decision SciencesAdvances in

Discrete MathematicsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom

Volume 2014 Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Stochastic AnalysisInternational Journal of

4 Discrete Dynamics in Nature and Society

0

1

2

3

0 05 1 15 2 25

k density fe2003

k density fe2007k density fe2012

x

Figure 1 Kernel density segment curve of financial efficiency

results in money flowing from low income groups to highincome groups due to the threshold and cost to obtain finan-cial services and the difference between regional resourcesendowment which will exacerbate income inequality espe-cially in developing countries With reference to the researchresults of Yao [30] we take urban-rural income ratio as aproxy variable where urban-rural income ratio = per capitadisposable income of urban residentsper capita net incomeof rural residents

Based on the input-output data of 281 prefecture-levelcities covering the period from 2003 to 2012 this paper mea-sures the regional financial efficiency by using MAXDEA64software The calculation results show that there exists greatdifference in regional financial efficiency and the averagefinancial efficiency is 0576 while the standard deviationis 0624 From a regional perspective the average financialefficiency in the eastern central and western regions isrespectively 0628 0503 and 0604 and the financial effi-ciency of the central region is lower than that of eastern andwestern regions In addition this paper makes Kernel densityestimation on regional financial efficiency in the three typicalyears (2003 2007 and 2012) by using nonparametric Gaussnormal Kernel density (nuclear density) segment curve Asshown in Figure 1 the horizontal axis represents the regionalfinancial efficiency value of prefecture-level cities while thevertical axis represents the corresponding Kernel densityvalues

First there shows tendency toward left deviation atregional financial efficiency in Chinarsquos prefecture-level citiesin the view of the specific distribution form which indicatesthat the financial efficiency in most regions are below thenational average level Second themain peak level of financialefficiency gradually declines and moves to the left devia-tion during the three typical years (2003 2007 and 2012)which indicates that the level of financial efficiency graduallydecreases while the gap between the financial efficiency ofdifferent regions is widening In addition Chinarsquos regionalfinancial efficiency distribution form gradually evolves frommultipeak trend in 2003 to the double-peak trend in 2012

Table 1 Moranrsquos 119868 index of regional financial efficiency

Year Geographic distanceweighting matrix

Mixed geography-economyweighting matrix

2003 0152lowastlowastlowast 0130lowastlowastlowast

2004 0094lowastlowastlowast 0061lowastlowastlowast

2005 0094lowastlowastlowast 0028lowast

2006 0135lowastlowastlowast 0092lowastlowastlowast

2007 0137lowastlowastlowast 0100lowastlowastlowast

2008 0075lowastlowastlowast 0057lowastlowastlowast

2009 0019lowastlowastlowast 0024lowastlowastlowast

2010 0141lowastlowastlowast 0088lowastlowastlowast

2011 0140lowastlowastlowast 0071lowastlowastlowast

2012 0149lowastlowastlowast 0052lowastlowastlowast

Note lowastlowastlowast and lowast respectively represent the significant level of 1 and 10

which indicates that the development trend of regionalfinancial efficiency shifts frommultipolarity mode to double-polarity mode

In order to investigate the spatial clustering pattern andthe evolution trend of financial efficiency in 281 prefecture-level cities covering the period from 2003 to 2012 it is neededto further calculate Moranrsquos 119868 index of regional financialefficiency by using global spatial autocorrelation methodMoranrsquos 119868 index is the correlation coefficient of observedvalues and spatial lagged variables and the value of indexranges fromminus1 to 1 If the regional financial efficiency presentsspatial positive correlation the value of Moranrsquos 119868 indexranges from 0 to 1 which indicates that the regions withsimilar observed values are collectively distributed in thespatial range The closer Moranrsquos 119868 index to 1 the strongerthe spatial positive correlation On the contrary the regionswith similar observed value are discretely distributed in thegeographical space The specific measure formula of globalMoranrsquos 119868 index is as follows [31]

Moranrsquos 119868 =

119899sum119899

119894=1sum119899

119895=1119908119894119895(119909119894minus 119909) (119909

119895minus 119909)

sum119899

119894sum119899

119895=1119908119894119895sum119899

119894=1(119909119894minus 119909)2

(3)

In (3) 119909119894and 119909

119895 respectively represent region 119894 and

region 119895 and 119899 represents the number of regions while 119908119894119895

represents the spatial weightmatrixWewillmeasure regionalfinancial efficiencywith two spatial weightmatrixes includinggeographic distance weighting matrix and mixed geography-economy weighting matrix in the view of robustness ofresearch results The specific construction methods of thesetwo spatial weight matrixes are as in Section 4

Table 1 lists Moranrsquos 119868 indexes of regional financial effi-ciency calculatedwith two spatial weightmatrixes Accordingto the estimated results the values of Moranrsquos 119868 index calcu-lated with two spatial weight matrixes are positive and passthe 10 significance test which indicates that there exists sig-nificant positive autocorrelation (spatial dependence) amongregional financial efficiency in the geographical space Thespatial distribution of regional financial efficiency is notrandom which shows that the regions with similar observedvalues are collectively distributed in the geographical space

Discrete Dynamics in Nature and Society 5

fe2003

Wz

z

minus2 minus1 0 1 2 3 4 5 6minus1

0

1

2

3

76

61

75135138

17

113

199

143

20

9414277118

117

33

185

125

116115

1371227

8

55

226

1501395149177178

183

97

18012110

146

181

44911746

179

317124

259526413226136410128158

221

260

219

871585

211

184253100

276

16112913115180

212

167

42

188

186

213

41

89252

215436242

223

83

1283419

12

10286108105781659332

187

1531884

190

119144148152

56

182

25579

227215

2611689274156

176

96133

22

40

126279

269

52

25821

114163

25765

198

170262

229236

166164

243

171

253

3570

204

27215949

47155

50

110

63

8899

235

249

201

54

248

169

210

14029

9

145

206

127

46

244

216

173453873

12010490

57

277233

37

240172112

162157

111

202

124

200

72

60

30134147

24126398160

123256

195

237

11

196

228

64

4369

268

23103107109

230

278

192

13082

27231247

224

23259267

214

207

239

5366139

48

3968

265

238

250

141

194

14

271

189218

275

197

270

220

254

106

24527458

222

208225

175

193

203

205

273251

217

1

209

5162

281

266280

24616

234

67

81

191

Moran scatterplot (Moranrsquos I = 0130)

Figure 2 Moranrsquos 119868 index scatter diagram of financial efficiency in2003

fe2012

Wz

z

minus2 minus1 0 1 2 3 4 5minus1

0

1

2

3

4

258259

180

21

22

181166

173162261257

19155

11

23188

54

187186

218

1492646152

142

220

158

55

125185

126

10418413

212214

12910

1542632332721077

20

237103

9

1592815

51

8211260100

213

37256144

53

312261795150

190

49

239116245

50

15133106261826161121

38

117

241113231

112

177

271

219

16023013744

199

222

1713228122138

210

174

27097

115

221

240153

75248

232143163

179

362940

223244

165

178

242101

23456

102

195

253

471574518

193

229246235

268

108128109

4698

11876273127120

13513242

148

94

139

96

114

7116879

48

77236

167

9389276

169

41

99

74

57

145172

9192

140

12

274

34

27717065133

217

183

24327

198

80

164176

250

16123

255

130

13684175

254

110249

86

204

4514635

262

209

12478252

1526743

60

119

215

131134

203

196

147

88

25

83

72265

269

216

3966

85

247141227

8782

275

278156

192

280

90

224

194

202

68

206

32

207200

6930

205

21111

58

62

189

63

64

5981

208

251

197

14

2387370

225

201

281

24105

279

266

191

267

Moran scatterplot (Moranrsquos I = 0052)

Figure 3 Moranrsquos 119868 index scatter diagram of financial efficiency in2012

Namely it presents a spatial correlation structure that theregions with high financial efficiency are inclined to beadjacent to other high-efficiency regions while the regionswith low efficiency are inclined to be adjacent to other low-efficiency regions

In order to depict the clustering pattern of regionalfinancial efficiency more intuitively we will draw Moranrsquos119868 index scatter diagrams of regional financial efficiency intwo typical years namely 2003 and 2012 with the use of theeconomic-geographical mixed weighting matrix

Moranrsquos 119868 index scatter diagrams divide financial effi-ciency clustering distribution into four quadrants of whichthe first quadrant (HH high efficiency-high spatial lag) andthird quadrant (LL low efficiency-low spatial lag) reflectpositive spatial correlation and the second quadrant (LHlow efficiency-high spatial lag) and fourth quadrant (HL highefficiency-low spatial lag) reflect negative spatial correlationAs shown in Figures 2 and 3 most regions are in thefirst quadrant and the third quadrant which further con-firmed that there exists significant positive spatial correlationbetween regional financial efficiency

3 Model Specification

In order to investigate how the fiscal decentralization influ-ences the financial efficiency and strategic competitionamong regional financial development econometric modelis initially expressed as

fe119894119905

= 1205730+ 120573119883119894119905

+ 120583119894+ ]119905+ 120576119894119905 (4)

In (4) 119894 and 119905 respectively indicate the region 119894 andyear 119905 120583

119894is a regional disturbance ]

119905is timing disturbance

120576119894119905is the random error fe

119894119905represents financial efficiency

and 119909119894119905contains fiscal expenditure decentralization revenue

decentralization and other controlled variablesFlowing of capital and technology across sectors and

practitionersrsquo communication and cooperation lead to obvi-ous clustering and rivalry among regional financial devel-opment which indicates that the financial developmentand efficiency in geographically neighboring regions or theregions with similar economic structure are interdependentrather than independent to some degree Considering thefiscal decentralization system strategic competition causedby public policy spillover and ldquofree riderrdquo behavior furtherenhance the spatial correlation of financial efficiency andfinancial development among regions we find that it isnecessary to introduce the interactive dependencies amongregions into the spatial econometric analysis framework toreflect the spatial spillover effects of financial developmentunder the fiscal decentralization system by constructing aspatial econometric model

It has been proved that there exists spatial correlationin regional financial efficiency by using Moran statisticalanalysis so we can generate spatial econometric models byintroducing spatial autocorrelation factors into the generalregression models According to different forms of errorshock on an observationrsquos spatial autocorrelation spatialeconometric models can be divided into the Spatial ErrorModel (SEM) and the Spatial Lag Model (SLM) In ourresearch the SEM assumes that spatial autocorrelation stemsfrom the error shock of neighboring regions on the depen-dent variables and examines the effects of neighboringregions under observation while the SLM assumes thatspatial autocorrelation stems from dependent variables andexamines the effects of financial efficiency in neighboringregions on local financial efficiency According to the prin-ciples of model construction the SEM reflects the indirectbandwagon effects caused by error terms whereas the SLMcontrols the direct bandwagon effects LeSage and Pace [32]further establish the Spatial Durbin Model (SDM) on thebasis of the SEM and the SLM The SDM includes spatial lagterms fromdependent variables and independent variables tocapture the spillover effects deriving from different variables

Wewill construct static Spatial DurbinModel (SDM) anddynamic Spatial Durbin Model (DSDM) to investigate howfiscal decentralization affects regional financial efficiencyMeanwhile in order to reflect the financial efficiencyrsquos pos-sible inertia characteristics in reality the lag item (fe

119894119905-1) offinancial efficiency will be introduced to the independent

6 Discrete Dynamics in Nature and Society

variables to measure the influence of self-reinforcing mecha-nism The specific models can be expressed as

fe119894119905

= 120588119882119894119895fe119894119905

+ 120573119883119894119905

+ 120579119882119894119895119883119894119905

+ 120583119894+ ]119905+ 120576119894119905

fe119894119905

= 120593fe119894119905-1 + 120588119882

119894119895fe119894119905

+ 120573119883119894119905

+ 120579119882119894119895119883119894119905

+ 120583119894+ ]119905

+ 120576119894119905

(5)

In (5) 120588 is the spatial lag coefficient reflecting thedirection and extent of how financial efficiency in theneighboring regions affects local financial efficiency and thecoefficient directly reflects the degree of spatial competitionin the financial efficiency 119882

119894119895is the spatial weight matrix of

119873119879 times 119873119879 119873 is the number of cross-sectional samples (281prefecture-level cities) and 119879 is time (2003ndash2012)

4 Variables Selection and Data Sources

41 Variables Definition and Measurement Methods Accord-ing to research purpose we will focus on the relationshipbetween fiscal decentralization (fiscal expenditure decentral-ization and revenue decentralization) and regional financialefficiency in the subnational citiesOther economic and socialfactors affecting regional financial efficiency include regionaleconomic development urbanization industrial structureinformation technology development fixed asset investmentand the level of internal openingTherefore wewill take theseindexes as controlled variables and introduce them into anempirical analysis model the detailed variable setting andmeasuring methods are as follows

411 The Dependent Variable The dependent variable isfinancial efficiency (fe) which is measured by the aforemen-tioned SBMmodel

412 Main Independent Variables Fiscal expenditure decen-tralization and revenue decentralization are our main inde-pendent variable When it comes to the measure method offiscal decentralization we mainly use research results of Guoand Jia [33] and Gong and Lei [34] for reference The specificmeasure formula of fiscal expenditure decentralization (sd) isas follows

sd

=

PFEcity

PFEcity + PFEprovince + PFEnation[1 minus

GDPcity

GDPnation]

(6)

In (6) PFEcity PFEprovince and PFEnation respectivelyrepresent per capita fiscal expenditure of each city theprovincial per capita fiscal expenditure and the national percapita fiscal expenditure GDPcity and GDPnation respectivelyrepresent the GDP in each city and the national GDP[1 minus GDPcityGDPnation] is the reduction factor for economicscale The index has proposed the effects of population scale

and economic scale The specific measure formula of fiscalrevenue decentralization (rd) is as follows

rd

=

PFRcity

PFRcity + PFRprovince + PFRnation[1 minus

GDPcity

GDPnation]

(7)

In (7) PFRcity PFRprovince and PFRnation respectivelyrepresent per capita fiscal revenue of each city the provincialper capita fiscal revenue and the national per capita fiscalrevenue GDPcity and GDPnation respectively represent theGDP in each city and the national GDP

413 Controlled Variable Controlled variables are as follows(1) Regional economic development (119901119892119889119901) regional

economy provides the necessary material and nonmaterialsupport for the financial development and innovation thehigher the level of regional economic development is themore beneficial to financial efficiency the improvement isIn this paper per capita actual GDP is used as a proxyvariable for regional economic development To exclude theinfluence of price factor the foregoing data that involve valueare adjusted in accordance with constant 2003 prices

(2) Urbanization (119906119903119887119886119899) Urbanization of population(the proportion of nonagricultural population accounts fortotal population) is used as the proxy variable in which thedata of nonagricultural population covering the period from2003 to 2008 is from ldquoChina Urban Statistical Yearbookrdquo(2004ndash2009) and the data of nonagricultural populationcovering the period from 2009 to 2012 is from the annualldquoStatistics on Population of Counties and Cities in Chinardquo

(3) Industrial structure (119894119905119906) the upgrading and adjust-ment of regional industrial structure constitutes a soft envi-ronment for financial development and innovation Thehigher the proportion of the third industry is the morebeneficial to financial efficiency the improvement is In thispaper the index of industrial structure upgrading is usedto represent industrial structure and the specific measureformula is as follows

itu =

119899

sum

119895=1

119902 (119895) lowast 119895 (8)

In (8) 119902(119895) is the proportion of the third industry accountsfor local GDP in region 119895 119899 = 3 and the value of index rangesfrom 1 to 3

(4) Information technology development (119894119899119905119890119903119899119890119905) it ismeasured by the number of internet users in each city

(5) Fixed asset investment (119894119891119886) it is measured by theactual fixed asset investment in each city

(6)The level of internal opening (119901119903119894V119886119905119890) with referenceto the research result of Fan et al [35] we use the developmentof private enterprises to measure the level of internal openingand assume that the higher the level of the developmentof private enterprise is the stronger the flexibility of localfinancial innovation activities is which is more conduciveto financial efficiency improvement The proportion of theemployees of private enterprises and self-employed employ-ees account for the local total population is used as a proxy to

Discrete Dynamics in Nature and Society 7

measure the development of private enterprises in differentregions

414 Spatial Weight Matrixes Strategic competition tendsto occur in the regions with the similar level of economicdevelopment simultaneously the concentration effect willgradually decay with the geographical distance [36] In orderto comprehensively reflect the spatial effects of geographicaland economic characteristics and accurately measure thespatial correlation among regions we will construct themixed geography-economy weighting matrix The specificmeasure formula is as follows

119882eco-geo =

10038161003816100381610038161003816119876119894minus 119876119895

10038161003816100381610038161003816

119889119894119895

2119894 = 119895

0 119894 = 119895

(9)

In (9) 119882eco-geo is mixed geography-economy weightingmatrix119876

119894and119876

119895 respectively represent the per capita actual

GDP of city 119894 and city 119895 (the mean of the actual GDP coveringthe period from 2003 to 2012) and 119889

119894119895is the geographical

distance between city 119894 and city 119895 calculated with the latitudeand longitude data of prefecture-level cityrsquos administrativeunit The specific measure formula is as follows

119889119894119895

= 119877 arccos (sin120593119860sin120593119861

+ cos120593119860cos120593119861cos (120582

119861minus 120582119860))

(10)

In (10) 119877 is the earth circle radius (6378 km) and 120582 and120593 respectively represent administrative centerrsquos longitudeand latitude of city 119894 and city 119895 Longitude and latitude dataare from the national fundamental geographic informationsystem All of the diagonal elements of matrix 119882eco-geo are0 And the weight matrix is normalized so that the sum ofline elements is 1 In addition considering the reliability androbustness of statistical analysis we will also construct geo-graphical distance weight matrix tomake a spatial economet-ric analysis and then compare the corresponding estimatedresultsThe specificmeasure formula of geographical distanceweight matrix is as follows

119882geo =

1

119889119894119895

2119894 = 119895

0 119894 = 119895

(11)

In (11) 119882geo is a geographical distance weighting matrixwhich is also normalized so that the sum of line elements is 1and other indexes are consistent with that of (10)

42 Data Sources and Description Considering the avail-ability of data and consistency of research samples weeventually choose 281 prefecture-level cities (Hong KongMacao and Taiwan are excluded) as samples and excludeBazhong Ziyang Lhasa Jinchang Baiyin Zhongwei andother cities due to a serious lack of available dataThe researchperiod is covering the period from 2003 to 2012 and thedata is from ldquoChina City Statistical Yearbookrdquo ldquoRegional

Economic China Statistical Yearbookrdquo and provincial statis-tical yearbooks covering the period from 2004 to 2013 Themissing data was partially filled by interpolation In orderto weaken the influence of heteroskedasticity outliers andanomaly terms on the data stability all the data are processedby the logarithm An overview of the descriptive statistics ofthese data is provided in Table 2

In order to avoid the deviation of the model settingand spurious regression we carry on unit root testing onrelated variables by LLC and Fisher-ADF testing methodsbefore spatial econometric regression analysis The resultsof stationary testing of panel data indicate that all variablesreject the null hypothesis of unit root which passes the testat the 5 significance level Moreover the panel data arestationary and the regression results will be significantlyreliable

5 Empirical Methodology

51 Spatial Econometric Analysis of Fiscal Decentralization onRegional Financial Efficiency Since the SDMmodel containsthe spatial lag item of the dependent variable and the spatiallag items of independent variables at the same time so itcanmore fully reflect the impact of spatial autocorrelation onthe regression results than SEM model and SLM model [32]Therefore we construct SDMmodels to test the spatial effectsof the fiscal revenue decentralization and fiscal expendituredecentralization on the regional financial efficiency The 119875

values of Hausman test of the spatial econometric models areless than 005 which indicates that the estimated results offixed effect estimation are better than that of random effectestimation Therefore we will specifically carry out analysisaccording to the estimated results of spatially fixed effectmodels in the following parts Meanwhile a robust test iscarried by constructing DSDM models It is found that theestimated coefficients of SDMmodel and DSDMmodel havethe same direction while the values of estimated coefficientsand level of significance are also very close indicating that theresults of SDM model are significantly reliable Comparingthe values of 119877

2 and Log-likelihood of SDM model andDSDM model we find that the fitting degree of SDM modelis higher than that of DSDMmodel so the following analysisis mainly based on the econometric results of SDMmodel

As shown in Table 3 the overall spatial correlationcoefficients 120588 of four models are positive and pass thetests at the 1 significance level indicating that there arepositive spatial dependent relationships between the localfinancial efficiency and the neighboring regionsrsquo financialefficiency that is the economic activities of geographyneighborhood regions or regions with similar level of eco-nomic development have obvious spillover effect on localfinancial efficiency thereby causing imitation and strate-gic competition of financial development and innovationamong regions On one hand the geographical adjacencyand similarity of economic development in different regionsbring huge convenience to cooperation between financialenterprises sharing of infrastructure information exchangeand communication innovation and diffusion of knowledge

8 Discrete Dynamics in Nature and Society

Table 2 Statistical description of samplersquos variables

Index Variables Mean Standarddeviation Minimum 25 quantile 50 quantile 75 quantile Maximum

Finance efficiency fe 0576 0624 0133 0380 0500 0653 3958Fiscal expendituredecentralization sd 0368 0108 0138 0294 0351 0421 0888

Fiscal revenuedecentralization rd 0242 0146 0037 0133 0202 0318 0871

Per capita real GDP(yuanperson) pgdp 27119890 + 04 32119890 + 04 1891200 9837380 17119890 + 04 31119890 + 04 44119890 + 05

Urbanization () Urban 0345 0195 0013 0202 0296 0451 100000Industrial structure itu 2203 0136 1318 2110 2190 2283 2797Informationtechnologydevelopment (tenthousand families)

Internet 41066 128506 0024 8316 17139 36800 5174000

Fixed assetinvestment (tenthousand yuan)

ifa 15119890 + 04 15119890 + 04 654299 4606420 9841960 20119890 + 04 17119890 + 05

The level of internalopening () Private 0086 0106 0003 0036 0059 0096 100000

and technology flowing of financial resources among regionswhich causes regional financial development and innovationand imitation and strategic competition among regions [37]on the other hand the restriction of transportation cost onfinancial sector is not strong and the rapid development ofinternet technology leads to cross-regional financial transac-tions and innovation thus further expanding the clusteringeffect effect of economic scale and spatial spillover effectsof financial development The coefficients of lag item (fe

119894119905-1)of financial efficiency in DSDM models are positive andthey are respectively 0211 and 0212 both of them pass thetests at the 1 significance level indicating that the previousinputs and outputs of production will be present through thefinancial efficiency increasing in later stage (or later period)and financial efficiency increasing is a dynamic systemicprocess The specific impact of each variable on financialefficiency is analyzed as follows

The estimated coefficients of Lnsd and Lnrd are positiveand all of them pass the tests at the 1 significance levelwhich indicates that the fiscal expenditure and revenuedecentralization play a significant positive role in promotinglocal financial efficiency Under the dual pressures of makingup the fiscal gap and political promotion the local govern-ments regarding economic development as main missionhave instinctive strong interest in scarce flowing economicfactors [38] As the excellent combination of capital talenttechnology management and marketing financial resourceswill be the important target for battling among local govern-ments With the expansion of fiscal expenditure decentral-ization and revenue decentralization local government usesmany policy tools such as fiscal subsidies tax concessionsand government procurement to compete for getting thefinancial resources by attracting financial capital humancapital and technological innovation to flow into the localregion thus contributing to the financial innovation and

improvement of financial efficiencyThe influence coefficientof fiscal expenditure decentralization is 0357 greater thanthe influence coefficient (0153) of fiscal revenue decentral-ization indicating that the impacts of fiscal expendituredecentralization and revenue decentralization on regionalfinancial efficiency have unbalanced characteristics and thepossible reason is that local governments under provinceget very limited taxation autonomy resulting in the factthat revenue decentralization is often lower than expendi-ture decentralization In addition under the capital-driveneconomic growth model the flowing direction of capitaldepends on the level of infrastructure [39] thus the inflowof financial resources as well as financial innovation oftenoccurs in the regions with rapid economic developmentexcellent investment environment and great consumptionability so that local governments have to increase fiscalexpenditure to enhance the supply of the associated support-ing infrastructure Comparatively speaking tax preferenceof undeveloped regions is less attractive than the superiorityconditions of developed regions

We will further analyze the estimated coefficients andsignificance levels of spatial lags of fiscal expenditure andrevenue decentralization and investigate the strategic com-petitive effects of financial efficiency among neighboringregions The spatial lag of expenditure decentralization(WLnsd) does not pass the significance test suggestingthat spatial spillover effect of fiscal expenditure decentral-ization is uncertain and needs further examination whilethe spatial lag of fiscal revenue decentralization (WLnrd)is negative at the 5 significance level and the estimatedcoefficient is minus0228 indicating that with the increase offiscal revenue decentralization the negative externalities onfinancial development and innovation will be increasinglyprominent thus inhibiting the improvement of financialefficiency in neighboring regions The reason may be that

Discrete Dynamics in Nature and Society 9

Table 3 Estimated results of regional financial efficiency under theSpatial Durbin Model (economic-geographic weighting matrix)

Independentvariables

Dependent variable sd Dependent variable rd

SDMmodel DSDMmodel SDMmodel DSDM

model

feitminus10211lowastlowastlowast(1283)

0212lowastlowastlowast(1283)

Lnsd 0357lowastlowastlowast(626)

0334lowastlowastlowast(570)

Lnrd 0153lowastlowastlowast(470)

0143lowastlowastlowast(408)

Lnpgdp 0169lowastlowastlowast(462)

0120lowastlowastlowast(308)

0138lowastlowastlowast(360)

0099lowastlowast(244)

Lnurban minus0050lowastlowast(minus237)

minus0044lowastlowast(minus209)

minus0050lowastlowast(minus237)

minus0045lowastlowast(minus215)

Lnitu 0433lowastlowast(227)

0438lowastlowast(215)

0525lowastlowastlowast(273)

0531lowastlowastlowast(259)

Lninternet minus0295lowastlowastlowast(minus2970)

minus0299lowastlowastlowast(minus2716)

minus0292lowastlowastlowast(minus2945)

minus0297lowastlowastlowast(minus2690)

Lnifa minus0446lowastlowastlowast(minus2298)

minus0386lowastlowastlowast(minus1832)

minus0437lowastlowastlowast(minus2321)

minus0374lowastlowastlowast(minus1806)

Lnprivate 0038lowastlowastlowast(322)

0037lowastlowastlowast(290)

0039lowastlowastlowast(325)

0038lowastlowastlowast(302)

WLnsd 0179(099)

minus0041(minus022)

WLnrd minus0228lowastlowast(minus251)

minus0186lowast(minus179)

WLnpgdp 0016(027)

0064(101)

0058(096)

0097(150)

WLnurban 0097lowastlowast(236)

0082lowastlowast(198)

0107lowastlowastlowast(259)

0089lowastlowast(213)

WLnitu 1159lowast(193)

0516(082)

0615(100)

0228(035)

WLninternet 0177lowastlowastlowast(647)

0181lowastlowastlowast(622)

0180lowastlowastlowast(666)

0180lowastlowastlowast(619)

WLnifa 0237lowastlowastlowast(539)

0251lowastlowastlowast(532)

0275lowastlowastlowast(704)

0260lowastlowastlowast(600)

WLnprivate 0065lowast(185)

0045(115)

0083lowastlowast(232)

0060(153)

120588

0420lowastlowastlowast(1165)

0462lowastlowastlowast(1251)

0429lowastlowastlowast(1205)

0463lowastlowastlowast(1266)

1198772 0448 0443 0441 0438

log 119871 599812 595228 595228 587624Observations 2810 2529 2810 2529Note the values in parentheses represent the 119905-statistics values of thecorresponding estimated coefficientslowastlowastlowastlowastlowast andlowast respectively representsignificance level of 1 5 and 10

local governmentsrsquo motivation of pursuing revenues underthe fiscal revenue decentralization system is not conduciveto free flowing of financial resources among regions whichdirectly affects the financial efficiency in other regions

When it comes to the estimated coefficients of othercontrolled variables we found that not all the spatial spillovereffects of variables are significant Overall various economicactivities have direct and indirect influence on financial

efficiency and promote the strategic financial competitionamong regions The specific effects are as follows the levelof open-into-inside can not only improve the local financialefficiency significantly but also increase the financial effi-ciency correspondingly in neighboring regions The strongdemand for various financial products and services of privateenterprises promotes financial innovation and improves theefficiency of resource allocation in financial markets Onthe contrary urbanization information technology develop-ment and fixed assets investment suppress the improvementof local financial efficiency but play a positive role in promot-ing the financial efficiency in neighboring regionsThe accel-erating of urbanization process the application of internettechnology and investment in fixed assets promote economicexchanges and cooperation among regions so that the spatialspillover effects of financial development and innovation areincreasingly significant Furthermore although the regionaleconomic development and industrial structure significantlycontribute to the improvement of regional financial efficiencyin general the spillover effects among regions are not obvious

52 Decomposition of Fiscal Decentralizationrsquos Effects onRegional Financial Efficiency Whether there truly existspillover effects among the variables relying solely on the spa-tial lag coefficients of SDMmodel and DSDMmodel is likelyto lead to misinterpretation of the estimated results Spatialeffects can be further decomposed into direct effect indirecteffect (spillover effects) and total effect with reference to theresearch result of LeSage and Pace [32] The indirect effectsrepresent the independent variablesrsquo influence on the depen-dent variable in all other regions through spatial interactionsTherefore we will further verify the existence of spillovereffects by using partial differential matrix analysis Table 4illustrates the direct effect indirect effect and total effect ofthe fiscal expenditure and revenue decentralization in SDMmodel The results show that the estimated coefficients ofindependent variablesrsquo direct effect have the same directionwith the estimated coefficients of SDMmodel in Table 3 andthe estimated coefficients and significance levels are also veryclose

According to the estimated results of SDM model inTable 4 the direct effect indirect effect and total effectof fiscal expenditure decentralization on regional financialefficiency are positive and pass the 10 significance leveltest The regression coefficient of direct effect is 0368 theregression coefficient of indirect effect is 0560 and theregression coefficient of total effect is 0928 indicating thatthe degree of fiscal expenditure decentralization increases by1 while regional financial efficiency increases by 0928Fiscal expenditure decentralization plays a significant posi-tive impact on financial development and financial efficiencywhich is opposite to the research result of Wang and Li [23]The opposite conclusions are possibly due to the differencesof sample selection and indexesThe rise of fiscal expendituredecentralization in the local region leads to an increaseby 0368 of the financial efficiency while the increase ofexpenditure decentralization in the neighboring region leadsto financial efficiency increasing by 0560 indicating that

10 Discrete Dynamics in Nature and Society

Table 4 Decomposition of direct effect indirect effect and total effect under SDMmodel

Independent variablesDependent variable sd Dependent variable sd

Direct effect Indirect effect Total effect Direct effect Indirect effect Total effect

Lnsd 0368lowastlowastlowast(750)

0560lowast(195)

0928lowastlowastlowast(310)

Lnrd 0146lowastlowastlowast(529)

minus0283lowast(minus195)

minus0137lowast(minus192)

Lnpgdp 0175lowastlowastlowast(435)

0132(153)

0307lowastlowastlowast(359)

0145lowastlowastlowast(346)

0187lowastlowast(218)

0332lowastlowastlowast(392)

Lnurban minus0046lowastlowast(minus207)

0127lowastlowast(200)

0081(130)

minus0045lowastlowast(minus204)

0146lowastlowast(226)

0101(158)

Lnitu 0486lowastlowastlowast(275)

2321lowastlowast(237)

2807lowastlowastlowast(292)

0559lowastlowastlowast(315)

1488(151)

2047lowastlowast(211)

Lninternet minus0291lowastlowastlowast(minus3055)

0086lowast(190)

minus0205lowastlowastlowast(minus432)

minus0288lowastlowastlowast(minus3029)

0091lowastlowast(200)

minus0197lowastlowastlowast(minus410)

Lnifa minus0441lowastlowastlowast(minus2260)

0087(149)

minus0353lowastlowastlowast(minus594)

minus0430lowastlowastlowast(minus2254)

0154lowastlowastlowast(283)

minus0276lowastlowastlowast(minus494)

Lnprivate 0041lowastlowastlowast(319)

0134lowastlowast(235)

0175lowastlowastlowast(285)

0042lowastlowastlowast(327)

0166lowastlowastlowast(283)

0208lowastlowastlowast(330)

Note lowast lowast lowast lowastlowast and lowast respectively represent the significant level of 1 5 and 10

fiscal expenditure decentralization significantly promotesthe competition of financial development and innovationamong regions The expansion of the autonomy of localgovernment expenditure strengthens the regulatory capacityof local governments which is conducive to acceleratingmarket-oriented reform processing and achieves the organicintegration between fiscal system and market mechanism inthe allocation of financial resources which is beneficial tocreating a favorable institutional environment is also con-ducive to the flowing of financial resources among regionsand greatly promotes financial efficiency

The fiscal revenue decentralizationrsquos direct effect onregional financial efficiency is significantly positive whilethe spillover effect is significantly negative indicating thatthe impact of fiscal revenue decentralization on regionalfinancial efficiency presents the feature of ldquodouble-edged-swordrdquo Fiscal revenue decentralization has a significant posi-tive impact on the local financial efficiencyHowever there aresignificant negative spillover effects on the financial efficiencyin other regions The regression coefficient of the directeffect is 0146 the regression coefficient of the indirect effectis minus0283 and the regression coefficient of the total effectis minus0137 which indicates that the degree of fiscal revenuedecentralization increasing by 1 will lead to decrease by0137 of regional financial efficiency Furthermore whilethe increase of local fiscal revenue decentralization leads tofinancial efficiencyrsquos increase by 0146 the increase of thefiscal revenue decentralization in neighboring regions resultsin the financial efficiencyrsquos decrease by 0283 Comparingthe estimated coefficient of direct effect and that of indirecteffect it can be found that the negative spillover effectsof fiscal revenue decentralization are significantly greaterthan the direct scale effect so the negative externalitiesof fiscal revenue decentralization on financial efficiencyare greater than the positive externality On one hand inorder to obtain high-quality financial resources the local

governments ignore the differences like the level of economicdevelopment factor endowments and other aspects amongregions and seek to imitate policies and instruments inother regions thus leading to chain reaction of absorbinginvestment by other regions On the other hand whenthe neighboring regions or the regions with similar levelof economic development take some measures to attractliquidity factors the local government driven by yardstickcompetitionwill adopt similar policies to avoid the outflow ofelements which results in convergence of competitive tactics[40 41] Tax incentives reduction of fee and other wayswill become important means of competition and imitationbetween local governments in different regions on the basisof the independent revenue authority However such kind ofcompetitive mode between local governments for resourceswill consume huge fiscal resources and weaken the capacityof local governmentrsquos provision of public goods and servicesUltimately it is harmful to provide healthy stable andsustainable economic development environment for financialinnovation and increase of financial efficiency

53 Robustness Test To ensure the robustness and reliabilityof research results we will calculate the spatial lags withthe geographical distance weight matrix and do econometricregression by taking financial efficiency as the dependentvariable so that we can comparatively analyze the differenceof influence with different spatial weight setting based on theabove The robustness test results are as in Table 5

As shown in Table 5 the estimated results indicate thatthe sign of regression coefficients of the fiscal expendituredecentralization and revenue decentralization is consistentwith that of corresponding variable in the model of Table 3and the regression results in themodel are basically consistentwith the conclusions in Table 3 Comparing the estimatedresults based on two different weightmatrixes it is found thatthere are some differences in the influence coefficients of two

Discrete Dynamics in Nature and Society 11

Table 5 Robustness test on Spatial Durbin Models (geographicaldistance weight matrix)

Independentvariables

Dependent variable sd Dependent variable rdSDMmodel

DSDMmodel

SDMmodel

DSDMmodel

feitminus1 mdash 0202lowastlowastlowast(1276) mdash 0200lowastlowastlowast

(1255)

Lnsd or Lnrd 0364lowastlowastlowast(636)

0348lowastlowastlowast(592)

0173lowastlowastlowast(522)

0161lowastlowastlowast(458)

Lnpgdp 0268lowastlowastlowast(735)

0225lowastlowastlowast(578)

0242lowastlowastlowast(637)

0209lowastlowastlowast(522)

Lnurban minus0026(minus112)

minus0026(minus112)

minus0028(minus120)

minus0027(minus115)

Lnitu 0465lowastlowast(246)

0478lowastlowast(238)

0527lowastlowastlowast(278)

0542lowastlowastlowast(269)

Lninternet minus0295lowastlowastlowast(minus3038)

minus0300lowastlowastlowast(minus2799)

minus0292lowastlowastlowast(minus3005)

minus0298lowastlowastlowast(minus2771)

Lnifa minus0484lowastlowastlowast(minus2479)

minus0427lowastlowastlowast(minus2024)

minus0472lowastlowastlowast(minus2448)

minus0417lowastlowastlowast(minus1988)

Lnprivate 0036lowastlowastlowast(312)

0035lowastlowastlowast(284)

0035lowastlowastlowast(307)

0035lowastlowastlowast(288)

WLnsd orWLnrd

minus0264(minus149)

minus0418lowastlowast(minus225)

minus0277lowastlowastlowast(minus370)

minus0241lowastlowastlowast(minus276)

WLnpgdp minus0269lowastlowastlowast(minus427)

minus0286lowastlowastlowast(minus428)

minus0237lowastlowastlowast(minus369)

minus0267lowastlowastlowast(minus394)

WLnurban 0053(147)

0059lowast(167)

0059(163)

0062lowast(173)

WLnitu 0104(020)

minus0085(minus016)

minus0205(minus039)

minus0290(minus051)

WLninternet 0217lowastlowastlowast(838)

0239lowastlowastlowast(853)

0214lowastlowastlowast(831)

0235lowastlowastlowast(841)

WLnifa 0434lowastlowastlowast(1052)

0464lowastlowastlowast(1054)

0433lowastlowastlowast(1165)

0450lowastlowastlowast(1129)

WLnprivate 0062lowast(187)

0038(106)

0081lowastlowast(241)

0050(139)

120588

0547lowastlowastlowast(1858)

0580lowastlowastlowast(1938)

0541lowastlowastlowast(1839)

0570lowastlowastlowast(1898)

1198772 0444 0428 0443 0430

log 119871 659741 665804 665804 659509Observations 2810 2529 2810 2529Note lowastlowastlowast lowastlowast and lowast respectively represent the significant level of 1 5and 10

kinds of fiscal decentralizations indicators which show thatthe gap between the levels of economic developmentwill havean impact on role of fiscal decentralization acting on regionalfinancial efficiency On one hand under the fiscal decen-tralization system because of the heterogeneity of regionaleconomic development level the local governmentsrsquo ability ofmaking up the gap between fiscal revenue and expenditure isdifferent and the fiscal pressure is also different therefore thedegree of interference to the regional market environment bythe local government is different which leads to the regionaldifferences in financial efficiency in a certain degree [21]On the other hand the occurrence of financial innovationdepends on the support of the economic foundation and

the different economic and social environment will lead tothe fact that the financial innovation activities are more andmore concentrated in the regions with the advantages offinancial location resulting in nonequilibrium of financialefficiency in different regions [42] Under the performanceassessment method focusing on GDP local governments willnot only consider the strategic behaviors of the neighboringregions but also tend to think about the behaviors of theregions with the similar level of economy In additioncross-regional knowledge diffusion technology spillover andhuman capital flow are more prone to occur among theregions with similar level of economic development so thatthe financial exchanges and contacts among regions are morefrequent It also shows that besides considering the impact ofgeographical distance the economic distance factor shouldbe taken into consideration while we are researching thespatial effects of fiscal decentralization on regional financialinnovation in addition to the consideration of the impact ofgeographical distance the economic distance factor shouldbe considered

6 Conclusions and Enlightenments

Based on the panel data of 281 prefecture-level cities in Chinacovering the period of 2003ndash2012 the regional financialefficiency is measured by the superefficiency SBMmodel thegeographic distance weight matrix and economic-geographymixed weight matrix which can reflect the regional similarityare introduced into the SDM and the spatial effects of thefiscal expenditure and revenue decentralization on regionalfinancial efficiency under province are investigated as wellas the strategic financial competition arising from it Basedon the results of previous empirical analysis we can draw thefollowing basic conclusions

First the geographical distribution of Chinarsquos prefecture-level citiesrsquo financial efficiency presents significant features oftime inertia and spatial dependency Financial developmentand innovation among regions particularly neighboringregions is not independent but shows significant spatialspillover effects The financial efficiency of the local regionwill be significantly affected by the financial efficiency ofadjacent regions The spatial effect is increasingly obviousin the process of financial development and innovation andthe size of the spillover effects is closely related to spatialweight (which is used to measure regional similarities) andmainly shows that there is competition for financial efficiencyamong regions with similar economic attributes and adjacentregions

Second it cannot be ignored that the institutional factorsespecially fiscal decentralization have effect on the financialefficiency and the factors have a decisive role in the allocationof financial resource and the regional distribution of financialinnovation The improvement of fiscal decentralization levelwill help to attract more financial resources thereby promot-ing an increase in financial efficiencyThe improvement of thefinancial autonomy in neighboring jurisdictions will reducethe financial resources flowing into the local region andsuppresses the occurrence of the local financial innovationIn addition there exist significant spatial spillover effects of

12 Discrete Dynamics in Nature and Society

financial development and efficiency caused by fiscal decen-tralization The scope of spatial spillovers is mainly reflectedin two aspects Thus the local governmentsrsquo competition notonly strongly increases the financial efficiency of the regionbut also accelerates interregional competition for financialefficiency

Third fiscal expenditure decentralization and revenuedecentralizationrsquos impact on regional financial efficiencyhave nonequilibrium characteristics First of all there existsasymmetry between fiscal expenditure decentralization andrevenue decentralizationrsquos impact on regional financial effi-ciency Fiscal expenditure decentralization can promoteoptimization of the fiscal expenditure structure and fiscalrevenue decentralization can increase the local governmentsrsquorevenue share so as to provide positive incentives [43] bothof them have a positive and promoting effect on the improve-ment of local financial efficiency and such effect largelydepends on the symmetry between the fiscal expenditure andrevenue decentralization If it is asymmetric between theirproperty rights and the expenditure responsibilities underfiscal decentralization system the local governments maymake excessive intervention to regional financial resourcesallocation tomeet their own demands for financial resourcesNext there exists nonconformance between fiscal expendi-ture decentralization and revenue decentralizationrsquos impacton financial efficiency in neighboring regions Comparedwith the fiscal expenditure decentralization the fiscal revenuedecentralization has more apparent inhibitory effect on thefinancial efficiency in neighboring regions

Our research conclusions have important policy implica-tions that illuminate the relationship between fiscal decen-tralization and financial efficiency

(1) The regional financial efficiency is closely relatedto the location and the economic development of the sur-rounding cities The regional financial efficiency is affectednot only by the financial development and innovation inneighboring regions but also by the impact of the struc-tural differences among regions In this case breaking theadministrativemonopoly of local protectionism and carryingon the ldquorichening neighborrdquo policy of financial developmentare the best choices for the local government to developthe regional economy When making policies of financialdevelopment local governments need to consider their owneconomic development environment and economic featuresin neighboring regions and strengthen communication andcoordination with the surrounding regions so as to avoidthe possible contradictions and conflicts on financial devel-opment policies among regions

(2) The spatial spillover effects of the regional financialefficiency should be divided into two kinds of effects includ-ing global geographic spillover effects and local geographicspillover effects In order to promote the increase of financialefficiency in a better way it should also be based on thehierarchy of spatial effect of and the degree of spillover effectto define the relationship between the government and themarket and central and local governments and among thelocal governments and their role in regional financial devel-opment and innovation so as to better promote financialefficiency

(3) There is a two-way interactive mechanism betweenthe fiscal system and financial development such effectsmay affect the behaviors of the local governments and theallocation of resources and ultimately act on macroeconomicperformance [10] Under the fiscal decentralization systemthe strategic competitive behaviors among local governmentssignificantly affect the liquidity of financial resources and theoccurrence of financial innovation the local governmentsrsquoexpansion of fiscal expenditure and revenue shortfall becomethe main driving force of each region competing for financialresources In order to better play the role of incentivesand constraints of fiscal competition mechanism under thefiscal decentralization system and avoid the negative impactbrought by the vicious interregional strategic financial com-petitive behaviors on the regional economic development thelocal governmentsrsquo short-sighted behaviors should be strictlyregulated and then the interregional financial cooperationwith mutual benefit as the core should be actively promoted

(4) Constructing a scientific and rational framework forfiscal decentralization will be an important force to accelerateeffectively the improvement of regional financial efficiencyChinarsquos future subprovincial fiscal decentralization reformshould adapt to the new normal financial development andfurther rationalize the provincial and subprovincial fiscaldecentralization system to ensure a balanced and appropriatefiscal expenditure and revenue decentralization On onehand we should weaken local governmentsrsquo negative inter-vention in financial system by reducing local governmentsrsquofinancial pressure caused by the asymmetry between thefiscal resources and expenditure responsibilities On the otherhandwe should also actively play the function of fiscal systemto adjust the financial development and innovation and fullyplay public financersquos role in promoting the flowing of finan-cial resources among regions the balanced development ofregional finance and the improvement of financial efficiency

Competing Interests

There are no competing interests in regard to the contentdiscussed in this paper

Authorsrsquo Contributions

All authors conceived and designed the research XiaomeiHucollected and processed the data Jianmin Liu analyzed thedata and interpreted the results all authors were also involvedin writing the paper and have all approved the submittedform

Acknowledgments

The research of this paper is jointly funded by the NationalSocial Science Foundation of China under Project no14BJY159 and the Hunan Philosophy and Social ScienceFund Project under Grant no 13JD09 The authors gratefullyacknowledge the support of Jinguang Wu from HunanUniversity of Finance and Economics China

Discrete Dynamics in Nature and Society 13

References

[1] J Zhang and X Jin ldquoRe-examine of the relationship betweenChinarsquos financial deepening and productivity 1987ndash2001rdquo Eco-nomic Research Journal vol 11 pp 35ndash43 2005

[2] K Chen A L Hillman and Q Y Gu ldquoFiscal centralization andchanges of local governmentsrsquo behaviorsmdashfrom lsquohands for helprsquoto lsquohands for grabrsquordquoChina EconomicQuarterly vol 1 pp 111ndash1302002

[3] L Zhou and Z M Wang ldquoEmpirical analysis of regionalfinancial development of China and economic growth 1978ndash2000rdquo Financial Research vol 10 pp 1ndash13 2002

[4] L Zhou ldquolsquoSecond financersquo of Chinarsquos financial industry andfinancial division during the reforming periodrdquo The Journal ofWorld Economy vol 6 pp 72ndash79 2003

[5] L Zhou ldquoGradual transition state capacity and fiscal functionof financerdquo Journal of Finance and Economics vol 2 pp 26ndash372005

[6] E L He and W Liu ldquoFinancial capital local governmentsrsquointervention and industrial structuremdashan empirical analysisbased on the data of central regionsrdquo Economic Issues vol 5pp 37ndash40 2011

[7] X L Dou and W J Wang ldquoThe differences and trends of Chi-nese regional financial development and local governmentsrsquobehavior and analysis of tax sharing reform on the differencesand trends of Chinese regional financial developmentrdquo Shang-hai Finance vol 2 pp 9ndash15 2011

[8] J Zhang and K R Shen ldquoLocal government interventionregional financial development and the transformation ofChinarsquos economic growth modemdashan empirical study under thefiscal decentralization systemrdquo Nankai Economic Studies vol 6pp 122ndash141 2008

[9] G Chen X G Yin and Y Pan ldquoChinarsquos financial developmenttax sharing reform and economic growthrdquo Financial Researchvol 2 pp 99ndash109 2006

[10] P P Ding and Y Fu ldquoBehaviors of local governments therelationship of fiscal and finance and macroeconomic volatilityassociated with Chinamdashan analysis under Chinese-style fiscaldecentralization systemrdquo Comparative Economic amp Social Sys-tems vol 6 pp 87ndash97 2012

[11] X G Yin G Chen and Y Pan ldquoTax sharing reform local gov-ernmentsrsquo intervention and financial development efficiencyrdquoJournal of Finance and Economics vol 10 pp 92ndash101 2006

[12] M G Yu and H B Pan ldquoGovernment intervention the ruleof law financial development and bank loans of state-ownedenterprisesrdquo Financial Research vol 9 pp 1ndash23 2008

[13] D X Wang J Liu and L Li ldquoFiscal decentralization bankcredit and total factor productivityrdquo Journal of Finance andEconomics vol 4 pp 69ndash80 2011

[14] Y X Jiang and R B Jiang ldquoRegional financial innovationefficiency evaluation environmental impact and gap analysisrdquoJournal of Zhejiang University (Humanities and Social Sciences)vol 43 no 4 pp 52ndash65 2013

[15] H W Yang L W Cheng Y Xu and Y W Qi ldquoMeasure andinfluencing factors of financial resources allocation efficiencyunder fiscal decentralization system based on super-efficiencyDEA-TOBIT two-step methodrdquo Journal of Jiangxi NormalUniversity (Natural Sciences Edition) vol 6 pp 587ndash592 2014

[16] N Shen F Z Liu and J Q Zhao ldquoFiscal decentralizationfinancial deepening and regional international poverty prone todeveloprdquo Finance amp Trade Economics vol 1 pp 41ndash46 2006

[17] Q Y Du G P Li and N Shen ldquoFinancial efficiency andoptimization of the regional industrial structure under thecurrent fiscal systemrdquo Xinjiang Social Science vol 4 pp 16ndash212006

[18] T Feng Y W Song and Y Lu ldquoFiscal decentralization localgovernmentsrsquo behavior and regional financial developmentrdquoJournal of Xirsquoan Jiaotong University vol 5 pp 20ndash27 2007

[19] Y A Zhou X J Ma and J Y Zhao ldquoGovernment behaviorsfinancial development and economic growthrdquo Henan SocialSciences vol 1 pp 55ndash60 2007

[20] C Hao ldquoDevelopment of financial intermediation and eco-nomic growth the Chinese experiencerdquo China EconomicReview vol 17 no 4 pp 347ndash362 2006

[21] J Zhang Fiscal Decentralization Regional Financial Devel-opment and Chinarsquos Economic Growth Performance NanjingUniversity Press Nanjing China 2011

[22] Y J Yao ldquoGovernment intervention the development of finan-cial intermediation and economic growthrdquo Journal of Financeand Economics vol 8 pp 52ndash58 2010

[23] X Wang and L Li ldquoFiscal decentralization and regional finan-cial development the theoretical and empirical analysis fromthe perspective of central governmentrdquo Public Finance Researchvol 4 pp 31ndash35 2012

[24] W L Zhang and C L Qin ldquoFinancial development fiscaldecentralization and regional economic differencesrdquo EconomicGeography vol 1 pp 583ndash588 2009

[25] Y Ma J Li and B Chen ldquoProvincial differences in Chinesefiscal decentralization financial development industrializationand economic growthrdquo EconomicTheory and Business Manage-ment vol 2 pp 5ndash19 2015

[26] Y X Jiang and R B Jiang ldquoThe spatial external effects ofregional financial innovation financial competition and finan-cial agglomerationrdquo Social Science Front vol 3 pp 79ndash84 2014

[27] K Tone ldquoA slacks-based measure of super-efficiency indata envelopment analysisrdquo European Journal of OperationalResearch vol 143 no 1 pp 32ndash41 2002

[28] Y Q Lu and D G Zhang ldquoChinarsquos regional financial efficiencyand differences of efficiency measurementrdquo Economic Geogra-phy vol 1 pp 96ndash101 2012

[29] A N Berger and D B Humphrey ldquoEfficiency of financialinstitutions international survey and directions for futureresearchrdquo European Journal of Operational Research vol 98 no2 pp 175ndash212 1997

[30] Y J Yao ldquoEmpirical analysis of financial development andurban-rural income gaprdquo Journal of Finance and Economics vol2 pp 49ndash59 2005

[31] Q Li J Song E Wang H Hu J Zhang and Y WangldquoEconomic growth and pollutant emissions in China a spatialeconometric analysisrdquo Stochastic Environmental Research andRisk Assessment vol 28 no 2 pp 429ndash442 2014

[32] J P LeSage and R K Pace Introduction to Spatial EconometricsCRCPress Taylor amp Francis Group Boca Raton Fla USA 2009

[33] Q W Guo and J X Jia ldquoFiscal decentralization organizationalstructure of local governments and the size of governmentexpenditurerdquoEconomic Research Journal vol 11 pp 59ndash87 2010httpwwwcnkicomcnArticleCJFDTotal-JJYJ201011008htm

[34] F Gong and X Lei ldquoThe measurement of Chinese-style fiscaldecentralizationrdquo Statistical Research vol 10 pp 47ndash55 2010

[35] G Fan X L Wang and H P Zhu Chinese Marketing IndexRegional Marketing Process in 2011 Economic Science PressBeijing China 2011

14 Discrete Dynamics in Nature and Society

[36] S S Rosenthal and W C Strange ldquoGeography industrialorganization and agglomerationrdquoTheReview of Economics andStatistics vol 85 no 2 pp 377ndash393 2003

[37] S Park Yoon International Banking and Financial CentersKluwer Academic Boston Mass USA 1989

[38] B Y Qiao J Y Fan and X Y Feng ldquoChinarsquos fiscal decentral-ization and compulsory primary educationrdquo Social Sciences inChina vol 6 pp 37ndash46 2005

[39] W JWang and C L Qin ldquoRegional differences in the behaviorsof local governments and growth effect of fiscal decentralizationbased on the analysis of judgment hypothesis and testingrdquoManagement World vol 1 pp 9ndash21 2008

[40] J K Brueckner ldquoStrategic interaction among governments anoverview of empirical studiesrdquo International Regional ScienceReview vol 26 no 2 pp 175ndash188 2003

[41] F Revelli ldquoOn spatial public finance empiricsrdquo International Taxand Public Finance vol 12 no 4 pp 475ndash492 2005

[42] N Shen and T Y He ldquoFinancial efficiency and regional eco-nomic growth from the perspective of fiscal decentralizationrdquoReform vol 1 pp 61ndash65 2006

[43] J M Liu X M Hu and J G Wu ldquoThe threshold effect ofsub-national fiscal decentralization revenue and expendituredecentralization on provincial industrial transformation andupgrading based on the data of 14 cities in Hunan ProvincerdquoPublic Finance Research vol 8 pp 49ndash52 2014

Submit your manuscripts athttpwwwhindawicom

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

MathematicsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Mathematical Problems in Engineering

Hindawi Publishing Corporationhttpwwwhindawicom

Differential EquationsInternational Journal of

Volume 2014

Applied MathematicsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Probability and StatisticsHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Mathematical PhysicsAdvances in

Complex AnalysisJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

OptimizationJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

CombinatoricsHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Operations ResearchAdvances in

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Function Spaces

Abstract and Applied AnalysisHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of Mathematics and Mathematical Sciences

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Algebra

Discrete Dynamics in Nature and Society

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Decision SciencesAdvances in

Discrete MathematicsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom

Volume 2014 Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Stochastic AnalysisInternational Journal of

Discrete Dynamics in Nature and Society 5

fe2003

Wz

z

minus2 minus1 0 1 2 3 4 5 6minus1

0

1

2

3

76

61

75135138

17

113

199

143

20

9414277118

117

33

185

125

116115

1371227

8

55

226

1501395149177178

183

97

18012110

146

181

44911746

179

317124

259526413226136410128158

221

260

219

871585

211

184253100

276

16112913115180

212

167

42

188

186

213

41

89252

215436242

223

83

1283419

12

10286108105781659332

187

1531884

190

119144148152

56

182

25579

227215

2611689274156

176

96133

22

40

126279

269

52

25821

114163

25765

198

170262

229236

166164

243

171

253

3570

204

27215949

47155

50

110

63

8899

235

249

201

54

248

169

210

14029

9

145

206

127

46

244

216

173453873

12010490

57

277233

37

240172112

162157

111

202

124

200

72

60

30134147

24126398160

123256

195

237

11

196

228

64

4369

268

23103107109

230

278

192

13082

27231247

224

23259267

214

207

239

5366139

48

3968

265

238

250

141

194

14

271

189218

275

197

270

220

254

106

24527458

222

208225

175

193

203

205

273251

217

1

209

5162

281

266280

24616

234

67

81

191

Moran scatterplot (Moranrsquos I = 0130)

Figure 2 Moranrsquos 119868 index scatter diagram of financial efficiency in2003

fe2012

Wz

z

minus2 minus1 0 1 2 3 4 5minus1

0

1

2

3

4

258259

180

21

22

181166

173162261257

19155

11

23188

54

187186

218

1492646152

142

220

158

55

125185

126

10418413

212214

12910

1542632332721077

20

237103

9

1592815

51

8211260100

213

37256144

53

312261795150

190

49

239116245

50

15133106261826161121

38

117

241113231

112

177

271

219

16023013744

199

222

1713228122138

210

174

27097

115

221

240153

75248

232143163

179

362940

223244

165

178

242101

23456

102

195

253

471574518

193

229246235

268

108128109

4698

11876273127120

13513242

148

94

139

96

114

7116879

48

77236

167

9389276

169

41

99

74

57

145172

9192

140

12

274

34

27717065133

217

183

24327

198

80

164176

250

16123

255

130

13684175

254

110249

86

204

4514635

262

209

12478252

1526743

60

119

215

131134

203

196

147

88

25

83

72265

269

216

3966

85

247141227

8782

275

278156

192

280

90

224

194

202

68

206

32

207200

6930

205

21111

58

62

189

63

64

5981

208

251

197

14

2387370

225

201

281

24105

279

266

191

267

Moran scatterplot (Moranrsquos I = 0052)

Figure 3 Moranrsquos 119868 index scatter diagram of financial efficiency in2012

Namely it presents a spatial correlation structure that theregions with high financial efficiency are inclined to beadjacent to other high-efficiency regions while the regionswith low efficiency are inclined to be adjacent to other low-efficiency regions

In order to depict the clustering pattern of regionalfinancial efficiency more intuitively we will draw Moranrsquos119868 index scatter diagrams of regional financial efficiency intwo typical years namely 2003 and 2012 with the use of theeconomic-geographical mixed weighting matrix

Moranrsquos 119868 index scatter diagrams divide financial effi-ciency clustering distribution into four quadrants of whichthe first quadrant (HH high efficiency-high spatial lag) andthird quadrant (LL low efficiency-low spatial lag) reflectpositive spatial correlation and the second quadrant (LHlow efficiency-high spatial lag) and fourth quadrant (HL highefficiency-low spatial lag) reflect negative spatial correlationAs shown in Figures 2 and 3 most regions are in thefirst quadrant and the third quadrant which further con-firmed that there exists significant positive spatial correlationbetween regional financial efficiency

3 Model Specification

In order to investigate how the fiscal decentralization influ-ences the financial efficiency and strategic competitionamong regional financial development econometric modelis initially expressed as

fe119894119905

= 1205730+ 120573119883119894119905

+ 120583119894+ ]119905+ 120576119894119905 (4)

In (4) 119894 and 119905 respectively indicate the region 119894 andyear 119905 120583

119894is a regional disturbance ]

119905is timing disturbance

120576119894119905is the random error fe

119894119905represents financial efficiency

and 119909119894119905contains fiscal expenditure decentralization revenue

decentralization and other controlled variablesFlowing of capital and technology across sectors and

practitionersrsquo communication and cooperation lead to obvi-ous clustering and rivalry among regional financial devel-opment which indicates that the financial developmentand efficiency in geographically neighboring regions or theregions with similar economic structure are interdependentrather than independent to some degree Considering thefiscal decentralization system strategic competition causedby public policy spillover and ldquofree riderrdquo behavior furtherenhance the spatial correlation of financial efficiency andfinancial development among regions we find that it isnecessary to introduce the interactive dependencies amongregions into the spatial econometric analysis framework toreflect the spatial spillover effects of financial developmentunder the fiscal decentralization system by constructing aspatial econometric model

It has been proved that there exists spatial correlationin regional financial efficiency by using Moran statisticalanalysis so we can generate spatial econometric models byintroducing spatial autocorrelation factors into the generalregression models According to different forms of errorshock on an observationrsquos spatial autocorrelation spatialeconometric models can be divided into the Spatial ErrorModel (SEM) and the Spatial Lag Model (SLM) In ourresearch the SEM assumes that spatial autocorrelation stemsfrom the error shock of neighboring regions on the depen-dent variables and examines the effects of neighboringregions under observation while the SLM assumes thatspatial autocorrelation stems from dependent variables andexamines the effects of financial efficiency in neighboringregions on local financial efficiency According to the prin-ciples of model construction the SEM reflects the indirectbandwagon effects caused by error terms whereas the SLMcontrols the direct bandwagon effects LeSage and Pace [32]further establish the Spatial Durbin Model (SDM) on thebasis of the SEM and the SLM The SDM includes spatial lagterms fromdependent variables and independent variables tocapture the spillover effects deriving from different variables

Wewill construct static Spatial DurbinModel (SDM) anddynamic Spatial Durbin Model (DSDM) to investigate howfiscal decentralization affects regional financial efficiencyMeanwhile in order to reflect the financial efficiencyrsquos pos-sible inertia characteristics in reality the lag item (fe

119894119905-1) offinancial efficiency will be introduced to the independent

6 Discrete Dynamics in Nature and Society

variables to measure the influence of self-reinforcing mecha-nism The specific models can be expressed as

fe119894119905

= 120588119882119894119895fe119894119905

+ 120573119883119894119905

+ 120579119882119894119895119883119894119905

+ 120583119894+ ]119905+ 120576119894119905

fe119894119905

= 120593fe119894119905-1 + 120588119882

119894119895fe119894119905

+ 120573119883119894119905

+ 120579119882119894119895119883119894119905

+ 120583119894+ ]119905

+ 120576119894119905

(5)

In (5) 120588 is the spatial lag coefficient reflecting thedirection and extent of how financial efficiency in theneighboring regions affects local financial efficiency and thecoefficient directly reflects the degree of spatial competitionin the financial efficiency 119882

119894119895is the spatial weight matrix of

119873119879 times 119873119879 119873 is the number of cross-sectional samples (281prefecture-level cities) and 119879 is time (2003ndash2012)

4 Variables Selection and Data Sources

41 Variables Definition and Measurement Methods Accord-ing to research purpose we will focus on the relationshipbetween fiscal decentralization (fiscal expenditure decentral-ization and revenue decentralization) and regional financialefficiency in the subnational citiesOther economic and socialfactors affecting regional financial efficiency include regionaleconomic development urbanization industrial structureinformation technology development fixed asset investmentand the level of internal openingTherefore wewill take theseindexes as controlled variables and introduce them into anempirical analysis model the detailed variable setting andmeasuring methods are as follows

411 The Dependent Variable The dependent variable isfinancial efficiency (fe) which is measured by the aforemen-tioned SBMmodel

412 Main Independent Variables Fiscal expenditure decen-tralization and revenue decentralization are our main inde-pendent variable When it comes to the measure method offiscal decentralization we mainly use research results of Guoand Jia [33] and Gong and Lei [34] for reference The specificmeasure formula of fiscal expenditure decentralization (sd) isas follows

sd

=

PFEcity

PFEcity + PFEprovince + PFEnation[1 minus

GDPcity

GDPnation]

(6)

In (6) PFEcity PFEprovince and PFEnation respectivelyrepresent per capita fiscal expenditure of each city theprovincial per capita fiscal expenditure and the national percapita fiscal expenditure GDPcity and GDPnation respectivelyrepresent the GDP in each city and the national GDP[1 minus GDPcityGDPnation] is the reduction factor for economicscale The index has proposed the effects of population scale

and economic scale The specific measure formula of fiscalrevenue decentralization (rd) is as follows

rd

=

PFRcity

PFRcity + PFRprovince + PFRnation[1 minus

GDPcity

GDPnation]

(7)

In (7) PFRcity PFRprovince and PFRnation respectivelyrepresent per capita fiscal revenue of each city the provincialper capita fiscal revenue and the national per capita fiscalrevenue GDPcity and GDPnation respectively represent theGDP in each city and the national GDP

413 Controlled Variable Controlled variables are as follows(1) Regional economic development (119901119892119889119901) regional

economy provides the necessary material and nonmaterialsupport for the financial development and innovation thehigher the level of regional economic development is themore beneficial to financial efficiency the improvement isIn this paper per capita actual GDP is used as a proxyvariable for regional economic development To exclude theinfluence of price factor the foregoing data that involve valueare adjusted in accordance with constant 2003 prices

(2) Urbanization (119906119903119887119886119899) Urbanization of population(the proportion of nonagricultural population accounts fortotal population) is used as the proxy variable in which thedata of nonagricultural population covering the period from2003 to 2008 is from ldquoChina Urban Statistical Yearbookrdquo(2004ndash2009) and the data of nonagricultural populationcovering the period from 2009 to 2012 is from the annualldquoStatistics on Population of Counties and Cities in Chinardquo

(3) Industrial structure (119894119905119906) the upgrading and adjust-ment of regional industrial structure constitutes a soft envi-ronment for financial development and innovation Thehigher the proportion of the third industry is the morebeneficial to financial efficiency the improvement is In thispaper the index of industrial structure upgrading is usedto represent industrial structure and the specific measureformula is as follows

itu =

119899

sum

119895=1

119902 (119895) lowast 119895 (8)

In (8) 119902(119895) is the proportion of the third industry accountsfor local GDP in region 119895 119899 = 3 and the value of index rangesfrom 1 to 3

(4) Information technology development (119894119899119905119890119903119899119890119905) it ismeasured by the number of internet users in each city

(5) Fixed asset investment (119894119891119886) it is measured by theactual fixed asset investment in each city

(6)The level of internal opening (119901119903119894V119886119905119890) with referenceto the research result of Fan et al [35] we use the developmentof private enterprises to measure the level of internal openingand assume that the higher the level of the developmentof private enterprise is the stronger the flexibility of localfinancial innovation activities is which is more conduciveto financial efficiency improvement The proportion of theemployees of private enterprises and self-employed employ-ees account for the local total population is used as a proxy to

Discrete Dynamics in Nature and Society 7

measure the development of private enterprises in differentregions

414 Spatial Weight Matrixes Strategic competition tendsto occur in the regions with the similar level of economicdevelopment simultaneously the concentration effect willgradually decay with the geographical distance [36] In orderto comprehensively reflect the spatial effects of geographicaland economic characteristics and accurately measure thespatial correlation among regions we will construct themixed geography-economy weighting matrix The specificmeasure formula is as follows

119882eco-geo =

10038161003816100381610038161003816119876119894minus 119876119895

10038161003816100381610038161003816

119889119894119895

2119894 = 119895

0 119894 = 119895

(9)

In (9) 119882eco-geo is mixed geography-economy weightingmatrix119876

119894and119876

119895 respectively represent the per capita actual

GDP of city 119894 and city 119895 (the mean of the actual GDP coveringthe period from 2003 to 2012) and 119889

119894119895is the geographical

distance between city 119894 and city 119895 calculated with the latitudeand longitude data of prefecture-level cityrsquos administrativeunit The specific measure formula is as follows

119889119894119895

= 119877 arccos (sin120593119860sin120593119861

+ cos120593119860cos120593119861cos (120582

119861minus 120582119860))

(10)

In (10) 119877 is the earth circle radius (6378 km) and 120582 and120593 respectively represent administrative centerrsquos longitudeand latitude of city 119894 and city 119895 Longitude and latitude dataare from the national fundamental geographic informationsystem All of the diagonal elements of matrix 119882eco-geo are0 And the weight matrix is normalized so that the sum ofline elements is 1 In addition considering the reliability androbustness of statistical analysis we will also construct geo-graphical distance weight matrix tomake a spatial economet-ric analysis and then compare the corresponding estimatedresultsThe specificmeasure formula of geographical distanceweight matrix is as follows

119882geo =

1

119889119894119895

2119894 = 119895

0 119894 = 119895

(11)

In (11) 119882geo is a geographical distance weighting matrixwhich is also normalized so that the sum of line elements is 1and other indexes are consistent with that of (10)

42 Data Sources and Description Considering the avail-ability of data and consistency of research samples weeventually choose 281 prefecture-level cities (Hong KongMacao and Taiwan are excluded) as samples and excludeBazhong Ziyang Lhasa Jinchang Baiyin Zhongwei andother cities due to a serious lack of available dataThe researchperiod is covering the period from 2003 to 2012 and thedata is from ldquoChina City Statistical Yearbookrdquo ldquoRegional

Economic China Statistical Yearbookrdquo and provincial statis-tical yearbooks covering the period from 2004 to 2013 Themissing data was partially filled by interpolation In orderto weaken the influence of heteroskedasticity outliers andanomaly terms on the data stability all the data are processedby the logarithm An overview of the descriptive statistics ofthese data is provided in Table 2

In order to avoid the deviation of the model settingand spurious regression we carry on unit root testing onrelated variables by LLC and Fisher-ADF testing methodsbefore spatial econometric regression analysis The resultsof stationary testing of panel data indicate that all variablesreject the null hypothesis of unit root which passes the testat the 5 significance level Moreover the panel data arestationary and the regression results will be significantlyreliable

5 Empirical Methodology

51 Spatial Econometric Analysis of Fiscal Decentralization onRegional Financial Efficiency Since the SDMmodel containsthe spatial lag item of the dependent variable and the spatiallag items of independent variables at the same time so itcanmore fully reflect the impact of spatial autocorrelation onthe regression results than SEM model and SLM model [32]Therefore we construct SDMmodels to test the spatial effectsof the fiscal revenue decentralization and fiscal expendituredecentralization on the regional financial efficiency The 119875

values of Hausman test of the spatial econometric models areless than 005 which indicates that the estimated results offixed effect estimation are better than that of random effectestimation Therefore we will specifically carry out analysisaccording to the estimated results of spatially fixed effectmodels in the following parts Meanwhile a robust test iscarried by constructing DSDM models It is found that theestimated coefficients of SDMmodel and DSDMmodel havethe same direction while the values of estimated coefficientsand level of significance are also very close indicating that theresults of SDM model are significantly reliable Comparingthe values of 119877

2 and Log-likelihood of SDM model andDSDM model we find that the fitting degree of SDM modelis higher than that of DSDMmodel so the following analysisis mainly based on the econometric results of SDMmodel

As shown in Table 3 the overall spatial correlationcoefficients 120588 of four models are positive and pass thetests at the 1 significance level indicating that there arepositive spatial dependent relationships between the localfinancial efficiency and the neighboring regionsrsquo financialefficiency that is the economic activities of geographyneighborhood regions or regions with similar level of eco-nomic development have obvious spillover effect on localfinancial efficiency thereby causing imitation and strate-gic competition of financial development and innovationamong regions On one hand the geographical adjacencyand similarity of economic development in different regionsbring huge convenience to cooperation between financialenterprises sharing of infrastructure information exchangeand communication innovation and diffusion of knowledge

8 Discrete Dynamics in Nature and Society

Table 2 Statistical description of samplersquos variables

Index Variables Mean Standarddeviation Minimum 25 quantile 50 quantile 75 quantile Maximum

Finance efficiency fe 0576 0624 0133 0380 0500 0653 3958Fiscal expendituredecentralization sd 0368 0108 0138 0294 0351 0421 0888

Fiscal revenuedecentralization rd 0242 0146 0037 0133 0202 0318 0871

Per capita real GDP(yuanperson) pgdp 27119890 + 04 32119890 + 04 1891200 9837380 17119890 + 04 31119890 + 04 44119890 + 05

Urbanization () Urban 0345 0195 0013 0202 0296 0451 100000Industrial structure itu 2203 0136 1318 2110 2190 2283 2797Informationtechnologydevelopment (tenthousand families)

Internet 41066 128506 0024 8316 17139 36800 5174000

Fixed assetinvestment (tenthousand yuan)

ifa 15119890 + 04 15119890 + 04 654299 4606420 9841960 20119890 + 04 17119890 + 05

The level of internalopening () Private 0086 0106 0003 0036 0059 0096 100000

and technology flowing of financial resources among regionswhich causes regional financial development and innovationand imitation and strategic competition among regions [37]on the other hand the restriction of transportation cost onfinancial sector is not strong and the rapid development ofinternet technology leads to cross-regional financial transac-tions and innovation thus further expanding the clusteringeffect effect of economic scale and spatial spillover effectsof financial development The coefficients of lag item (fe

119894119905-1)of financial efficiency in DSDM models are positive andthey are respectively 0211 and 0212 both of them pass thetests at the 1 significance level indicating that the previousinputs and outputs of production will be present through thefinancial efficiency increasing in later stage (or later period)and financial efficiency increasing is a dynamic systemicprocess The specific impact of each variable on financialefficiency is analyzed as follows

The estimated coefficients of Lnsd and Lnrd are positiveand all of them pass the tests at the 1 significance levelwhich indicates that the fiscal expenditure and revenuedecentralization play a significant positive role in promotinglocal financial efficiency Under the dual pressures of makingup the fiscal gap and political promotion the local govern-ments regarding economic development as main missionhave instinctive strong interest in scarce flowing economicfactors [38] As the excellent combination of capital talenttechnology management and marketing financial resourceswill be the important target for battling among local govern-ments With the expansion of fiscal expenditure decentral-ization and revenue decentralization local government usesmany policy tools such as fiscal subsidies tax concessionsand government procurement to compete for getting thefinancial resources by attracting financial capital humancapital and technological innovation to flow into the localregion thus contributing to the financial innovation and

improvement of financial efficiencyThe influence coefficientof fiscal expenditure decentralization is 0357 greater thanthe influence coefficient (0153) of fiscal revenue decentral-ization indicating that the impacts of fiscal expendituredecentralization and revenue decentralization on regionalfinancial efficiency have unbalanced characteristics and thepossible reason is that local governments under provinceget very limited taxation autonomy resulting in the factthat revenue decentralization is often lower than expendi-ture decentralization In addition under the capital-driveneconomic growth model the flowing direction of capitaldepends on the level of infrastructure [39] thus the inflowof financial resources as well as financial innovation oftenoccurs in the regions with rapid economic developmentexcellent investment environment and great consumptionability so that local governments have to increase fiscalexpenditure to enhance the supply of the associated support-ing infrastructure Comparatively speaking tax preferenceof undeveloped regions is less attractive than the superiorityconditions of developed regions

We will further analyze the estimated coefficients andsignificance levels of spatial lags of fiscal expenditure andrevenue decentralization and investigate the strategic com-petitive effects of financial efficiency among neighboringregions The spatial lag of expenditure decentralization(WLnsd) does not pass the significance test suggestingthat spatial spillover effect of fiscal expenditure decentral-ization is uncertain and needs further examination whilethe spatial lag of fiscal revenue decentralization (WLnrd)is negative at the 5 significance level and the estimatedcoefficient is minus0228 indicating that with the increase offiscal revenue decentralization the negative externalities onfinancial development and innovation will be increasinglyprominent thus inhibiting the improvement of financialefficiency in neighboring regions The reason may be that

Discrete Dynamics in Nature and Society 9

Table 3 Estimated results of regional financial efficiency under theSpatial Durbin Model (economic-geographic weighting matrix)

Independentvariables

Dependent variable sd Dependent variable rd

SDMmodel DSDMmodel SDMmodel DSDM

model

feitminus10211lowastlowastlowast(1283)

0212lowastlowastlowast(1283)

Lnsd 0357lowastlowastlowast(626)

0334lowastlowastlowast(570)

Lnrd 0153lowastlowastlowast(470)

0143lowastlowastlowast(408)

Lnpgdp 0169lowastlowastlowast(462)

0120lowastlowastlowast(308)

0138lowastlowastlowast(360)

0099lowastlowast(244)

Lnurban minus0050lowastlowast(minus237)

minus0044lowastlowast(minus209)

minus0050lowastlowast(minus237)

minus0045lowastlowast(minus215)

Lnitu 0433lowastlowast(227)

0438lowastlowast(215)

0525lowastlowastlowast(273)

0531lowastlowastlowast(259)

Lninternet minus0295lowastlowastlowast(minus2970)

minus0299lowastlowastlowast(minus2716)

minus0292lowastlowastlowast(minus2945)

minus0297lowastlowastlowast(minus2690)

Lnifa minus0446lowastlowastlowast(minus2298)

minus0386lowastlowastlowast(minus1832)

minus0437lowastlowastlowast(minus2321)

minus0374lowastlowastlowast(minus1806)

Lnprivate 0038lowastlowastlowast(322)

0037lowastlowastlowast(290)

0039lowastlowastlowast(325)

0038lowastlowastlowast(302)

WLnsd 0179(099)

minus0041(minus022)

WLnrd minus0228lowastlowast(minus251)

minus0186lowast(minus179)

WLnpgdp 0016(027)

0064(101)

0058(096)

0097(150)

WLnurban 0097lowastlowast(236)

0082lowastlowast(198)

0107lowastlowastlowast(259)

0089lowastlowast(213)

WLnitu 1159lowast(193)

0516(082)

0615(100)

0228(035)

WLninternet 0177lowastlowastlowast(647)

0181lowastlowastlowast(622)

0180lowastlowastlowast(666)

0180lowastlowastlowast(619)

WLnifa 0237lowastlowastlowast(539)

0251lowastlowastlowast(532)

0275lowastlowastlowast(704)

0260lowastlowastlowast(600)

WLnprivate 0065lowast(185)

0045(115)

0083lowastlowast(232)

0060(153)

120588

0420lowastlowastlowast(1165)

0462lowastlowastlowast(1251)

0429lowastlowastlowast(1205)

0463lowastlowastlowast(1266)

1198772 0448 0443 0441 0438

log 119871 599812 595228 595228 587624Observations 2810 2529 2810 2529Note the values in parentheses represent the 119905-statistics values of thecorresponding estimated coefficientslowastlowastlowastlowastlowast andlowast respectively representsignificance level of 1 5 and 10

local governmentsrsquo motivation of pursuing revenues underthe fiscal revenue decentralization system is not conduciveto free flowing of financial resources among regions whichdirectly affects the financial efficiency in other regions

When it comes to the estimated coefficients of othercontrolled variables we found that not all the spatial spillovereffects of variables are significant Overall various economicactivities have direct and indirect influence on financial

efficiency and promote the strategic financial competitionamong regions The specific effects are as follows the levelof open-into-inside can not only improve the local financialefficiency significantly but also increase the financial effi-ciency correspondingly in neighboring regions The strongdemand for various financial products and services of privateenterprises promotes financial innovation and improves theefficiency of resource allocation in financial markets Onthe contrary urbanization information technology develop-ment and fixed assets investment suppress the improvementof local financial efficiency but play a positive role in promot-ing the financial efficiency in neighboring regionsThe accel-erating of urbanization process the application of internettechnology and investment in fixed assets promote economicexchanges and cooperation among regions so that the spatialspillover effects of financial development and innovation areincreasingly significant Furthermore although the regionaleconomic development and industrial structure significantlycontribute to the improvement of regional financial efficiencyin general the spillover effects among regions are not obvious

52 Decomposition of Fiscal Decentralizationrsquos Effects onRegional Financial Efficiency Whether there truly existspillover effects among the variables relying solely on the spa-tial lag coefficients of SDMmodel and DSDMmodel is likelyto lead to misinterpretation of the estimated results Spatialeffects can be further decomposed into direct effect indirecteffect (spillover effects) and total effect with reference to theresearch result of LeSage and Pace [32] The indirect effectsrepresent the independent variablesrsquo influence on the depen-dent variable in all other regions through spatial interactionsTherefore we will further verify the existence of spillovereffects by using partial differential matrix analysis Table 4illustrates the direct effect indirect effect and total effect ofthe fiscal expenditure and revenue decentralization in SDMmodel The results show that the estimated coefficients ofindependent variablesrsquo direct effect have the same directionwith the estimated coefficients of SDMmodel in Table 3 andthe estimated coefficients and significance levels are also veryclose

According to the estimated results of SDM model inTable 4 the direct effect indirect effect and total effectof fiscal expenditure decentralization on regional financialefficiency are positive and pass the 10 significance leveltest The regression coefficient of direct effect is 0368 theregression coefficient of indirect effect is 0560 and theregression coefficient of total effect is 0928 indicating thatthe degree of fiscal expenditure decentralization increases by1 while regional financial efficiency increases by 0928Fiscal expenditure decentralization plays a significant posi-tive impact on financial development and financial efficiencywhich is opposite to the research result of Wang and Li [23]The opposite conclusions are possibly due to the differencesof sample selection and indexesThe rise of fiscal expendituredecentralization in the local region leads to an increaseby 0368 of the financial efficiency while the increase ofexpenditure decentralization in the neighboring region leadsto financial efficiency increasing by 0560 indicating that

10 Discrete Dynamics in Nature and Society

Table 4 Decomposition of direct effect indirect effect and total effect under SDMmodel

Independent variablesDependent variable sd Dependent variable sd

Direct effect Indirect effect Total effect Direct effect Indirect effect Total effect

Lnsd 0368lowastlowastlowast(750)

0560lowast(195)

0928lowastlowastlowast(310)

Lnrd 0146lowastlowastlowast(529)

minus0283lowast(minus195)

minus0137lowast(minus192)

Lnpgdp 0175lowastlowastlowast(435)

0132(153)

0307lowastlowastlowast(359)

0145lowastlowastlowast(346)

0187lowastlowast(218)

0332lowastlowastlowast(392)

Lnurban minus0046lowastlowast(minus207)

0127lowastlowast(200)

0081(130)

minus0045lowastlowast(minus204)

0146lowastlowast(226)

0101(158)

Lnitu 0486lowastlowastlowast(275)

2321lowastlowast(237)

2807lowastlowastlowast(292)

0559lowastlowastlowast(315)

1488(151)

2047lowastlowast(211)

Lninternet minus0291lowastlowastlowast(minus3055)

0086lowast(190)

minus0205lowastlowastlowast(minus432)

minus0288lowastlowastlowast(minus3029)

0091lowastlowast(200)

minus0197lowastlowastlowast(minus410)

Lnifa minus0441lowastlowastlowast(minus2260)

0087(149)

minus0353lowastlowastlowast(minus594)

minus0430lowastlowastlowast(minus2254)

0154lowastlowastlowast(283)

minus0276lowastlowastlowast(minus494)

Lnprivate 0041lowastlowastlowast(319)

0134lowastlowast(235)

0175lowastlowastlowast(285)

0042lowastlowastlowast(327)

0166lowastlowastlowast(283)

0208lowastlowastlowast(330)

Note lowast lowast lowast lowastlowast and lowast respectively represent the significant level of 1 5 and 10

fiscal expenditure decentralization significantly promotesthe competition of financial development and innovationamong regions The expansion of the autonomy of localgovernment expenditure strengthens the regulatory capacityof local governments which is conducive to acceleratingmarket-oriented reform processing and achieves the organicintegration between fiscal system and market mechanism inthe allocation of financial resources which is beneficial tocreating a favorable institutional environment is also con-ducive to the flowing of financial resources among regionsand greatly promotes financial efficiency

The fiscal revenue decentralizationrsquos direct effect onregional financial efficiency is significantly positive whilethe spillover effect is significantly negative indicating thatthe impact of fiscal revenue decentralization on regionalfinancial efficiency presents the feature of ldquodouble-edged-swordrdquo Fiscal revenue decentralization has a significant posi-tive impact on the local financial efficiencyHowever there aresignificant negative spillover effects on the financial efficiencyin other regions The regression coefficient of the directeffect is 0146 the regression coefficient of the indirect effectis minus0283 and the regression coefficient of the total effectis minus0137 which indicates that the degree of fiscal revenuedecentralization increasing by 1 will lead to decrease by0137 of regional financial efficiency Furthermore whilethe increase of local fiscal revenue decentralization leads tofinancial efficiencyrsquos increase by 0146 the increase of thefiscal revenue decentralization in neighboring regions resultsin the financial efficiencyrsquos decrease by 0283 Comparingthe estimated coefficient of direct effect and that of indirecteffect it can be found that the negative spillover effectsof fiscal revenue decentralization are significantly greaterthan the direct scale effect so the negative externalitiesof fiscal revenue decentralization on financial efficiencyare greater than the positive externality On one hand inorder to obtain high-quality financial resources the local

governments ignore the differences like the level of economicdevelopment factor endowments and other aspects amongregions and seek to imitate policies and instruments inother regions thus leading to chain reaction of absorbinginvestment by other regions On the other hand whenthe neighboring regions or the regions with similar levelof economic development take some measures to attractliquidity factors the local government driven by yardstickcompetitionwill adopt similar policies to avoid the outflow ofelements which results in convergence of competitive tactics[40 41] Tax incentives reduction of fee and other wayswill become important means of competition and imitationbetween local governments in different regions on the basisof the independent revenue authority However such kind ofcompetitive mode between local governments for resourceswill consume huge fiscal resources and weaken the capacityof local governmentrsquos provision of public goods and servicesUltimately it is harmful to provide healthy stable andsustainable economic development environment for financialinnovation and increase of financial efficiency

53 Robustness Test To ensure the robustness and reliabilityof research results we will calculate the spatial lags withthe geographical distance weight matrix and do econometricregression by taking financial efficiency as the dependentvariable so that we can comparatively analyze the differenceof influence with different spatial weight setting based on theabove The robustness test results are as in Table 5

As shown in Table 5 the estimated results indicate thatthe sign of regression coefficients of the fiscal expendituredecentralization and revenue decentralization is consistentwith that of corresponding variable in the model of Table 3and the regression results in themodel are basically consistentwith the conclusions in Table 3 Comparing the estimatedresults based on two different weightmatrixes it is found thatthere are some differences in the influence coefficients of two

Discrete Dynamics in Nature and Society 11

Table 5 Robustness test on Spatial Durbin Models (geographicaldistance weight matrix)

Independentvariables

Dependent variable sd Dependent variable rdSDMmodel

DSDMmodel

SDMmodel

DSDMmodel

feitminus1 mdash 0202lowastlowastlowast(1276) mdash 0200lowastlowastlowast

(1255)

Lnsd or Lnrd 0364lowastlowastlowast(636)

0348lowastlowastlowast(592)

0173lowastlowastlowast(522)

0161lowastlowastlowast(458)

Lnpgdp 0268lowastlowastlowast(735)

0225lowastlowastlowast(578)

0242lowastlowastlowast(637)

0209lowastlowastlowast(522)

Lnurban minus0026(minus112)

minus0026(minus112)

minus0028(minus120)

minus0027(minus115)

Lnitu 0465lowastlowast(246)

0478lowastlowast(238)

0527lowastlowastlowast(278)

0542lowastlowastlowast(269)

Lninternet minus0295lowastlowastlowast(minus3038)

minus0300lowastlowastlowast(minus2799)

minus0292lowastlowastlowast(minus3005)

minus0298lowastlowastlowast(minus2771)

Lnifa minus0484lowastlowastlowast(minus2479)

minus0427lowastlowastlowast(minus2024)

minus0472lowastlowastlowast(minus2448)

minus0417lowastlowastlowast(minus1988)

Lnprivate 0036lowastlowastlowast(312)

0035lowastlowastlowast(284)

0035lowastlowastlowast(307)

0035lowastlowastlowast(288)

WLnsd orWLnrd

minus0264(minus149)

minus0418lowastlowast(minus225)

minus0277lowastlowastlowast(minus370)

minus0241lowastlowastlowast(minus276)

WLnpgdp minus0269lowastlowastlowast(minus427)

minus0286lowastlowastlowast(minus428)

minus0237lowastlowastlowast(minus369)

minus0267lowastlowastlowast(minus394)

WLnurban 0053(147)

0059lowast(167)

0059(163)

0062lowast(173)

WLnitu 0104(020)

minus0085(minus016)

minus0205(minus039)

minus0290(minus051)

WLninternet 0217lowastlowastlowast(838)

0239lowastlowastlowast(853)

0214lowastlowastlowast(831)

0235lowastlowastlowast(841)

WLnifa 0434lowastlowastlowast(1052)

0464lowastlowastlowast(1054)

0433lowastlowastlowast(1165)

0450lowastlowastlowast(1129)

WLnprivate 0062lowast(187)

0038(106)

0081lowastlowast(241)

0050(139)

120588

0547lowastlowastlowast(1858)

0580lowastlowastlowast(1938)

0541lowastlowastlowast(1839)

0570lowastlowastlowast(1898)

1198772 0444 0428 0443 0430

log 119871 659741 665804 665804 659509Observations 2810 2529 2810 2529Note lowastlowastlowast lowastlowast and lowast respectively represent the significant level of 1 5and 10

kinds of fiscal decentralizations indicators which show thatthe gap between the levels of economic developmentwill havean impact on role of fiscal decentralization acting on regionalfinancial efficiency On one hand under the fiscal decen-tralization system because of the heterogeneity of regionaleconomic development level the local governmentsrsquo ability ofmaking up the gap between fiscal revenue and expenditure isdifferent and the fiscal pressure is also different therefore thedegree of interference to the regional market environment bythe local government is different which leads to the regionaldifferences in financial efficiency in a certain degree [21]On the other hand the occurrence of financial innovationdepends on the support of the economic foundation and

the different economic and social environment will lead tothe fact that the financial innovation activities are more andmore concentrated in the regions with the advantages offinancial location resulting in nonequilibrium of financialefficiency in different regions [42] Under the performanceassessment method focusing on GDP local governments willnot only consider the strategic behaviors of the neighboringregions but also tend to think about the behaviors of theregions with the similar level of economy In additioncross-regional knowledge diffusion technology spillover andhuman capital flow are more prone to occur among theregions with similar level of economic development so thatthe financial exchanges and contacts among regions are morefrequent It also shows that besides considering the impact ofgeographical distance the economic distance factor shouldbe taken into consideration while we are researching thespatial effects of fiscal decentralization on regional financialinnovation in addition to the consideration of the impact ofgeographical distance the economic distance factor shouldbe considered

6 Conclusions and Enlightenments

Based on the panel data of 281 prefecture-level cities in Chinacovering the period of 2003ndash2012 the regional financialefficiency is measured by the superefficiency SBMmodel thegeographic distance weight matrix and economic-geographymixed weight matrix which can reflect the regional similarityare introduced into the SDM and the spatial effects of thefiscal expenditure and revenue decentralization on regionalfinancial efficiency under province are investigated as wellas the strategic financial competition arising from it Basedon the results of previous empirical analysis we can draw thefollowing basic conclusions

First the geographical distribution of Chinarsquos prefecture-level citiesrsquo financial efficiency presents significant features oftime inertia and spatial dependency Financial developmentand innovation among regions particularly neighboringregions is not independent but shows significant spatialspillover effects The financial efficiency of the local regionwill be significantly affected by the financial efficiency ofadjacent regions The spatial effect is increasingly obviousin the process of financial development and innovation andthe size of the spillover effects is closely related to spatialweight (which is used to measure regional similarities) andmainly shows that there is competition for financial efficiencyamong regions with similar economic attributes and adjacentregions

Second it cannot be ignored that the institutional factorsespecially fiscal decentralization have effect on the financialefficiency and the factors have a decisive role in the allocationof financial resource and the regional distribution of financialinnovation The improvement of fiscal decentralization levelwill help to attract more financial resources thereby promot-ing an increase in financial efficiencyThe improvement of thefinancial autonomy in neighboring jurisdictions will reducethe financial resources flowing into the local region andsuppresses the occurrence of the local financial innovationIn addition there exist significant spatial spillover effects of

12 Discrete Dynamics in Nature and Society

financial development and efficiency caused by fiscal decen-tralization The scope of spatial spillovers is mainly reflectedin two aspects Thus the local governmentsrsquo competition notonly strongly increases the financial efficiency of the regionbut also accelerates interregional competition for financialefficiency

Third fiscal expenditure decentralization and revenuedecentralizationrsquos impact on regional financial efficiencyhave nonequilibrium characteristics First of all there existsasymmetry between fiscal expenditure decentralization andrevenue decentralizationrsquos impact on regional financial effi-ciency Fiscal expenditure decentralization can promoteoptimization of the fiscal expenditure structure and fiscalrevenue decentralization can increase the local governmentsrsquorevenue share so as to provide positive incentives [43] bothof them have a positive and promoting effect on the improve-ment of local financial efficiency and such effect largelydepends on the symmetry between the fiscal expenditure andrevenue decentralization If it is asymmetric between theirproperty rights and the expenditure responsibilities underfiscal decentralization system the local governments maymake excessive intervention to regional financial resourcesallocation tomeet their own demands for financial resourcesNext there exists nonconformance between fiscal expendi-ture decentralization and revenue decentralizationrsquos impacton financial efficiency in neighboring regions Comparedwith the fiscal expenditure decentralization the fiscal revenuedecentralization has more apparent inhibitory effect on thefinancial efficiency in neighboring regions

Our research conclusions have important policy implica-tions that illuminate the relationship between fiscal decen-tralization and financial efficiency

(1) The regional financial efficiency is closely relatedto the location and the economic development of the sur-rounding cities The regional financial efficiency is affectednot only by the financial development and innovation inneighboring regions but also by the impact of the struc-tural differences among regions In this case breaking theadministrativemonopoly of local protectionism and carryingon the ldquorichening neighborrdquo policy of financial developmentare the best choices for the local government to developthe regional economy When making policies of financialdevelopment local governments need to consider their owneconomic development environment and economic featuresin neighboring regions and strengthen communication andcoordination with the surrounding regions so as to avoidthe possible contradictions and conflicts on financial devel-opment policies among regions

(2) The spatial spillover effects of the regional financialefficiency should be divided into two kinds of effects includ-ing global geographic spillover effects and local geographicspillover effects In order to promote the increase of financialefficiency in a better way it should also be based on thehierarchy of spatial effect of and the degree of spillover effectto define the relationship between the government and themarket and central and local governments and among thelocal governments and their role in regional financial devel-opment and innovation so as to better promote financialefficiency

(3) There is a two-way interactive mechanism betweenthe fiscal system and financial development such effectsmay affect the behaviors of the local governments and theallocation of resources and ultimately act on macroeconomicperformance [10] Under the fiscal decentralization systemthe strategic competitive behaviors among local governmentssignificantly affect the liquidity of financial resources and theoccurrence of financial innovation the local governmentsrsquoexpansion of fiscal expenditure and revenue shortfall becomethe main driving force of each region competing for financialresources In order to better play the role of incentivesand constraints of fiscal competition mechanism under thefiscal decentralization system and avoid the negative impactbrought by the vicious interregional strategic financial com-petitive behaviors on the regional economic development thelocal governmentsrsquo short-sighted behaviors should be strictlyregulated and then the interregional financial cooperationwith mutual benefit as the core should be actively promoted

(4) Constructing a scientific and rational framework forfiscal decentralization will be an important force to accelerateeffectively the improvement of regional financial efficiencyChinarsquos future subprovincial fiscal decentralization reformshould adapt to the new normal financial development andfurther rationalize the provincial and subprovincial fiscaldecentralization system to ensure a balanced and appropriatefiscal expenditure and revenue decentralization On onehand we should weaken local governmentsrsquo negative inter-vention in financial system by reducing local governmentsrsquofinancial pressure caused by the asymmetry between thefiscal resources and expenditure responsibilities On the otherhandwe should also actively play the function of fiscal systemto adjust the financial development and innovation and fullyplay public financersquos role in promoting the flowing of finan-cial resources among regions the balanced development ofregional finance and the improvement of financial efficiency

Competing Interests

There are no competing interests in regard to the contentdiscussed in this paper

Authorsrsquo Contributions

All authors conceived and designed the research XiaomeiHucollected and processed the data Jianmin Liu analyzed thedata and interpreted the results all authors were also involvedin writing the paper and have all approved the submittedform

Acknowledgments

The research of this paper is jointly funded by the NationalSocial Science Foundation of China under Project no14BJY159 and the Hunan Philosophy and Social ScienceFund Project under Grant no 13JD09 The authors gratefullyacknowledge the support of Jinguang Wu from HunanUniversity of Finance and Economics China

Discrete Dynamics in Nature and Society 13

References

[1] J Zhang and X Jin ldquoRe-examine of the relationship betweenChinarsquos financial deepening and productivity 1987ndash2001rdquo Eco-nomic Research Journal vol 11 pp 35ndash43 2005

[2] K Chen A L Hillman and Q Y Gu ldquoFiscal centralization andchanges of local governmentsrsquo behaviorsmdashfrom lsquohands for helprsquoto lsquohands for grabrsquordquoChina EconomicQuarterly vol 1 pp 111ndash1302002

[3] L Zhou and Z M Wang ldquoEmpirical analysis of regionalfinancial development of China and economic growth 1978ndash2000rdquo Financial Research vol 10 pp 1ndash13 2002

[4] L Zhou ldquolsquoSecond financersquo of Chinarsquos financial industry andfinancial division during the reforming periodrdquo The Journal ofWorld Economy vol 6 pp 72ndash79 2003

[5] L Zhou ldquoGradual transition state capacity and fiscal functionof financerdquo Journal of Finance and Economics vol 2 pp 26ndash372005

[6] E L He and W Liu ldquoFinancial capital local governmentsrsquointervention and industrial structuremdashan empirical analysisbased on the data of central regionsrdquo Economic Issues vol 5pp 37ndash40 2011

[7] X L Dou and W J Wang ldquoThe differences and trends of Chi-nese regional financial development and local governmentsrsquobehavior and analysis of tax sharing reform on the differencesand trends of Chinese regional financial developmentrdquo Shang-hai Finance vol 2 pp 9ndash15 2011

[8] J Zhang and K R Shen ldquoLocal government interventionregional financial development and the transformation ofChinarsquos economic growth modemdashan empirical study under thefiscal decentralization systemrdquo Nankai Economic Studies vol 6pp 122ndash141 2008

[9] G Chen X G Yin and Y Pan ldquoChinarsquos financial developmenttax sharing reform and economic growthrdquo Financial Researchvol 2 pp 99ndash109 2006

[10] P P Ding and Y Fu ldquoBehaviors of local governments therelationship of fiscal and finance and macroeconomic volatilityassociated with Chinamdashan analysis under Chinese-style fiscaldecentralization systemrdquo Comparative Economic amp Social Sys-tems vol 6 pp 87ndash97 2012

[11] X G Yin G Chen and Y Pan ldquoTax sharing reform local gov-ernmentsrsquo intervention and financial development efficiencyrdquoJournal of Finance and Economics vol 10 pp 92ndash101 2006

[12] M G Yu and H B Pan ldquoGovernment intervention the ruleof law financial development and bank loans of state-ownedenterprisesrdquo Financial Research vol 9 pp 1ndash23 2008

[13] D X Wang J Liu and L Li ldquoFiscal decentralization bankcredit and total factor productivityrdquo Journal of Finance andEconomics vol 4 pp 69ndash80 2011

[14] Y X Jiang and R B Jiang ldquoRegional financial innovationefficiency evaluation environmental impact and gap analysisrdquoJournal of Zhejiang University (Humanities and Social Sciences)vol 43 no 4 pp 52ndash65 2013

[15] H W Yang L W Cheng Y Xu and Y W Qi ldquoMeasure andinfluencing factors of financial resources allocation efficiencyunder fiscal decentralization system based on super-efficiencyDEA-TOBIT two-step methodrdquo Journal of Jiangxi NormalUniversity (Natural Sciences Edition) vol 6 pp 587ndash592 2014

[16] N Shen F Z Liu and J Q Zhao ldquoFiscal decentralizationfinancial deepening and regional international poverty prone todeveloprdquo Finance amp Trade Economics vol 1 pp 41ndash46 2006

[17] Q Y Du G P Li and N Shen ldquoFinancial efficiency andoptimization of the regional industrial structure under thecurrent fiscal systemrdquo Xinjiang Social Science vol 4 pp 16ndash212006

[18] T Feng Y W Song and Y Lu ldquoFiscal decentralization localgovernmentsrsquo behavior and regional financial developmentrdquoJournal of Xirsquoan Jiaotong University vol 5 pp 20ndash27 2007

[19] Y A Zhou X J Ma and J Y Zhao ldquoGovernment behaviorsfinancial development and economic growthrdquo Henan SocialSciences vol 1 pp 55ndash60 2007

[20] C Hao ldquoDevelopment of financial intermediation and eco-nomic growth the Chinese experiencerdquo China EconomicReview vol 17 no 4 pp 347ndash362 2006

[21] J Zhang Fiscal Decentralization Regional Financial Devel-opment and Chinarsquos Economic Growth Performance NanjingUniversity Press Nanjing China 2011

[22] Y J Yao ldquoGovernment intervention the development of finan-cial intermediation and economic growthrdquo Journal of Financeand Economics vol 8 pp 52ndash58 2010

[23] X Wang and L Li ldquoFiscal decentralization and regional finan-cial development the theoretical and empirical analysis fromthe perspective of central governmentrdquo Public Finance Researchvol 4 pp 31ndash35 2012

[24] W L Zhang and C L Qin ldquoFinancial development fiscaldecentralization and regional economic differencesrdquo EconomicGeography vol 1 pp 583ndash588 2009

[25] Y Ma J Li and B Chen ldquoProvincial differences in Chinesefiscal decentralization financial development industrializationand economic growthrdquo EconomicTheory and Business Manage-ment vol 2 pp 5ndash19 2015

[26] Y X Jiang and R B Jiang ldquoThe spatial external effects ofregional financial innovation financial competition and finan-cial agglomerationrdquo Social Science Front vol 3 pp 79ndash84 2014

[27] K Tone ldquoA slacks-based measure of super-efficiency indata envelopment analysisrdquo European Journal of OperationalResearch vol 143 no 1 pp 32ndash41 2002

[28] Y Q Lu and D G Zhang ldquoChinarsquos regional financial efficiencyand differences of efficiency measurementrdquo Economic Geogra-phy vol 1 pp 96ndash101 2012

[29] A N Berger and D B Humphrey ldquoEfficiency of financialinstitutions international survey and directions for futureresearchrdquo European Journal of Operational Research vol 98 no2 pp 175ndash212 1997

[30] Y J Yao ldquoEmpirical analysis of financial development andurban-rural income gaprdquo Journal of Finance and Economics vol2 pp 49ndash59 2005

[31] Q Li J Song E Wang H Hu J Zhang and Y WangldquoEconomic growth and pollutant emissions in China a spatialeconometric analysisrdquo Stochastic Environmental Research andRisk Assessment vol 28 no 2 pp 429ndash442 2014

[32] J P LeSage and R K Pace Introduction to Spatial EconometricsCRCPress Taylor amp Francis Group Boca Raton Fla USA 2009

[33] Q W Guo and J X Jia ldquoFiscal decentralization organizationalstructure of local governments and the size of governmentexpenditurerdquoEconomic Research Journal vol 11 pp 59ndash87 2010httpwwwcnkicomcnArticleCJFDTotal-JJYJ201011008htm

[34] F Gong and X Lei ldquoThe measurement of Chinese-style fiscaldecentralizationrdquo Statistical Research vol 10 pp 47ndash55 2010

[35] G Fan X L Wang and H P Zhu Chinese Marketing IndexRegional Marketing Process in 2011 Economic Science PressBeijing China 2011

14 Discrete Dynamics in Nature and Society

[36] S S Rosenthal and W C Strange ldquoGeography industrialorganization and agglomerationrdquoTheReview of Economics andStatistics vol 85 no 2 pp 377ndash393 2003

[37] S Park Yoon International Banking and Financial CentersKluwer Academic Boston Mass USA 1989

[38] B Y Qiao J Y Fan and X Y Feng ldquoChinarsquos fiscal decentral-ization and compulsory primary educationrdquo Social Sciences inChina vol 6 pp 37ndash46 2005

[39] W JWang and C L Qin ldquoRegional differences in the behaviorsof local governments and growth effect of fiscal decentralizationbased on the analysis of judgment hypothesis and testingrdquoManagement World vol 1 pp 9ndash21 2008

[40] J K Brueckner ldquoStrategic interaction among governments anoverview of empirical studiesrdquo International Regional ScienceReview vol 26 no 2 pp 175ndash188 2003

[41] F Revelli ldquoOn spatial public finance empiricsrdquo International Taxand Public Finance vol 12 no 4 pp 475ndash492 2005

[42] N Shen and T Y He ldquoFinancial efficiency and regional eco-nomic growth from the perspective of fiscal decentralizationrdquoReform vol 1 pp 61ndash65 2006

[43] J M Liu X M Hu and J G Wu ldquoThe threshold effect ofsub-national fiscal decentralization revenue and expendituredecentralization on provincial industrial transformation andupgrading based on the data of 14 cities in Hunan ProvincerdquoPublic Finance Research vol 8 pp 49ndash52 2014

Submit your manuscripts athttpwwwhindawicom

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

MathematicsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Mathematical Problems in Engineering

Hindawi Publishing Corporationhttpwwwhindawicom

Differential EquationsInternational Journal of

Volume 2014

Applied MathematicsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Probability and StatisticsHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Mathematical PhysicsAdvances in

Complex AnalysisJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

OptimizationJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

CombinatoricsHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Operations ResearchAdvances in

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Function Spaces

Abstract and Applied AnalysisHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of Mathematics and Mathematical Sciences

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Algebra

Discrete Dynamics in Nature and Society

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Decision SciencesAdvances in

Discrete MathematicsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom

Volume 2014 Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Stochastic AnalysisInternational Journal of

6 Discrete Dynamics in Nature and Society

variables to measure the influence of self-reinforcing mecha-nism The specific models can be expressed as

fe119894119905

= 120588119882119894119895fe119894119905

+ 120573119883119894119905

+ 120579119882119894119895119883119894119905

+ 120583119894+ ]119905+ 120576119894119905

fe119894119905

= 120593fe119894119905-1 + 120588119882

119894119895fe119894119905

+ 120573119883119894119905

+ 120579119882119894119895119883119894119905

+ 120583119894+ ]119905

+ 120576119894119905

(5)

In (5) 120588 is the spatial lag coefficient reflecting thedirection and extent of how financial efficiency in theneighboring regions affects local financial efficiency and thecoefficient directly reflects the degree of spatial competitionin the financial efficiency 119882

119894119895is the spatial weight matrix of

119873119879 times 119873119879 119873 is the number of cross-sectional samples (281prefecture-level cities) and 119879 is time (2003ndash2012)

4 Variables Selection and Data Sources

41 Variables Definition and Measurement Methods Accord-ing to research purpose we will focus on the relationshipbetween fiscal decentralization (fiscal expenditure decentral-ization and revenue decentralization) and regional financialefficiency in the subnational citiesOther economic and socialfactors affecting regional financial efficiency include regionaleconomic development urbanization industrial structureinformation technology development fixed asset investmentand the level of internal openingTherefore wewill take theseindexes as controlled variables and introduce them into anempirical analysis model the detailed variable setting andmeasuring methods are as follows

411 The Dependent Variable The dependent variable isfinancial efficiency (fe) which is measured by the aforemen-tioned SBMmodel

412 Main Independent Variables Fiscal expenditure decen-tralization and revenue decentralization are our main inde-pendent variable When it comes to the measure method offiscal decentralization we mainly use research results of Guoand Jia [33] and Gong and Lei [34] for reference The specificmeasure formula of fiscal expenditure decentralization (sd) isas follows

sd

=

PFEcity

PFEcity + PFEprovince + PFEnation[1 minus

GDPcity

GDPnation]

(6)

In (6) PFEcity PFEprovince and PFEnation respectivelyrepresent per capita fiscal expenditure of each city theprovincial per capita fiscal expenditure and the national percapita fiscal expenditure GDPcity and GDPnation respectivelyrepresent the GDP in each city and the national GDP[1 minus GDPcityGDPnation] is the reduction factor for economicscale The index has proposed the effects of population scale

and economic scale The specific measure formula of fiscalrevenue decentralization (rd) is as follows

rd

=

PFRcity

PFRcity + PFRprovince + PFRnation[1 minus

GDPcity

GDPnation]

(7)

In (7) PFRcity PFRprovince and PFRnation respectivelyrepresent per capita fiscal revenue of each city the provincialper capita fiscal revenue and the national per capita fiscalrevenue GDPcity and GDPnation respectively represent theGDP in each city and the national GDP

413 Controlled Variable Controlled variables are as follows(1) Regional economic development (119901119892119889119901) regional

economy provides the necessary material and nonmaterialsupport for the financial development and innovation thehigher the level of regional economic development is themore beneficial to financial efficiency the improvement isIn this paper per capita actual GDP is used as a proxyvariable for regional economic development To exclude theinfluence of price factor the foregoing data that involve valueare adjusted in accordance with constant 2003 prices

(2) Urbanization (119906119903119887119886119899) Urbanization of population(the proportion of nonagricultural population accounts fortotal population) is used as the proxy variable in which thedata of nonagricultural population covering the period from2003 to 2008 is from ldquoChina Urban Statistical Yearbookrdquo(2004ndash2009) and the data of nonagricultural populationcovering the period from 2009 to 2012 is from the annualldquoStatistics on Population of Counties and Cities in Chinardquo

(3) Industrial structure (119894119905119906) the upgrading and adjust-ment of regional industrial structure constitutes a soft envi-ronment for financial development and innovation Thehigher the proportion of the third industry is the morebeneficial to financial efficiency the improvement is In thispaper the index of industrial structure upgrading is usedto represent industrial structure and the specific measureformula is as follows

itu =

119899

sum

119895=1

119902 (119895) lowast 119895 (8)

In (8) 119902(119895) is the proportion of the third industry accountsfor local GDP in region 119895 119899 = 3 and the value of index rangesfrom 1 to 3

(4) Information technology development (119894119899119905119890119903119899119890119905) it ismeasured by the number of internet users in each city

(5) Fixed asset investment (119894119891119886) it is measured by theactual fixed asset investment in each city

(6)The level of internal opening (119901119903119894V119886119905119890) with referenceto the research result of Fan et al [35] we use the developmentof private enterprises to measure the level of internal openingand assume that the higher the level of the developmentof private enterprise is the stronger the flexibility of localfinancial innovation activities is which is more conduciveto financial efficiency improvement The proportion of theemployees of private enterprises and self-employed employ-ees account for the local total population is used as a proxy to

Discrete Dynamics in Nature and Society 7

measure the development of private enterprises in differentregions

414 Spatial Weight Matrixes Strategic competition tendsto occur in the regions with the similar level of economicdevelopment simultaneously the concentration effect willgradually decay with the geographical distance [36] In orderto comprehensively reflect the spatial effects of geographicaland economic characteristics and accurately measure thespatial correlation among regions we will construct themixed geography-economy weighting matrix The specificmeasure formula is as follows

119882eco-geo =

10038161003816100381610038161003816119876119894minus 119876119895

10038161003816100381610038161003816

119889119894119895

2119894 = 119895

0 119894 = 119895

(9)

In (9) 119882eco-geo is mixed geography-economy weightingmatrix119876

119894and119876

119895 respectively represent the per capita actual

GDP of city 119894 and city 119895 (the mean of the actual GDP coveringthe period from 2003 to 2012) and 119889

119894119895is the geographical

distance between city 119894 and city 119895 calculated with the latitudeand longitude data of prefecture-level cityrsquos administrativeunit The specific measure formula is as follows

119889119894119895

= 119877 arccos (sin120593119860sin120593119861

+ cos120593119860cos120593119861cos (120582

119861minus 120582119860))

(10)

In (10) 119877 is the earth circle radius (6378 km) and 120582 and120593 respectively represent administrative centerrsquos longitudeand latitude of city 119894 and city 119895 Longitude and latitude dataare from the national fundamental geographic informationsystem All of the diagonal elements of matrix 119882eco-geo are0 And the weight matrix is normalized so that the sum ofline elements is 1 In addition considering the reliability androbustness of statistical analysis we will also construct geo-graphical distance weight matrix tomake a spatial economet-ric analysis and then compare the corresponding estimatedresultsThe specificmeasure formula of geographical distanceweight matrix is as follows

119882geo =

1

119889119894119895

2119894 = 119895

0 119894 = 119895

(11)

In (11) 119882geo is a geographical distance weighting matrixwhich is also normalized so that the sum of line elements is 1and other indexes are consistent with that of (10)

42 Data Sources and Description Considering the avail-ability of data and consistency of research samples weeventually choose 281 prefecture-level cities (Hong KongMacao and Taiwan are excluded) as samples and excludeBazhong Ziyang Lhasa Jinchang Baiyin Zhongwei andother cities due to a serious lack of available dataThe researchperiod is covering the period from 2003 to 2012 and thedata is from ldquoChina City Statistical Yearbookrdquo ldquoRegional

Economic China Statistical Yearbookrdquo and provincial statis-tical yearbooks covering the period from 2004 to 2013 Themissing data was partially filled by interpolation In orderto weaken the influence of heteroskedasticity outliers andanomaly terms on the data stability all the data are processedby the logarithm An overview of the descriptive statistics ofthese data is provided in Table 2

In order to avoid the deviation of the model settingand spurious regression we carry on unit root testing onrelated variables by LLC and Fisher-ADF testing methodsbefore spatial econometric regression analysis The resultsof stationary testing of panel data indicate that all variablesreject the null hypothesis of unit root which passes the testat the 5 significance level Moreover the panel data arestationary and the regression results will be significantlyreliable

5 Empirical Methodology

51 Spatial Econometric Analysis of Fiscal Decentralization onRegional Financial Efficiency Since the SDMmodel containsthe spatial lag item of the dependent variable and the spatiallag items of independent variables at the same time so itcanmore fully reflect the impact of spatial autocorrelation onthe regression results than SEM model and SLM model [32]Therefore we construct SDMmodels to test the spatial effectsof the fiscal revenue decentralization and fiscal expendituredecentralization on the regional financial efficiency The 119875

values of Hausman test of the spatial econometric models areless than 005 which indicates that the estimated results offixed effect estimation are better than that of random effectestimation Therefore we will specifically carry out analysisaccording to the estimated results of spatially fixed effectmodels in the following parts Meanwhile a robust test iscarried by constructing DSDM models It is found that theestimated coefficients of SDMmodel and DSDMmodel havethe same direction while the values of estimated coefficientsand level of significance are also very close indicating that theresults of SDM model are significantly reliable Comparingthe values of 119877

2 and Log-likelihood of SDM model andDSDM model we find that the fitting degree of SDM modelis higher than that of DSDMmodel so the following analysisis mainly based on the econometric results of SDMmodel

As shown in Table 3 the overall spatial correlationcoefficients 120588 of four models are positive and pass thetests at the 1 significance level indicating that there arepositive spatial dependent relationships between the localfinancial efficiency and the neighboring regionsrsquo financialefficiency that is the economic activities of geographyneighborhood regions or regions with similar level of eco-nomic development have obvious spillover effect on localfinancial efficiency thereby causing imitation and strate-gic competition of financial development and innovationamong regions On one hand the geographical adjacencyand similarity of economic development in different regionsbring huge convenience to cooperation between financialenterprises sharing of infrastructure information exchangeand communication innovation and diffusion of knowledge

8 Discrete Dynamics in Nature and Society

Table 2 Statistical description of samplersquos variables

Index Variables Mean Standarddeviation Minimum 25 quantile 50 quantile 75 quantile Maximum

Finance efficiency fe 0576 0624 0133 0380 0500 0653 3958Fiscal expendituredecentralization sd 0368 0108 0138 0294 0351 0421 0888

Fiscal revenuedecentralization rd 0242 0146 0037 0133 0202 0318 0871

Per capita real GDP(yuanperson) pgdp 27119890 + 04 32119890 + 04 1891200 9837380 17119890 + 04 31119890 + 04 44119890 + 05

Urbanization () Urban 0345 0195 0013 0202 0296 0451 100000Industrial structure itu 2203 0136 1318 2110 2190 2283 2797Informationtechnologydevelopment (tenthousand families)

Internet 41066 128506 0024 8316 17139 36800 5174000

Fixed assetinvestment (tenthousand yuan)

ifa 15119890 + 04 15119890 + 04 654299 4606420 9841960 20119890 + 04 17119890 + 05

The level of internalopening () Private 0086 0106 0003 0036 0059 0096 100000

and technology flowing of financial resources among regionswhich causes regional financial development and innovationand imitation and strategic competition among regions [37]on the other hand the restriction of transportation cost onfinancial sector is not strong and the rapid development ofinternet technology leads to cross-regional financial transac-tions and innovation thus further expanding the clusteringeffect effect of economic scale and spatial spillover effectsof financial development The coefficients of lag item (fe

119894119905-1)of financial efficiency in DSDM models are positive andthey are respectively 0211 and 0212 both of them pass thetests at the 1 significance level indicating that the previousinputs and outputs of production will be present through thefinancial efficiency increasing in later stage (or later period)and financial efficiency increasing is a dynamic systemicprocess The specific impact of each variable on financialefficiency is analyzed as follows

The estimated coefficients of Lnsd and Lnrd are positiveand all of them pass the tests at the 1 significance levelwhich indicates that the fiscal expenditure and revenuedecentralization play a significant positive role in promotinglocal financial efficiency Under the dual pressures of makingup the fiscal gap and political promotion the local govern-ments regarding economic development as main missionhave instinctive strong interest in scarce flowing economicfactors [38] As the excellent combination of capital talenttechnology management and marketing financial resourceswill be the important target for battling among local govern-ments With the expansion of fiscal expenditure decentral-ization and revenue decentralization local government usesmany policy tools such as fiscal subsidies tax concessionsand government procurement to compete for getting thefinancial resources by attracting financial capital humancapital and technological innovation to flow into the localregion thus contributing to the financial innovation and

improvement of financial efficiencyThe influence coefficientof fiscal expenditure decentralization is 0357 greater thanthe influence coefficient (0153) of fiscal revenue decentral-ization indicating that the impacts of fiscal expendituredecentralization and revenue decentralization on regionalfinancial efficiency have unbalanced characteristics and thepossible reason is that local governments under provinceget very limited taxation autonomy resulting in the factthat revenue decentralization is often lower than expendi-ture decentralization In addition under the capital-driveneconomic growth model the flowing direction of capitaldepends on the level of infrastructure [39] thus the inflowof financial resources as well as financial innovation oftenoccurs in the regions with rapid economic developmentexcellent investment environment and great consumptionability so that local governments have to increase fiscalexpenditure to enhance the supply of the associated support-ing infrastructure Comparatively speaking tax preferenceof undeveloped regions is less attractive than the superiorityconditions of developed regions

We will further analyze the estimated coefficients andsignificance levels of spatial lags of fiscal expenditure andrevenue decentralization and investigate the strategic com-petitive effects of financial efficiency among neighboringregions The spatial lag of expenditure decentralization(WLnsd) does not pass the significance test suggestingthat spatial spillover effect of fiscal expenditure decentral-ization is uncertain and needs further examination whilethe spatial lag of fiscal revenue decentralization (WLnrd)is negative at the 5 significance level and the estimatedcoefficient is minus0228 indicating that with the increase offiscal revenue decentralization the negative externalities onfinancial development and innovation will be increasinglyprominent thus inhibiting the improvement of financialefficiency in neighboring regions The reason may be that

Discrete Dynamics in Nature and Society 9

Table 3 Estimated results of regional financial efficiency under theSpatial Durbin Model (economic-geographic weighting matrix)

Independentvariables

Dependent variable sd Dependent variable rd

SDMmodel DSDMmodel SDMmodel DSDM

model

feitminus10211lowastlowastlowast(1283)

0212lowastlowastlowast(1283)

Lnsd 0357lowastlowastlowast(626)

0334lowastlowastlowast(570)

Lnrd 0153lowastlowastlowast(470)

0143lowastlowastlowast(408)

Lnpgdp 0169lowastlowastlowast(462)

0120lowastlowastlowast(308)

0138lowastlowastlowast(360)

0099lowastlowast(244)

Lnurban minus0050lowastlowast(minus237)

minus0044lowastlowast(minus209)

minus0050lowastlowast(minus237)

minus0045lowastlowast(minus215)

Lnitu 0433lowastlowast(227)

0438lowastlowast(215)

0525lowastlowastlowast(273)

0531lowastlowastlowast(259)

Lninternet minus0295lowastlowastlowast(minus2970)

minus0299lowastlowastlowast(minus2716)

minus0292lowastlowastlowast(minus2945)

minus0297lowastlowastlowast(minus2690)

Lnifa minus0446lowastlowastlowast(minus2298)

minus0386lowastlowastlowast(minus1832)

minus0437lowastlowastlowast(minus2321)

minus0374lowastlowastlowast(minus1806)

Lnprivate 0038lowastlowastlowast(322)

0037lowastlowastlowast(290)

0039lowastlowastlowast(325)

0038lowastlowastlowast(302)

WLnsd 0179(099)

minus0041(minus022)

WLnrd minus0228lowastlowast(minus251)

minus0186lowast(minus179)

WLnpgdp 0016(027)

0064(101)

0058(096)

0097(150)

WLnurban 0097lowastlowast(236)

0082lowastlowast(198)

0107lowastlowastlowast(259)

0089lowastlowast(213)

WLnitu 1159lowast(193)

0516(082)

0615(100)

0228(035)

WLninternet 0177lowastlowastlowast(647)

0181lowastlowastlowast(622)

0180lowastlowastlowast(666)

0180lowastlowastlowast(619)

WLnifa 0237lowastlowastlowast(539)

0251lowastlowastlowast(532)

0275lowastlowastlowast(704)

0260lowastlowastlowast(600)

WLnprivate 0065lowast(185)

0045(115)

0083lowastlowast(232)

0060(153)

120588

0420lowastlowastlowast(1165)

0462lowastlowastlowast(1251)

0429lowastlowastlowast(1205)

0463lowastlowastlowast(1266)

1198772 0448 0443 0441 0438

log 119871 599812 595228 595228 587624Observations 2810 2529 2810 2529Note the values in parentheses represent the 119905-statistics values of thecorresponding estimated coefficientslowastlowastlowastlowastlowast andlowast respectively representsignificance level of 1 5 and 10

local governmentsrsquo motivation of pursuing revenues underthe fiscal revenue decentralization system is not conduciveto free flowing of financial resources among regions whichdirectly affects the financial efficiency in other regions

When it comes to the estimated coefficients of othercontrolled variables we found that not all the spatial spillovereffects of variables are significant Overall various economicactivities have direct and indirect influence on financial

efficiency and promote the strategic financial competitionamong regions The specific effects are as follows the levelof open-into-inside can not only improve the local financialefficiency significantly but also increase the financial effi-ciency correspondingly in neighboring regions The strongdemand for various financial products and services of privateenterprises promotes financial innovation and improves theefficiency of resource allocation in financial markets Onthe contrary urbanization information technology develop-ment and fixed assets investment suppress the improvementof local financial efficiency but play a positive role in promot-ing the financial efficiency in neighboring regionsThe accel-erating of urbanization process the application of internettechnology and investment in fixed assets promote economicexchanges and cooperation among regions so that the spatialspillover effects of financial development and innovation areincreasingly significant Furthermore although the regionaleconomic development and industrial structure significantlycontribute to the improvement of regional financial efficiencyin general the spillover effects among regions are not obvious

52 Decomposition of Fiscal Decentralizationrsquos Effects onRegional Financial Efficiency Whether there truly existspillover effects among the variables relying solely on the spa-tial lag coefficients of SDMmodel and DSDMmodel is likelyto lead to misinterpretation of the estimated results Spatialeffects can be further decomposed into direct effect indirecteffect (spillover effects) and total effect with reference to theresearch result of LeSage and Pace [32] The indirect effectsrepresent the independent variablesrsquo influence on the depen-dent variable in all other regions through spatial interactionsTherefore we will further verify the existence of spillovereffects by using partial differential matrix analysis Table 4illustrates the direct effect indirect effect and total effect ofthe fiscal expenditure and revenue decentralization in SDMmodel The results show that the estimated coefficients ofindependent variablesrsquo direct effect have the same directionwith the estimated coefficients of SDMmodel in Table 3 andthe estimated coefficients and significance levels are also veryclose

According to the estimated results of SDM model inTable 4 the direct effect indirect effect and total effectof fiscal expenditure decentralization on regional financialefficiency are positive and pass the 10 significance leveltest The regression coefficient of direct effect is 0368 theregression coefficient of indirect effect is 0560 and theregression coefficient of total effect is 0928 indicating thatthe degree of fiscal expenditure decentralization increases by1 while regional financial efficiency increases by 0928Fiscal expenditure decentralization plays a significant posi-tive impact on financial development and financial efficiencywhich is opposite to the research result of Wang and Li [23]The opposite conclusions are possibly due to the differencesof sample selection and indexesThe rise of fiscal expendituredecentralization in the local region leads to an increaseby 0368 of the financial efficiency while the increase ofexpenditure decentralization in the neighboring region leadsto financial efficiency increasing by 0560 indicating that

10 Discrete Dynamics in Nature and Society

Table 4 Decomposition of direct effect indirect effect and total effect under SDMmodel

Independent variablesDependent variable sd Dependent variable sd

Direct effect Indirect effect Total effect Direct effect Indirect effect Total effect

Lnsd 0368lowastlowastlowast(750)

0560lowast(195)

0928lowastlowastlowast(310)

Lnrd 0146lowastlowastlowast(529)

minus0283lowast(minus195)

minus0137lowast(minus192)

Lnpgdp 0175lowastlowastlowast(435)

0132(153)

0307lowastlowastlowast(359)

0145lowastlowastlowast(346)

0187lowastlowast(218)

0332lowastlowastlowast(392)

Lnurban minus0046lowastlowast(minus207)

0127lowastlowast(200)

0081(130)

minus0045lowastlowast(minus204)

0146lowastlowast(226)

0101(158)

Lnitu 0486lowastlowastlowast(275)

2321lowastlowast(237)

2807lowastlowastlowast(292)

0559lowastlowastlowast(315)

1488(151)

2047lowastlowast(211)

Lninternet minus0291lowastlowastlowast(minus3055)

0086lowast(190)

minus0205lowastlowastlowast(minus432)

minus0288lowastlowastlowast(minus3029)

0091lowastlowast(200)

minus0197lowastlowastlowast(minus410)

Lnifa minus0441lowastlowastlowast(minus2260)

0087(149)

minus0353lowastlowastlowast(minus594)

minus0430lowastlowastlowast(minus2254)

0154lowastlowastlowast(283)

minus0276lowastlowastlowast(minus494)

Lnprivate 0041lowastlowastlowast(319)

0134lowastlowast(235)

0175lowastlowastlowast(285)

0042lowastlowastlowast(327)

0166lowastlowastlowast(283)

0208lowastlowastlowast(330)

Note lowast lowast lowast lowastlowast and lowast respectively represent the significant level of 1 5 and 10

fiscal expenditure decentralization significantly promotesthe competition of financial development and innovationamong regions The expansion of the autonomy of localgovernment expenditure strengthens the regulatory capacityof local governments which is conducive to acceleratingmarket-oriented reform processing and achieves the organicintegration between fiscal system and market mechanism inthe allocation of financial resources which is beneficial tocreating a favorable institutional environment is also con-ducive to the flowing of financial resources among regionsand greatly promotes financial efficiency

The fiscal revenue decentralizationrsquos direct effect onregional financial efficiency is significantly positive whilethe spillover effect is significantly negative indicating thatthe impact of fiscal revenue decentralization on regionalfinancial efficiency presents the feature of ldquodouble-edged-swordrdquo Fiscal revenue decentralization has a significant posi-tive impact on the local financial efficiencyHowever there aresignificant negative spillover effects on the financial efficiencyin other regions The regression coefficient of the directeffect is 0146 the regression coefficient of the indirect effectis minus0283 and the regression coefficient of the total effectis minus0137 which indicates that the degree of fiscal revenuedecentralization increasing by 1 will lead to decrease by0137 of regional financial efficiency Furthermore whilethe increase of local fiscal revenue decentralization leads tofinancial efficiencyrsquos increase by 0146 the increase of thefiscal revenue decentralization in neighboring regions resultsin the financial efficiencyrsquos decrease by 0283 Comparingthe estimated coefficient of direct effect and that of indirecteffect it can be found that the negative spillover effectsof fiscal revenue decentralization are significantly greaterthan the direct scale effect so the negative externalitiesof fiscal revenue decentralization on financial efficiencyare greater than the positive externality On one hand inorder to obtain high-quality financial resources the local

governments ignore the differences like the level of economicdevelopment factor endowments and other aspects amongregions and seek to imitate policies and instruments inother regions thus leading to chain reaction of absorbinginvestment by other regions On the other hand whenthe neighboring regions or the regions with similar levelof economic development take some measures to attractliquidity factors the local government driven by yardstickcompetitionwill adopt similar policies to avoid the outflow ofelements which results in convergence of competitive tactics[40 41] Tax incentives reduction of fee and other wayswill become important means of competition and imitationbetween local governments in different regions on the basisof the independent revenue authority However such kind ofcompetitive mode between local governments for resourceswill consume huge fiscal resources and weaken the capacityof local governmentrsquos provision of public goods and servicesUltimately it is harmful to provide healthy stable andsustainable economic development environment for financialinnovation and increase of financial efficiency

53 Robustness Test To ensure the robustness and reliabilityof research results we will calculate the spatial lags withthe geographical distance weight matrix and do econometricregression by taking financial efficiency as the dependentvariable so that we can comparatively analyze the differenceof influence with different spatial weight setting based on theabove The robustness test results are as in Table 5

As shown in Table 5 the estimated results indicate thatthe sign of regression coefficients of the fiscal expendituredecentralization and revenue decentralization is consistentwith that of corresponding variable in the model of Table 3and the regression results in themodel are basically consistentwith the conclusions in Table 3 Comparing the estimatedresults based on two different weightmatrixes it is found thatthere are some differences in the influence coefficients of two

Discrete Dynamics in Nature and Society 11

Table 5 Robustness test on Spatial Durbin Models (geographicaldistance weight matrix)

Independentvariables

Dependent variable sd Dependent variable rdSDMmodel

DSDMmodel

SDMmodel

DSDMmodel

feitminus1 mdash 0202lowastlowastlowast(1276) mdash 0200lowastlowastlowast

(1255)

Lnsd or Lnrd 0364lowastlowastlowast(636)

0348lowastlowastlowast(592)

0173lowastlowastlowast(522)

0161lowastlowastlowast(458)

Lnpgdp 0268lowastlowastlowast(735)

0225lowastlowastlowast(578)

0242lowastlowastlowast(637)

0209lowastlowastlowast(522)

Lnurban minus0026(minus112)

minus0026(minus112)

minus0028(minus120)

minus0027(minus115)

Lnitu 0465lowastlowast(246)

0478lowastlowast(238)

0527lowastlowastlowast(278)

0542lowastlowastlowast(269)

Lninternet minus0295lowastlowastlowast(minus3038)

minus0300lowastlowastlowast(minus2799)

minus0292lowastlowastlowast(minus3005)

minus0298lowastlowastlowast(minus2771)

Lnifa minus0484lowastlowastlowast(minus2479)

minus0427lowastlowastlowast(minus2024)

minus0472lowastlowastlowast(minus2448)

minus0417lowastlowastlowast(minus1988)

Lnprivate 0036lowastlowastlowast(312)

0035lowastlowastlowast(284)

0035lowastlowastlowast(307)

0035lowastlowastlowast(288)

WLnsd orWLnrd

minus0264(minus149)

minus0418lowastlowast(minus225)

minus0277lowastlowastlowast(minus370)

minus0241lowastlowastlowast(minus276)

WLnpgdp minus0269lowastlowastlowast(minus427)

minus0286lowastlowastlowast(minus428)

minus0237lowastlowastlowast(minus369)

minus0267lowastlowastlowast(minus394)

WLnurban 0053(147)

0059lowast(167)

0059(163)

0062lowast(173)

WLnitu 0104(020)

minus0085(minus016)

minus0205(minus039)

minus0290(minus051)

WLninternet 0217lowastlowastlowast(838)

0239lowastlowastlowast(853)

0214lowastlowastlowast(831)

0235lowastlowastlowast(841)

WLnifa 0434lowastlowastlowast(1052)

0464lowastlowastlowast(1054)

0433lowastlowastlowast(1165)

0450lowastlowastlowast(1129)

WLnprivate 0062lowast(187)

0038(106)

0081lowastlowast(241)

0050(139)

120588

0547lowastlowastlowast(1858)

0580lowastlowastlowast(1938)

0541lowastlowastlowast(1839)

0570lowastlowastlowast(1898)

1198772 0444 0428 0443 0430

log 119871 659741 665804 665804 659509Observations 2810 2529 2810 2529Note lowastlowastlowast lowastlowast and lowast respectively represent the significant level of 1 5and 10

kinds of fiscal decentralizations indicators which show thatthe gap between the levels of economic developmentwill havean impact on role of fiscal decentralization acting on regionalfinancial efficiency On one hand under the fiscal decen-tralization system because of the heterogeneity of regionaleconomic development level the local governmentsrsquo ability ofmaking up the gap between fiscal revenue and expenditure isdifferent and the fiscal pressure is also different therefore thedegree of interference to the regional market environment bythe local government is different which leads to the regionaldifferences in financial efficiency in a certain degree [21]On the other hand the occurrence of financial innovationdepends on the support of the economic foundation and

the different economic and social environment will lead tothe fact that the financial innovation activities are more andmore concentrated in the regions with the advantages offinancial location resulting in nonequilibrium of financialefficiency in different regions [42] Under the performanceassessment method focusing on GDP local governments willnot only consider the strategic behaviors of the neighboringregions but also tend to think about the behaviors of theregions with the similar level of economy In additioncross-regional knowledge diffusion technology spillover andhuman capital flow are more prone to occur among theregions with similar level of economic development so thatthe financial exchanges and contacts among regions are morefrequent It also shows that besides considering the impact ofgeographical distance the economic distance factor shouldbe taken into consideration while we are researching thespatial effects of fiscal decentralization on regional financialinnovation in addition to the consideration of the impact ofgeographical distance the economic distance factor shouldbe considered

6 Conclusions and Enlightenments

Based on the panel data of 281 prefecture-level cities in Chinacovering the period of 2003ndash2012 the regional financialefficiency is measured by the superefficiency SBMmodel thegeographic distance weight matrix and economic-geographymixed weight matrix which can reflect the regional similarityare introduced into the SDM and the spatial effects of thefiscal expenditure and revenue decentralization on regionalfinancial efficiency under province are investigated as wellas the strategic financial competition arising from it Basedon the results of previous empirical analysis we can draw thefollowing basic conclusions

First the geographical distribution of Chinarsquos prefecture-level citiesrsquo financial efficiency presents significant features oftime inertia and spatial dependency Financial developmentand innovation among regions particularly neighboringregions is not independent but shows significant spatialspillover effects The financial efficiency of the local regionwill be significantly affected by the financial efficiency ofadjacent regions The spatial effect is increasingly obviousin the process of financial development and innovation andthe size of the spillover effects is closely related to spatialweight (which is used to measure regional similarities) andmainly shows that there is competition for financial efficiencyamong regions with similar economic attributes and adjacentregions

Second it cannot be ignored that the institutional factorsespecially fiscal decentralization have effect on the financialefficiency and the factors have a decisive role in the allocationof financial resource and the regional distribution of financialinnovation The improvement of fiscal decentralization levelwill help to attract more financial resources thereby promot-ing an increase in financial efficiencyThe improvement of thefinancial autonomy in neighboring jurisdictions will reducethe financial resources flowing into the local region andsuppresses the occurrence of the local financial innovationIn addition there exist significant spatial spillover effects of

12 Discrete Dynamics in Nature and Society

financial development and efficiency caused by fiscal decen-tralization The scope of spatial spillovers is mainly reflectedin two aspects Thus the local governmentsrsquo competition notonly strongly increases the financial efficiency of the regionbut also accelerates interregional competition for financialefficiency

Third fiscal expenditure decentralization and revenuedecentralizationrsquos impact on regional financial efficiencyhave nonequilibrium characteristics First of all there existsasymmetry between fiscal expenditure decentralization andrevenue decentralizationrsquos impact on regional financial effi-ciency Fiscal expenditure decentralization can promoteoptimization of the fiscal expenditure structure and fiscalrevenue decentralization can increase the local governmentsrsquorevenue share so as to provide positive incentives [43] bothof them have a positive and promoting effect on the improve-ment of local financial efficiency and such effect largelydepends on the symmetry between the fiscal expenditure andrevenue decentralization If it is asymmetric between theirproperty rights and the expenditure responsibilities underfiscal decentralization system the local governments maymake excessive intervention to regional financial resourcesallocation tomeet their own demands for financial resourcesNext there exists nonconformance between fiscal expendi-ture decentralization and revenue decentralizationrsquos impacton financial efficiency in neighboring regions Comparedwith the fiscal expenditure decentralization the fiscal revenuedecentralization has more apparent inhibitory effect on thefinancial efficiency in neighboring regions

Our research conclusions have important policy implica-tions that illuminate the relationship between fiscal decen-tralization and financial efficiency

(1) The regional financial efficiency is closely relatedto the location and the economic development of the sur-rounding cities The regional financial efficiency is affectednot only by the financial development and innovation inneighboring regions but also by the impact of the struc-tural differences among regions In this case breaking theadministrativemonopoly of local protectionism and carryingon the ldquorichening neighborrdquo policy of financial developmentare the best choices for the local government to developthe regional economy When making policies of financialdevelopment local governments need to consider their owneconomic development environment and economic featuresin neighboring regions and strengthen communication andcoordination with the surrounding regions so as to avoidthe possible contradictions and conflicts on financial devel-opment policies among regions

(2) The spatial spillover effects of the regional financialefficiency should be divided into two kinds of effects includ-ing global geographic spillover effects and local geographicspillover effects In order to promote the increase of financialefficiency in a better way it should also be based on thehierarchy of spatial effect of and the degree of spillover effectto define the relationship between the government and themarket and central and local governments and among thelocal governments and their role in regional financial devel-opment and innovation so as to better promote financialefficiency

(3) There is a two-way interactive mechanism betweenthe fiscal system and financial development such effectsmay affect the behaviors of the local governments and theallocation of resources and ultimately act on macroeconomicperformance [10] Under the fiscal decentralization systemthe strategic competitive behaviors among local governmentssignificantly affect the liquidity of financial resources and theoccurrence of financial innovation the local governmentsrsquoexpansion of fiscal expenditure and revenue shortfall becomethe main driving force of each region competing for financialresources In order to better play the role of incentivesand constraints of fiscal competition mechanism under thefiscal decentralization system and avoid the negative impactbrought by the vicious interregional strategic financial com-petitive behaviors on the regional economic development thelocal governmentsrsquo short-sighted behaviors should be strictlyregulated and then the interregional financial cooperationwith mutual benefit as the core should be actively promoted

(4) Constructing a scientific and rational framework forfiscal decentralization will be an important force to accelerateeffectively the improvement of regional financial efficiencyChinarsquos future subprovincial fiscal decentralization reformshould adapt to the new normal financial development andfurther rationalize the provincial and subprovincial fiscaldecentralization system to ensure a balanced and appropriatefiscal expenditure and revenue decentralization On onehand we should weaken local governmentsrsquo negative inter-vention in financial system by reducing local governmentsrsquofinancial pressure caused by the asymmetry between thefiscal resources and expenditure responsibilities On the otherhandwe should also actively play the function of fiscal systemto adjust the financial development and innovation and fullyplay public financersquos role in promoting the flowing of finan-cial resources among regions the balanced development ofregional finance and the improvement of financial efficiency

Competing Interests

There are no competing interests in regard to the contentdiscussed in this paper

Authorsrsquo Contributions

All authors conceived and designed the research XiaomeiHucollected and processed the data Jianmin Liu analyzed thedata and interpreted the results all authors were also involvedin writing the paper and have all approved the submittedform

Acknowledgments

The research of this paper is jointly funded by the NationalSocial Science Foundation of China under Project no14BJY159 and the Hunan Philosophy and Social ScienceFund Project under Grant no 13JD09 The authors gratefullyacknowledge the support of Jinguang Wu from HunanUniversity of Finance and Economics China

Discrete Dynamics in Nature and Society 13

References

[1] J Zhang and X Jin ldquoRe-examine of the relationship betweenChinarsquos financial deepening and productivity 1987ndash2001rdquo Eco-nomic Research Journal vol 11 pp 35ndash43 2005

[2] K Chen A L Hillman and Q Y Gu ldquoFiscal centralization andchanges of local governmentsrsquo behaviorsmdashfrom lsquohands for helprsquoto lsquohands for grabrsquordquoChina EconomicQuarterly vol 1 pp 111ndash1302002

[3] L Zhou and Z M Wang ldquoEmpirical analysis of regionalfinancial development of China and economic growth 1978ndash2000rdquo Financial Research vol 10 pp 1ndash13 2002

[4] L Zhou ldquolsquoSecond financersquo of Chinarsquos financial industry andfinancial division during the reforming periodrdquo The Journal ofWorld Economy vol 6 pp 72ndash79 2003

[5] L Zhou ldquoGradual transition state capacity and fiscal functionof financerdquo Journal of Finance and Economics vol 2 pp 26ndash372005

[6] E L He and W Liu ldquoFinancial capital local governmentsrsquointervention and industrial structuremdashan empirical analysisbased on the data of central regionsrdquo Economic Issues vol 5pp 37ndash40 2011

[7] X L Dou and W J Wang ldquoThe differences and trends of Chi-nese regional financial development and local governmentsrsquobehavior and analysis of tax sharing reform on the differencesand trends of Chinese regional financial developmentrdquo Shang-hai Finance vol 2 pp 9ndash15 2011

[8] J Zhang and K R Shen ldquoLocal government interventionregional financial development and the transformation ofChinarsquos economic growth modemdashan empirical study under thefiscal decentralization systemrdquo Nankai Economic Studies vol 6pp 122ndash141 2008

[9] G Chen X G Yin and Y Pan ldquoChinarsquos financial developmenttax sharing reform and economic growthrdquo Financial Researchvol 2 pp 99ndash109 2006

[10] P P Ding and Y Fu ldquoBehaviors of local governments therelationship of fiscal and finance and macroeconomic volatilityassociated with Chinamdashan analysis under Chinese-style fiscaldecentralization systemrdquo Comparative Economic amp Social Sys-tems vol 6 pp 87ndash97 2012

[11] X G Yin G Chen and Y Pan ldquoTax sharing reform local gov-ernmentsrsquo intervention and financial development efficiencyrdquoJournal of Finance and Economics vol 10 pp 92ndash101 2006

[12] M G Yu and H B Pan ldquoGovernment intervention the ruleof law financial development and bank loans of state-ownedenterprisesrdquo Financial Research vol 9 pp 1ndash23 2008

[13] D X Wang J Liu and L Li ldquoFiscal decentralization bankcredit and total factor productivityrdquo Journal of Finance andEconomics vol 4 pp 69ndash80 2011

[14] Y X Jiang and R B Jiang ldquoRegional financial innovationefficiency evaluation environmental impact and gap analysisrdquoJournal of Zhejiang University (Humanities and Social Sciences)vol 43 no 4 pp 52ndash65 2013

[15] H W Yang L W Cheng Y Xu and Y W Qi ldquoMeasure andinfluencing factors of financial resources allocation efficiencyunder fiscal decentralization system based on super-efficiencyDEA-TOBIT two-step methodrdquo Journal of Jiangxi NormalUniversity (Natural Sciences Edition) vol 6 pp 587ndash592 2014

[16] N Shen F Z Liu and J Q Zhao ldquoFiscal decentralizationfinancial deepening and regional international poverty prone todeveloprdquo Finance amp Trade Economics vol 1 pp 41ndash46 2006

[17] Q Y Du G P Li and N Shen ldquoFinancial efficiency andoptimization of the regional industrial structure under thecurrent fiscal systemrdquo Xinjiang Social Science vol 4 pp 16ndash212006

[18] T Feng Y W Song and Y Lu ldquoFiscal decentralization localgovernmentsrsquo behavior and regional financial developmentrdquoJournal of Xirsquoan Jiaotong University vol 5 pp 20ndash27 2007

[19] Y A Zhou X J Ma and J Y Zhao ldquoGovernment behaviorsfinancial development and economic growthrdquo Henan SocialSciences vol 1 pp 55ndash60 2007

[20] C Hao ldquoDevelopment of financial intermediation and eco-nomic growth the Chinese experiencerdquo China EconomicReview vol 17 no 4 pp 347ndash362 2006

[21] J Zhang Fiscal Decentralization Regional Financial Devel-opment and Chinarsquos Economic Growth Performance NanjingUniversity Press Nanjing China 2011

[22] Y J Yao ldquoGovernment intervention the development of finan-cial intermediation and economic growthrdquo Journal of Financeand Economics vol 8 pp 52ndash58 2010

[23] X Wang and L Li ldquoFiscal decentralization and regional finan-cial development the theoretical and empirical analysis fromthe perspective of central governmentrdquo Public Finance Researchvol 4 pp 31ndash35 2012

[24] W L Zhang and C L Qin ldquoFinancial development fiscaldecentralization and regional economic differencesrdquo EconomicGeography vol 1 pp 583ndash588 2009

[25] Y Ma J Li and B Chen ldquoProvincial differences in Chinesefiscal decentralization financial development industrializationand economic growthrdquo EconomicTheory and Business Manage-ment vol 2 pp 5ndash19 2015

[26] Y X Jiang and R B Jiang ldquoThe spatial external effects ofregional financial innovation financial competition and finan-cial agglomerationrdquo Social Science Front vol 3 pp 79ndash84 2014

[27] K Tone ldquoA slacks-based measure of super-efficiency indata envelopment analysisrdquo European Journal of OperationalResearch vol 143 no 1 pp 32ndash41 2002

[28] Y Q Lu and D G Zhang ldquoChinarsquos regional financial efficiencyand differences of efficiency measurementrdquo Economic Geogra-phy vol 1 pp 96ndash101 2012

[29] A N Berger and D B Humphrey ldquoEfficiency of financialinstitutions international survey and directions for futureresearchrdquo European Journal of Operational Research vol 98 no2 pp 175ndash212 1997

[30] Y J Yao ldquoEmpirical analysis of financial development andurban-rural income gaprdquo Journal of Finance and Economics vol2 pp 49ndash59 2005

[31] Q Li J Song E Wang H Hu J Zhang and Y WangldquoEconomic growth and pollutant emissions in China a spatialeconometric analysisrdquo Stochastic Environmental Research andRisk Assessment vol 28 no 2 pp 429ndash442 2014

[32] J P LeSage and R K Pace Introduction to Spatial EconometricsCRCPress Taylor amp Francis Group Boca Raton Fla USA 2009

[33] Q W Guo and J X Jia ldquoFiscal decentralization organizationalstructure of local governments and the size of governmentexpenditurerdquoEconomic Research Journal vol 11 pp 59ndash87 2010httpwwwcnkicomcnArticleCJFDTotal-JJYJ201011008htm

[34] F Gong and X Lei ldquoThe measurement of Chinese-style fiscaldecentralizationrdquo Statistical Research vol 10 pp 47ndash55 2010

[35] G Fan X L Wang and H P Zhu Chinese Marketing IndexRegional Marketing Process in 2011 Economic Science PressBeijing China 2011

14 Discrete Dynamics in Nature and Society

[36] S S Rosenthal and W C Strange ldquoGeography industrialorganization and agglomerationrdquoTheReview of Economics andStatistics vol 85 no 2 pp 377ndash393 2003

[37] S Park Yoon International Banking and Financial CentersKluwer Academic Boston Mass USA 1989

[38] B Y Qiao J Y Fan and X Y Feng ldquoChinarsquos fiscal decentral-ization and compulsory primary educationrdquo Social Sciences inChina vol 6 pp 37ndash46 2005

[39] W JWang and C L Qin ldquoRegional differences in the behaviorsof local governments and growth effect of fiscal decentralizationbased on the analysis of judgment hypothesis and testingrdquoManagement World vol 1 pp 9ndash21 2008

[40] J K Brueckner ldquoStrategic interaction among governments anoverview of empirical studiesrdquo International Regional ScienceReview vol 26 no 2 pp 175ndash188 2003

[41] F Revelli ldquoOn spatial public finance empiricsrdquo International Taxand Public Finance vol 12 no 4 pp 475ndash492 2005

[42] N Shen and T Y He ldquoFinancial efficiency and regional eco-nomic growth from the perspective of fiscal decentralizationrdquoReform vol 1 pp 61ndash65 2006

[43] J M Liu X M Hu and J G Wu ldquoThe threshold effect ofsub-national fiscal decentralization revenue and expendituredecentralization on provincial industrial transformation andupgrading based on the data of 14 cities in Hunan ProvincerdquoPublic Finance Research vol 8 pp 49ndash52 2014

Submit your manuscripts athttpwwwhindawicom

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

MathematicsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Mathematical Problems in Engineering

Hindawi Publishing Corporationhttpwwwhindawicom

Differential EquationsInternational Journal of

Volume 2014

Applied MathematicsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Probability and StatisticsHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Mathematical PhysicsAdvances in

Complex AnalysisJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

OptimizationJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

CombinatoricsHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Operations ResearchAdvances in

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Function Spaces

Abstract and Applied AnalysisHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of Mathematics and Mathematical Sciences

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Algebra

Discrete Dynamics in Nature and Society

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Decision SciencesAdvances in

Discrete MathematicsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom

Volume 2014 Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Stochastic AnalysisInternational Journal of

Discrete Dynamics in Nature and Society 7

measure the development of private enterprises in differentregions

414 Spatial Weight Matrixes Strategic competition tendsto occur in the regions with the similar level of economicdevelopment simultaneously the concentration effect willgradually decay with the geographical distance [36] In orderto comprehensively reflect the spatial effects of geographicaland economic characteristics and accurately measure thespatial correlation among regions we will construct themixed geography-economy weighting matrix The specificmeasure formula is as follows

119882eco-geo =

10038161003816100381610038161003816119876119894minus 119876119895

10038161003816100381610038161003816

119889119894119895

2119894 = 119895

0 119894 = 119895

(9)

In (9) 119882eco-geo is mixed geography-economy weightingmatrix119876

119894and119876

119895 respectively represent the per capita actual

GDP of city 119894 and city 119895 (the mean of the actual GDP coveringthe period from 2003 to 2012) and 119889

119894119895is the geographical

distance between city 119894 and city 119895 calculated with the latitudeand longitude data of prefecture-level cityrsquos administrativeunit The specific measure formula is as follows

119889119894119895

= 119877 arccos (sin120593119860sin120593119861

+ cos120593119860cos120593119861cos (120582

119861minus 120582119860))

(10)

In (10) 119877 is the earth circle radius (6378 km) and 120582 and120593 respectively represent administrative centerrsquos longitudeand latitude of city 119894 and city 119895 Longitude and latitude dataare from the national fundamental geographic informationsystem All of the diagonal elements of matrix 119882eco-geo are0 And the weight matrix is normalized so that the sum ofline elements is 1 In addition considering the reliability androbustness of statistical analysis we will also construct geo-graphical distance weight matrix tomake a spatial economet-ric analysis and then compare the corresponding estimatedresultsThe specificmeasure formula of geographical distanceweight matrix is as follows

119882geo =

1

119889119894119895

2119894 = 119895

0 119894 = 119895

(11)

In (11) 119882geo is a geographical distance weighting matrixwhich is also normalized so that the sum of line elements is 1and other indexes are consistent with that of (10)

42 Data Sources and Description Considering the avail-ability of data and consistency of research samples weeventually choose 281 prefecture-level cities (Hong KongMacao and Taiwan are excluded) as samples and excludeBazhong Ziyang Lhasa Jinchang Baiyin Zhongwei andother cities due to a serious lack of available dataThe researchperiod is covering the period from 2003 to 2012 and thedata is from ldquoChina City Statistical Yearbookrdquo ldquoRegional

Economic China Statistical Yearbookrdquo and provincial statis-tical yearbooks covering the period from 2004 to 2013 Themissing data was partially filled by interpolation In orderto weaken the influence of heteroskedasticity outliers andanomaly terms on the data stability all the data are processedby the logarithm An overview of the descriptive statistics ofthese data is provided in Table 2

In order to avoid the deviation of the model settingand spurious regression we carry on unit root testing onrelated variables by LLC and Fisher-ADF testing methodsbefore spatial econometric regression analysis The resultsof stationary testing of panel data indicate that all variablesreject the null hypothesis of unit root which passes the testat the 5 significance level Moreover the panel data arestationary and the regression results will be significantlyreliable

5 Empirical Methodology

51 Spatial Econometric Analysis of Fiscal Decentralization onRegional Financial Efficiency Since the SDMmodel containsthe spatial lag item of the dependent variable and the spatiallag items of independent variables at the same time so itcanmore fully reflect the impact of spatial autocorrelation onthe regression results than SEM model and SLM model [32]Therefore we construct SDMmodels to test the spatial effectsof the fiscal revenue decentralization and fiscal expendituredecentralization on the regional financial efficiency The 119875

values of Hausman test of the spatial econometric models areless than 005 which indicates that the estimated results offixed effect estimation are better than that of random effectestimation Therefore we will specifically carry out analysisaccording to the estimated results of spatially fixed effectmodels in the following parts Meanwhile a robust test iscarried by constructing DSDM models It is found that theestimated coefficients of SDMmodel and DSDMmodel havethe same direction while the values of estimated coefficientsand level of significance are also very close indicating that theresults of SDM model are significantly reliable Comparingthe values of 119877

2 and Log-likelihood of SDM model andDSDM model we find that the fitting degree of SDM modelis higher than that of DSDMmodel so the following analysisis mainly based on the econometric results of SDMmodel

As shown in Table 3 the overall spatial correlationcoefficients 120588 of four models are positive and pass thetests at the 1 significance level indicating that there arepositive spatial dependent relationships between the localfinancial efficiency and the neighboring regionsrsquo financialefficiency that is the economic activities of geographyneighborhood regions or regions with similar level of eco-nomic development have obvious spillover effect on localfinancial efficiency thereby causing imitation and strate-gic competition of financial development and innovationamong regions On one hand the geographical adjacencyand similarity of economic development in different regionsbring huge convenience to cooperation between financialenterprises sharing of infrastructure information exchangeand communication innovation and diffusion of knowledge

8 Discrete Dynamics in Nature and Society

Table 2 Statistical description of samplersquos variables

Index Variables Mean Standarddeviation Minimum 25 quantile 50 quantile 75 quantile Maximum

Finance efficiency fe 0576 0624 0133 0380 0500 0653 3958Fiscal expendituredecentralization sd 0368 0108 0138 0294 0351 0421 0888

Fiscal revenuedecentralization rd 0242 0146 0037 0133 0202 0318 0871

Per capita real GDP(yuanperson) pgdp 27119890 + 04 32119890 + 04 1891200 9837380 17119890 + 04 31119890 + 04 44119890 + 05

Urbanization () Urban 0345 0195 0013 0202 0296 0451 100000Industrial structure itu 2203 0136 1318 2110 2190 2283 2797Informationtechnologydevelopment (tenthousand families)

Internet 41066 128506 0024 8316 17139 36800 5174000

Fixed assetinvestment (tenthousand yuan)

ifa 15119890 + 04 15119890 + 04 654299 4606420 9841960 20119890 + 04 17119890 + 05

The level of internalopening () Private 0086 0106 0003 0036 0059 0096 100000

and technology flowing of financial resources among regionswhich causes regional financial development and innovationand imitation and strategic competition among regions [37]on the other hand the restriction of transportation cost onfinancial sector is not strong and the rapid development ofinternet technology leads to cross-regional financial transac-tions and innovation thus further expanding the clusteringeffect effect of economic scale and spatial spillover effectsof financial development The coefficients of lag item (fe

119894119905-1)of financial efficiency in DSDM models are positive andthey are respectively 0211 and 0212 both of them pass thetests at the 1 significance level indicating that the previousinputs and outputs of production will be present through thefinancial efficiency increasing in later stage (or later period)and financial efficiency increasing is a dynamic systemicprocess The specific impact of each variable on financialefficiency is analyzed as follows

The estimated coefficients of Lnsd and Lnrd are positiveand all of them pass the tests at the 1 significance levelwhich indicates that the fiscal expenditure and revenuedecentralization play a significant positive role in promotinglocal financial efficiency Under the dual pressures of makingup the fiscal gap and political promotion the local govern-ments regarding economic development as main missionhave instinctive strong interest in scarce flowing economicfactors [38] As the excellent combination of capital talenttechnology management and marketing financial resourceswill be the important target for battling among local govern-ments With the expansion of fiscal expenditure decentral-ization and revenue decentralization local government usesmany policy tools such as fiscal subsidies tax concessionsand government procurement to compete for getting thefinancial resources by attracting financial capital humancapital and technological innovation to flow into the localregion thus contributing to the financial innovation and

improvement of financial efficiencyThe influence coefficientof fiscal expenditure decentralization is 0357 greater thanthe influence coefficient (0153) of fiscal revenue decentral-ization indicating that the impacts of fiscal expendituredecentralization and revenue decentralization on regionalfinancial efficiency have unbalanced characteristics and thepossible reason is that local governments under provinceget very limited taxation autonomy resulting in the factthat revenue decentralization is often lower than expendi-ture decentralization In addition under the capital-driveneconomic growth model the flowing direction of capitaldepends on the level of infrastructure [39] thus the inflowof financial resources as well as financial innovation oftenoccurs in the regions with rapid economic developmentexcellent investment environment and great consumptionability so that local governments have to increase fiscalexpenditure to enhance the supply of the associated support-ing infrastructure Comparatively speaking tax preferenceof undeveloped regions is less attractive than the superiorityconditions of developed regions

We will further analyze the estimated coefficients andsignificance levels of spatial lags of fiscal expenditure andrevenue decentralization and investigate the strategic com-petitive effects of financial efficiency among neighboringregions The spatial lag of expenditure decentralization(WLnsd) does not pass the significance test suggestingthat spatial spillover effect of fiscal expenditure decentral-ization is uncertain and needs further examination whilethe spatial lag of fiscal revenue decentralization (WLnrd)is negative at the 5 significance level and the estimatedcoefficient is minus0228 indicating that with the increase offiscal revenue decentralization the negative externalities onfinancial development and innovation will be increasinglyprominent thus inhibiting the improvement of financialefficiency in neighboring regions The reason may be that

Discrete Dynamics in Nature and Society 9

Table 3 Estimated results of regional financial efficiency under theSpatial Durbin Model (economic-geographic weighting matrix)

Independentvariables

Dependent variable sd Dependent variable rd

SDMmodel DSDMmodel SDMmodel DSDM

model

feitminus10211lowastlowastlowast(1283)

0212lowastlowastlowast(1283)

Lnsd 0357lowastlowastlowast(626)

0334lowastlowastlowast(570)

Lnrd 0153lowastlowastlowast(470)

0143lowastlowastlowast(408)

Lnpgdp 0169lowastlowastlowast(462)

0120lowastlowastlowast(308)

0138lowastlowastlowast(360)

0099lowastlowast(244)

Lnurban minus0050lowastlowast(minus237)

minus0044lowastlowast(minus209)

minus0050lowastlowast(minus237)

minus0045lowastlowast(minus215)

Lnitu 0433lowastlowast(227)

0438lowastlowast(215)

0525lowastlowastlowast(273)

0531lowastlowastlowast(259)

Lninternet minus0295lowastlowastlowast(minus2970)

minus0299lowastlowastlowast(minus2716)

minus0292lowastlowastlowast(minus2945)

minus0297lowastlowastlowast(minus2690)

Lnifa minus0446lowastlowastlowast(minus2298)

minus0386lowastlowastlowast(minus1832)

minus0437lowastlowastlowast(minus2321)

minus0374lowastlowastlowast(minus1806)

Lnprivate 0038lowastlowastlowast(322)

0037lowastlowastlowast(290)

0039lowastlowastlowast(325)

0038lowastlowastlowast(302)

WLnsd 0179(099)

minus0041(minus022)

WLnrd minus0228lowastlowast(minus251)

minus0186lowast(minus179)

WLnpgdp 0016(027)

0064(101)

0058(096)

0097(150)

WLnurban 0097lowastlowast(236)

0082lowastlowast(198)

0107lowastlowastlowast(259)

0089lowastlowast(213)

WLnitu 1159lowast(193)

0516(082)

0615(100)

0228(035)

WLninternet 0177lowastlowastlowast(647)

0181lowastlowastlowast(622)

0180lowastlowastlowast(666)

0180lowastlowastlowast(619)

WLnifa 0237lowastlowastlowast(539)

0251lowastlowastlowast(532)

0275lowastlowastlowast(704)

0260lowastlowastlowast(600)

WLnprivate 0065lowast(185)

0045(115)

0083lowastlowast(232)

0060(153)

120588

0420lowastlowastlowast(1165)

0462lowastlowastlowast(1251)

0429lowastlowastlowast(1205)

0463lowastlowastlowast(1266)

1198772 0448 0443 0441 0438

log 119871 599812 595228 595228 587624Observations 2810 2529 2810 2529Note the values in parentheses represent the 119905-statistics values of thecorresponding estimated coefficientslowastlowastlowastlowastlowast andlowast respectively representsignificance level of 1 5 and 10

local governmentsrsquo motivation of pursuing revenues underthe fiscal revenue decentralization system is not conduciveto free flowing of financial resources among regions whichdirectly affects the financial efficiency in other regions

When it comes to the estimated coefficients of othercontrolled variables we found that not all the spatial spillovereffects of variables are significant Overall various economicactivities have direct and indirect influence on financial

efficiency and promote the strategic financial competitionamong regions The specific effects are as follows the levelof open-into-inside can not only improve the local financialefficiency significantly but also increase the financial effi-ciency correspondingly in neighboring regions The strongdemand for various financial products and services of privateenterprises promotes financial innovation and improves theefficiency of resource allocation in financial markets Onthe contrary urbanization information technology develop-ment and fixed assets investment suppress the improvementof local financial efficiency but play a positive role in promot-ing the financial efficiency in neighboring regionsThe accel-erating of urbanization process the application of internettechnology and investment in fixed assets promote economicexchanges and cooperation among regions so that the spatialspillover effects of financial development and innovation areincreasingly significant Furthermore although the regionaleconomic development and industrial structure significantlycontribute to the improvement of regional financial efficiencyin general the spillover effects among regions are not obvious

52 Decomposition of Fiscal Decentralizationrsquos Effects onRegional Financial Efficiency Whether there truly existspillover effects among the variables relying solely on the spa-tial lag coefficients of SDMmodel and DSDMmodel is likelyto lead to misinterpretation of the estimated results Spatialeffects can be further decomposed into direct effect indirecteffect (spillover effects) and total effect with reference to theresearch result of LeSage and Pace [32] The indirect effectsrepresent the independent variablesrsquo influence on the depen-dent variable in all other regions through spatial interactionsTherefore we will further verify the existence of spillovereffects by using partial differential matrix analysis Table 4illustrates the direct effect indirect effect and total effect ofthe fiscal expenditure and revenue decentralization in SDMmodel The results show that the estimated coefficients ofindependent variablesrsquo direct effect have the same directionwith the estimated coefficients of SDMmodel in Table 3 andthe estimated coefficients and significance levels are also veryclose

According to the estimated results of SDM model inTable 4 the direct effect indirect effect and total effectof fiscal expenditure decentralization on regional financialefficiency are positive and pass the 10 significance leveltest The regression coefficient of direct effect is 0368 theregression coefficient of indirect effect is 0560 and theregression coefficient of total effect is 0928 indicating thatthe degree of fiscal expenditure decentralization increases by1 while regional financial efficiency increases by 0928Fiscal expenditure decentralization plays a significant posi-tive impact on financial development and financial efficiencywhich is opposite to the research result of Wang and Li [23]The opposite conclusions are possibly due to the differencesof sample selection and indexesThe rise of fiscal expendituredecentralization in the local region leads to an increaseby 0368 of the financial efficiency while the increase ofexpenditure decentralization in the neighboring region leadsto financial efficiency increasing by 0560 indicating that

10 Discrete Dynamics in Nature and Society

Table 4 Decomposition of direct effect indirect effect and total effect under SDMmodel

Independent variablesDependent variable sd Dependent variable sd

Direct effect Indirect effect Total effect Direct effect Indirect effect Total effect

Lnsd 0368lowastlowastlowast(750)

0560lowast(195)

0928lowastlowastlowast(310)

Lnrd 0146lowastlowastlowast(529)

minus0283lowast(minus195)

minus0137lowast(minus192)

Lnpgdp 0175lowastlowastlowast(435)

0132(153)

0307lowastlowastlowast(359)

0145lowastlowastlowast(346)

0187lowastlowast(218)

0332lowastlowastlowast(392)

Lnurban minus0046lowastlowast(minus207)

0127lowastlowast(200)

0081(130)

minus0045lowastlowast(minus204)

0146lowastlowast(226)

0101(158)

Lnitu 0486lowastlowastlowast(275)

2321lowastlowast(237)

2807lowastlowastlowast(292)

0559lowastlowastlowast(315)

1488(151)

2047lowastlowast(211)

Lninternet minus0291lowastlowastlowast(minus3055)

0086lowast(190)

minus0205lowastlowastlowast(minus432)

minus0288lowastlowastlowast(minus3029)

0091lowastlowast(200)

minus0197lowastlowastlowast(minus410)

Lnifa minus0441lowastlowastlowast(minus2260)

0087(149)

minus0353lowastlowastlowast(minus594)

minus0430lowastlowastlowast(minus2254)

0154lowastlowastlowast(283)

minus0276lowastlowastlowast(minus494)

Lnprivate 0041lowastlowastlowast(319)

0134lowastlowast(235)

0175lowastlowastlowast(285)

0042lowastlowastlowast(327)

0166lowastlowastlowast(283)

0208lowastlowastlowast(330)

Note lowast lowast lowast lowastlowast and lowast respectively represent the significant level of 1 5 and 10

fiscal expenditure decentralization significantly promotesthe competition of financial development and innovationamong regions The expansion of the autonomy of localgovernment expenditure strengthens the regulatory capacityof local governments which is conducive to acceleratingmarket-oriented reform processing and achieves the organicintegration between fiscal system and market mechanism inthe allocation of financial resources which is beneficial tocreating a favorable institutional environment is also con-ducive to the flowing of financial resources among regionsand greatly promotes financial efficiency

The fiscal revenue decentralizationrsquos direct effect onregional financial efficiency is significantly positive whilethe spillover effect is significantly negative indicating thatthe impact of fiscal revenue decentralization on regionalfinancial efficiency presents the feature of ldquodouble-edged-swordrdquo Fiscal revenue decentralization has a significant posi-tive impact on the local financial efficiencyHowever there aresignificant negative spillover effects on the financial efficiencyin other regions The regression coefficient of the directeffect is 0146 the regression coefficient of the indirect effectis minus0283 and the regression coefficient of the total effectis minus0137 which indicates that the degree of fiscal revenuedecentralization increasing by 1 will lead to decrease by0137 of regional financial efficiency Furthermore whilethe increase of local fiscal revenue decentralization leads tofinancial efficiencyrsquos increase by 0146 the increase of thefiscal revenue decentralization in neighboring regions resultsin the financial efficiencyrsquos decrease by 0283 Comparingthe estimated coefficient of direct effect and that of indirecteffect it can be found that the negative spillover effectsof fiscal revenue decentralization are significantly greaterthan the direct scale effect so the negative externalitiesof fiscal revenue decentralization on financial efficiencyare greater than the positive externality On one hand inorder to obtain high-quality financial resources the local

governments ignore the differences like the level of economicdevelopment factor endowments and other aspects amongregions and seek to imitate policies and instruments inother regions thus leading to chain reaction of absorbinginvestment by other regions On the other hand whenthe neighboring regions or the regions with similar levelof economic development take some measures to attractliquidity factors the local government driven by yardstickcompetitionwill adopt similar policies to avoid the outflow ofelements which results in convergence of competitive tactics[40 41] Tax incentives reduction of fee and other wayswill become important means of competition and imitationbetween local governments in different regions on the basisof the independent revenue authority However such kind ofcompetitive mode between local governments for resourceswill consume huge fiscal resources and weaken the capacityof local governmentrsquos provision of public goods and servicesUltimately it is harmful to provide healthy stable andsustainable economic development environment for financialinnovation and increase of financial efficiency

53 Robustness Test To ensure the robustness and reliabilityof research results we will calculate the spatial lags withthe geographical distance weight matrix and do econometricregression by taking financial efficiency as the dependentvariable so that we can comparatively analyze the differenceof influence with different spatial weight setting based on theabove The robustness test results are as in Table 5

As shown in Table 5 the estimated results indicate thatthe sign of regression coefficients of the fiscal expendituredecentralization and revenue decentralization is consistentwith that of corresponding variable in the model of Table 3and the regression results in themodel are basically consistentwith the conclusions in Table 3 Comparing the estimatedresults based on two different weightmatrixes it is found thatthere are some differences in the influence coefficients of two

Discrete Dynamics in Nature and Society 11

Table 5 Robustness test on Spatial Durbin Models (geographicaldistance weight matrix)

Independentvariables

Dependent variable sd Dependent variable rdSDMmodel

DSDMmodel

SDMmodel

DSDMmodel

feitminus1 mdash 0202lowastlowastlowast(1276) mdash 0200lowastlowastlowast

(1255)

Lnsd or Lnrd 0364lowastlowastlowast(636)

0348lowastlowastlowast(592)

0173lowastlowastlowast(522)

0161lowastlowastlowast(458)

Lnpgdp 0268lowastlowastlowast(735)

0225lowastlowastlowast(578)

0242lowastlowastlowast(637)

0209lowastlowastlowast(522)

Lnurban minus0026(minus112)

minus0026(minus112)

minus0028(minus120)

minus0027(minus115)

Lnitu 0465lowastlowast(246)

0478lowastlowast(238)

0527lowastlowastlowast(278)

0542lowastlowastlowast(269)

Lninternet minus0295lowastlowastlowast(minus3038)

minus0300lowastlowastlowast(minus2799)

minus0292lowastlowastlowast(minus3005)

minus0298lowastlowastlowast(minus2771)

Lnifa minus0484lowastlowastlowast(minus2479)

minus0427lowastlowastlowast(minus2024)

minus0472lowastlowastlowast(minus2448)

minus0417lowastlowastlowast(minus1988)

Lnprivate 0036lowastlowastlowast(312)

0035lowastlowastlowast(284)

0035lowastlowastlowast(307)

0035lowastlowastlowast(288)

WLnsd orWLnrd

minus0264(minus149)

minus0418lowastlowast(minus225)

minus0277lowastlowastlowast(minus370)

minus0241lowastlowastlowast(minus276)

WLnpgdp minus0269lowastlowastlowast(minus427)

minus0286lowastlowastlowast(minus428)

minus0237lowastlowastlowast(minus369)

minus0267lowastlowastlowast(minus394)

WLnurban 0053(147)

0059lowast(167)

0059(163)

0062lowast(173)

WLnitu 0104(020)

minus0085(minus016)

minus0205(minus039)

minus0290(minus051)

WLninternet 0217lowastlowastlowast(838)

0239lowastlowastlowast(853)

0214lowastlowastlowast(831)

0235lowastlowastlowast(841)

WLnifa 0434lowastlowastlowast(1052)

0464lowastlowastlowast(1054)

0433lowastlowastlowast(1165)

0450lowastlowastlowast(1129)

WLnprivate 0062lowast(187)

0038(106)

0081lowastlowast(241)

0050(139)

120588

0547lowastlowastlowast(1858)

0580lowastlowastlowast(1938)

0541lowastlowastlowast(1839)

0570lowastlowastlowast(1898)

1198772 0444 0428 0443 0430

log 119871 659741 665804 665804 659509Observations 2810 2529 2810 2529Note lowastlowastlowast lowastlowast and lowast respectively represent the significant level of 1 5and 10

kinds of fiscal decentralizations indicators which show thatthe gap between the levels of economic developmentwill havean impact on role of fiscal decentralization acting on regionalfinancial efficiency On one hand under the fiscal decen-tralization system because of the heterogeneity of regionaleconomic development level the local governmentsrsquo ability ofmaking up the gap between fiscal revenue and expenditure isdifferent and the fiscal pressure is also different therefore thedegree of interference to the regional market environment bythe local government is different which leads to the regionaldifferences in financial efficiency in a certain degree [21]On the other hand the occurrence of financial innovationdepends on the support of the economic foundation and

the different economic and social environment will lead tothe fact that the financial innovation activities are more andmore concentrated in the regions with the advantages offinancial location resulting in nonequilibrium of financialefficiency in different regions [42] Under the performanceassessment method focusing on GDP local governments willnot only consider the strategic behaviors of the neighboringregions but also tend to think about the behaviors of theregions with the similar level of economy In additioncross-regional knowledge diffusion technology spillover andhuman capital flow are more prone to occur among theregions with similar level of economic development so thatthe financial exchanges and contacts among regions are morefrequent It also shows that besides considering the impact ofgeographical distance the economic distance factor shouldbe taken into consideration while we are researching thespatial effects of fiscal decentralization on regional financialinnovation in addition to the consideration of the impact ofgeographical distance the economic distance factor shouldbe considered

6 Conclusions and Enlightenments

Based on the panel data of 281 prefecture-level cities in Chinacovering the period of 2003ndash2012 the regional financialefficiency is measured by the superefficiency SBMmodel thegeographic distance weight matrix and economic-geographymixed weight matrix which can reflect the regional similarityare introduced into the SDM and the spatial effects of thefiscal expenditure and revenue decentralization on regionalfinancial efficiency under province are investigated as wellas the strategic financial competition arising from it Basedon the results of previous empirical analysis we can draw thefollowing basic conclusions

First the geographical distribution of Chinarsquos prefecture-level citiesrsquo financial efficiency presents significant features oftime inertia and spatial dependency Financial developmentand innovation among regions particularly neighboringregions is not independent but shows significant spatialspillover effects The financial efficiency of the local regionwill be significantly affected by the financial efficiency ofadjacent regions The spatial effect is increasingly obviousin the process of financial development and innovation andthe size of the spillover effects is closely related to spatialweight (which is used to measure regional similarities) andmainly shows that there is competition for financial efficiencyamong regions with similar economic attributes and adjacentregions

Second it cannot be ignored that the institutional factorsespecially fiscal decentralization have effect on the financialefficiency and the factors have a decisive role in the allocationof financial resource and the regional distribution of financialinnovation The improvement of fiscal decentralization levelwill help to attract more financial resources thereby promot-ing an increase in financial efficiencyThe improvement of thefinancial autonomy in neighboring jurisdictions will reducethe financial resources flowing into the local region andsuppresses the occurrence of the local financial innovationIn addition there exist significant spatial spillover effects of

12 Discrete Dynamics in Nature and Society

financial development and efficiency caused by fiscal decen-tralization The scope of spatial spillovers is mainly reflectedin two aspects Thus the local governmentsrsquo competition notonly strongly increases the financial efficiency of the regionbut also accelerates interregional competition for financialefficiency

Third fiscal expenditure decentralization and revenuedecentralizationrsquos impact on regional financial efficiencyhave nonequilibrium characteristics First of all there existsasymmetry between fiscal expenditure decentralization andrevenue decentralizationrsquos impact on regional financial effi-ciency Fiscal expenditure decentralization can promoteoptimization of the fiscal expenditure structure and fiscalrevenue decentralization can increase the local governmentsrsquorevenue share so as to provide positive incentives [43] bothof them have a positive and promoting effect on the improve-ment of local financial efficiency and such effect largelydepends on the symmetry between the fiscal expenditure andrevenue decentralization If it is asymmetric between theirproperty rights and the expenditure responsibilities underfiscal decentralization system the local governments maymake excessive intervention to regional financial resourcesallocation tomeet their own demands for financial resourcesNext there exists nonconformance between fiscal expendi-ture decentralization and revenue decentralizationrsquos impacton financial efficiency in neighboring regions Comparedwith the fiscal expenditure decentralization the fiscal revenuedecentralization has more apparent inhibitory effect on thefinancial efficiency in neighboring regions

Our research conclusions have important policy implica-tions that illuminate the relationship between fiscal decen-tralization and financial efficiency

(1) The regional financial efficiency is closely relatedto the location and the economic development of the sur-rounding cities The regional financial efficiency is affectednot only by the financial development and innovation inneighboring regions but also by the impact of the struc-tural differences among regions In this case breaking theadministrativemonopoly of local protectionism and carryingon the ldquorichening neighborrdquo policy of financial developmentare the best choices for the local government to developthe regional economy When making policies of financialdevelopment local governments need to consider their owneconomic development environment and economic featuresin neighboring regions and strengthen communication andcoordination with the surrounding regions so as to avoidthe possible contradictions and conflicts on financial devel-opment policies among regions

(2) The spatial spillover effects of the regional financialefficiency should be divided into two kinds of effects includ-ing global geographic spillover effects and local geographicspillover effects In order to promote the increase of financialefficiency in a better way it should also be based on thehierarchy of spatial effect of and the degree of spillover effectto define the relationship between the government and themarket and central and local governments and among thelocal governments and their role in regional financial devel-opment and innovation so as to better promote financialefficiency

(3) There is a two-way interactive mechanism betweenthe fiscal system and financial development such effectsmay affect the behaviors of the local governments and theallocation of resources and ultimately act on macroeconomicperformance [10] Under the fiscal decentralization systemthe strategic competitive behaviors among local governmentssignificantly affect the liquidity of financial resources and theoccurrence of financial innovation the local governmentsrsquoexpansion of fiscal expenditure and revenue shortfall becomethe main driving force of each region competing for financialresources In order to better play the role of incentivesand constraints of fiscal competition mechanism under thefiscal decentralization system and avoid the negative impactbrought by the vicious interregional strategic financial com-petitive behaviors on the regional economic development thelocal governmentsrsquo short-sighted behaviors should be strictlyregulated and then the interregional financial cooperationwith mutual benefit as the core should be actively promoted

(4) Constructing a scientific and rational framework forfiscal decentralization will be an important force to accelerateeffectively the improvement of regional financial efficiencyChinarsquos future subprovincial fiscal decentralization reformshould adapt to the new normal financial development andfurther rationalize the provincial and subprovincial fiscaldecentralization system to ensure a balanced and appropriatefiscal expenditure and revenue decentralization On onehand we should weaken local governmentsrsquo negative inter-vention in financial system by reducing local governmentsrsquofinancial pressure caused by the asymmetry between thefiscal resources and expenditure responsibilities On the otherhandwe should also actively play the function of fiscal systemto adjust the financial development and innovation and fullyplay public financersquos role in promoting the flowing of finan-cial resources among regions the balanced development ofregional finance and the improvement of financial efficiency

Competing Interests

There are no competing interests in regard to the contentdiscussed in this paper

Authorsrsquo Contributions

All authors conceived and designed the research XiaomeiHucollected and processed the data Jianmin Liu analyzed thedata and interpreted the results all authors were also involvedin writing the paper and have all approved the submittedform

Acknowledgments

The research of this paper is jointly funded by the NationalSocial Science Foundation of China under Project no14BJY159 and the Hunan Philosophy and Social ScienceFund Project under Grant no 13JD09 The authors gratefullyacknowledge the support of Jinguang Wu from HunanUniversity of Finance and Economics China

Discrete Dynamics in Nature and Society 13

References

[1] J Zhang and X Jin ldquoRe-examine of the relationship betweenChinarsquos financial deepening and productivity 1987ndash2001rdquo Eco-nomic Research Journal vol 11 pp 35ndash43 2005

[2] K Chen A L Hillman and Q Y Gu ldquoFiscal centralization andchanges of local governmentsrsquo behaviorsmdashfrom lsquohands for helprsquoto lsquohands for grabrsquordquoChina EconomicQuarterly vol 1 pp 111ndash1302002

[3] L Zhou and Z M Wang ldquoEmpirical analysis of regionalfinancial development of China and economic growth 1978ndash2000rdquo Financial Research vol 10 pp 1ndash13 2002

[4] L Zhou ldquolsquoSecond financersquo of Chinarsquos financial industry andfinancial division during the reforming periodrdquo The Journal ofWorld Economy vol 6 pp 72ndash79 2003

[5] L Zhou ldquoGradual transition state capacity and fiscal functionof financerdquo Journal of Finance and Economics vol 2 pp 26ndash372005

[6] E L He and W Liu ldquoFinancial capital local governmentsrsquointervention and industrial structuremdashan empirical analysisbased on the data of central regionsrdquo Economic Issues vol 5pp 37ndash40 2011

[7] X L Dou and W J Wang ldquoThe differences and trends of Chi-nese regional financial development and local governmentsrsquobehavior and analysis of tax sharing reform on the differencesand trends of Chinese regional financial developmentrdquo Shang-hai Finance vol 2 pp 9ndash15 2011

[8] J Zhang and K R Shen ldquoLocal government interventionregional financial development and the transformation ofChinarsquos economic growth modemdashan empirical study under thefiscal decentralization systemrdquo Nankai Economic Studies vol 6pp 122ndash141 2008

[9] G Chen X G Yin and Y Pan ldquoChinarsquos financial developmenttax sharing reform and economic growthrdquo Financial Researchvol 2 pp 99ndash109 2006

[10] P P Ding and Y Fu ldquoBehaviors of local governments therelationship of fiscal and finance and macroeconomic volatilityassociated with Chinamdashan analysis under Chinese-style fiscaldecentralization systemrdquo Comparative Economic amp Social Sys-tems vol 6 pp 87ndash97 2012

[11] X G Yin G Chen and Y Pan ldquoTax sharing reform local gov-ernmentsrsquo intervention and financial development efficiencyrdquoJournal of Finance and Economics vol 10 pp 92ndash101 2006

[12] M G Yu and H B Pan ldquoGovernment intervention the ruleof law financial development and bank loans of state-ownedenterprisesrdquo Financial Research vol 9 pp 1ndash23 2008

[13] D X Wang J Liu and L Li ldquoFiscal decentralization bankcredit and total factor productivityrdquo Journal of Finance andEconomics vol 4 pp 69ndash80 2011

[14] Y X Jiang and R B Jiang ldquoRegional financial innovationefficiency evaluation environmental impact and gap analysisrdquoJournal of Zhejiang University (Humanities and Social Sciences)vol 43 no 4 pp 52ndash65 2013

[15] H W Yang L W Cheng Y Xu and Y W Qi ldquoMeasure andinfluencing factors of financial resources allocation efficiencyunder fiscal decentralization system based on super-efficiencyDEA-TOBIT two-step methodrdquo Journal of Jiangxi NormalUniversity (Natural Sciences Edition) vol 6 pp 587ndash592 2014

[16] N Shen F Z Liu and J Q Zhao ldquoFiscal decentralizationfinancial deepening and regional international poverty prone todeveloprdquo Finance amp Trade Economics vol 1 pp 41ndash46 2006

[17] Q Y Du G P Li and N Shen ldquoFinancial efficiency andoptimization of the regional industrial structure under thecurrent fiscal systemrdquo Xinjiang Social Science vol 4 pp 16ndash212006

[18] T Feng Y W Song and Y Lu ldquoFiscal decentralization localgovernmentsrsquo behavior and regional financial developmentrdquoJournal of Xirsquoan Jiaotong University vol 5 pp 20ndash27 2007

[19] Y A Zhou X J Ma and J Y Zhao ldquoGovernment behaviorsfinancial development and economic growthrdquo Henan SocialSciences vol 1 pp 55ndash60 2007

[20] C Hao ldquoDevelopment of financial intermediation and eco-nomic growth the Chinese experiencerdquo China EconomicReview vol 17 no 4 pp 347ndash362 2006

[21] J Zhang Fiscal Decentralization Regional Financial Devel-opment and Chinarsquos Economic Growth Performance NanjingUniversity Press Nanjing China 2011

[22] Y J Yao ldquoGovernment intervention the development of finan-cial intermediation and economic growthrdquo Journal of Financeand Economics vol 8 pp 52ndash58 2010

[23] X Wang and L Li ldquoFiscal decentralization and regional finan-cial development the theoretical and empirical analysis fromthe perspective of central governmentrdquo Public Finance Researchvol 4 pp 31ndash35 2012

[24] W L Zhang and C L Qin ldquoFinancial development fiscaldecentralization and regional economic differencesrdquo EconomicGeography vol 1 pp 583ndash588 2009

[25] Y Ma J Li and B Chen ldquoProvincial differences in Chinesefiscal decentralization financial development industrializationand economic growthrdquo EconomicTheory and Business Manage-ment vol 2 pp 5ndash19 2015

[26] Y X Jiang and R B Jiang ldquoThe spatial external effects ofregional financial innovation financial competition and finan-cial agglomerationrdquo Social Science Front vol 3 pp 79ndash84 2014

[27] K Tone ldquoA slacks-based measure of super-efficiency indata envelopment analysisrdquo European Journal of OperationalResearch vol 143 no 1 pp 32ndash41 2002

[28] Y Q Lu and D G Zhang ldquoChinarsquos regional financial efficiencyand differences of efficiency measurementrdquo Economic Geogra-phy vol 1 pp 96ndash101 2012

[29] A N Berger and D B Humphrey ldquoEfficiency of financialinstitutions international survey and directions for futureresearchrdquo European Journal of Operational Research vol 98 no2 pp 175ndash212 1997

[30] Y J Yao ldquoEmpirical analysis of financial development andurban-rural income gaprdquo Journal of Finance and Economics vol2 pp 49ndash59 2005

[31] Q Li J Song E Wang H Hu J Zhang and Y WangldquoEconomic growth and pollutant emissions in China a spatialeconometric analysisrdquo Stochastic Environmental Research andRisk Assessment vol 28 no 2 pp 429ndash442 2014

[32] J P LeSage and R K Pace Introduction to Spatial EconometricsCRCPress Taylor amp Francis Group Boca Raton Fla USA 2009

[33] Q W Guo and J X Jia ldquoFiscal decentralization organizationalstructure of local governments and the size of governmentexpenditurerdquoEconomic Research Journal vol 11 pp 59ndash87 2010httpwwwcnkicomcnArticleCJFDTotal-JJYJ201011008htm

[34] F Gong and X Lei ldquoThe measurement of Chinese-style fiscaldecentralizationrdquo Statistical Research vol 10 pp 47ndash55 2010

[35] G Fan X L Wang and H P Zhu Chinese Marketing IndexRegional Marketing Process in 2011 Economic Science PressBeijing China 2011

14 Discrete Dynamics in Nature and Society

[36] S S Rosenthal and W C Strange ldquoGeography industrialorganization and agglomerationrdquoTheReview of Economics andStatistics vol 85 no 2 pp 377ndash393 2003

[37] S Park Yoon International Banking and Financial CentersKluwer Academic Boston Mass USA 1989

[38] B Y Qiao J Y Fan and X Y Feng ldquoChinarsquos fiscal decentral-ization and compulsory primary educationrdquo Social Sciences inChina vol 6 pp 37ndash46 2005

[39] W JWang and C L Qin ldquoRegional differences in the behaviorsof local governments and growth effect of fiscal decentralizationbased on the analysis of judgment hypothesis and testingrdquoManagement World vol 1 pp 9ndash21 2008

[40] J K Brueckner ldquoStrategic interaction among governments anoverview of empirical studiesrdquo International Regional ScienceReview vol 26 no 2 pp 175ndash188 2003

[41] F Revelli ldquoOn spatial public finance empiricsrdquo International Taxand Public Finance vol 12 no 4 pp 475ndash492 2005

[42] N Shen and T Y He ldquoFinancial efficiency and regional eco-nomic growth from the perspective of fiscal decentralizationrdquoReform vol 1 pp 61ndash65 2006

[43] J M Liu X M Hu and J G Wu ldquoThe threshold effect ofsub-national fiscal decentralization revenue and expendituredecentralization on provincial industrial transformation andupgrading based on the data of 14 cities in Hunan ProvincerdquoPublic Finance Research vol 8 pp 49ndash52 2014

Submit your manuscripts athttpwwwhindawicom

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

MathematicsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Mathematical Problems in Engineering

Hindawi Publishing Corporationhttpwwwhindawicom

Differential EquationsInternational Journal of

Volume 2014

Applied MathematicsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Probability and StatisticsHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Mathematical PhysicsAdvances in

Complex AnalysisJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

OptimizationJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

CombinatoricsHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Operations ResearchAdvances in

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Function Spaces

Abstract and Applied AnalysisHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of Mathematics and Mathematical Sciences

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Algebra

Discrete Dynamics in Nature and Society

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Decision SciencesAdvances in

Discrete MathematicsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom

Volume 2014 Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Stochastic AnalysisInternational Journal of

8 Discrete Dynamics in Nature and Society

Table 2 Statistical description of samplersquos variables

Index Variables Mean Standarddeviation Minimum 25 quantile 50 quantile 75 quantile Maximum

Finance efficiency fe 0576 0624 0133 0380 0500 0653 3958Fiscal expendituredecentralization sd 0368 0108 0138 0294 0351 0421 0888

Fiscal revenuedecentralization rd 0242 0146 0037 0133 0202 0318 0871

Per capita real GDP(yuanperson) pgdp 27119890 + 04 32119890 + 04 1891200 9837380 17119890 + 04 31119890 + 04 44119890 + 05

Urbanization () Urban 0345 0195 0013 0202 0296 0451 100000Industrial structure itu 2203 0136 1318 2110 2190 2283 2797Informationtechnologydevelopment (tenthousand families)

Internet 41066 128506 0024 8316 17139 36800 5174000

Fixed assetinvestment (tenthousand yuan)

ifa 15119890 + 04 15119890 + 04 654299 4606420 9841960 20119890 + 04 17119890 + 05

The level of internalopening () Private 0086 0106 0003 0036 0059 0096 100000

and technology flowing of financial resources among regionswhich causes regional financial development and innovationand imitation and strategic competition among regions [37]on the other hand the restriction of transportation cost onfinancial sector is not strong and the rapid development ofinternet technology leads to cross-regional financial transac-tions and innovation thus further expanding the clusteringeffect effect of economic scale and spatial spillover effectsof financial development The coefficients of lag item (fe

119894119905-1)of financial efficiency in DSDM models are positive andthey are respectively 0211 and 0212 both of them pass thetests at the 1 significance level indicating that the previousinputs and outputs of production will be present through thefinancial efficiency increasing in later stage (or later period)and financial efficiency increasing is a dynamic systemicprocess The specific impact of each variable on financialefficiency is analyzed as follows

The estimated coefficients of Lnsd and Lnrd are positiveand all of them pass the tests at the 1 significance levelwhich indicates that the fiscal expenditure and revenuedecentralization play a significant positive role in promotinglocal financial efficiency Under the dual pressures of makingup the fiscal gap and political promotion the local govern-ments regarding economic development as main missionhave instinctive strong interest in scarce flowing economicfactors [38] As the excellent combination of capital talenttechnology management and marketing financial resourceswill be the important target for battling among local govern-ments With the expansion of fiscal expenditure decentral-ization and revenue decentralization local government usesmany policy tools such as fiscal subsidies tax concessionsand government procurement to compete for getting thefinancial resources by attracting financial capital humancapital and technological innovation to flow into the localregion thus contributing to the financial innovation and

improvement of financial efficiencyThe influence coefficientof fiscal expenditure decentralization is 0357 greater thanthe influence coefficient (0153) of fiscal revenue decentral-ization indicating that the impacts of fiscal expendituredecentralization and revenue decentralization on regionalfinancial efficiency have unbalanced characteristics and thepossible reason is that local governments under provinceget very limited taxation autonomy resulting in the factthat revenue decentralization is often lower than expendi-ture decentralization In addition under the capital-driveneconomic growth model the flowing direction of capitaldepends on the level of infrastructure [39] thus the inflowof financial resources as well as financial innovation oftenoccurs in the regions with rapid economic developmentexcellent investment environment and great consumptionability so that local governments have to increase fiscalexpenditure to enhance the supply of the associated support-ing infrastructure Comparatively speaking tax preferenceof undeveloped regions is less attractive than the superiorityconditions of developed regions

We will further analyze the estimated coefficients andsignificance levels of spatial lags of fiscal expenditure andrevenue decentralization and investigate the strategic com-petitive effects of financial efficiency among neighboringregions The spatial lag of expenditure decentralization(WLnsd) does not pass the significance test suggestingthat spatial spillover effect of fiscal expenditure decentral-ization is uncertain and needs further examination whilethe spatial lag of fiscal revenue decentralization (WLnrd)is negative at the 5 significance level and the estimatedcoefficient is minus0228 indicating that with the increase offiscal revenue decentralization the negative externalities onfinancial development and innovation will be increasinglyprominent thus inhibiting the improvement of financialefficiency in neighboring regions The reason may be that

Discrete Dynamics in Nature and Society 9

Table 3 Estimated results of regional financial efficiency under theSpatial Durbin Model (economic-geographic weighting matrix)

Independentvariables

Dependent variable sd Dependent variable rd

SDMmodel DSDMmodel SDMmodel DSDM

model

feitminus10211lowastlowastlowast(1283)

0212lowastlowastlowast(1283)

Lnsd 0357lowastlowastlowast(626)

0334lowastlowastlowast(570)

Lnrd 0153lowastlowastlowast(470)

0143lowastlowastlowast(408)

Lnpgdp 0169lowastlowastlowast(462)

0120lowastlowastlowast(308)

0138lowastlowastlowast(360)

0099lowastlowast(244)

Lnurban minus0050lowastlowast(minus237)

minus0044lowastlowast(minus209)

minus0050lowastlowast(minus237)

minus0045lowastlowast(minus215)

Lnitu 0433lowastlowast(227)

0438lowastlowast(215)

0525lowastlowastlowast(273)

0531lowastlowastlowast(259)

Lninternet minus0295lowastlowastlowast(minus2970)

minus0299lowastlowastlowast(minus2716)

minus0292lowastlowastlowast(minus2945)

minus0297lowastlowastlowast(minus2690)

Lnifa minus0446lowastlowastlowast(minus2298)

minus0386lowastlowastlowast(minus1832)

minus0437lowastlowastlowast(minus2321)

minus0374lowastlowastlowast(minus1806)

Lnprivate 0038lowastlowastlowast(322)

0037lowastlowastlowast(290)

0039lowastlowastlowast(325)

0038lowastlowastlowast(302)

WLnsd 0179(099)

minus0041(minus022)

WLnrd minus0228lowastlowast(minus251)

minus0186lowast(minus179)

WLnpgdp 0016(027)

0064(101)

0058(096)

0097(150)

WLnurban 0097lowastlowast(236)

0082lowastlowast(198)

0107lowastlowastlowast(259)

0089lowastlowast(213)

WLnitu 1159lowast(193)

0516(082)

0615(100)

0228(035)

WLninternet 0177lowastlowastlowast(647)

0181lowastlowastlowast(622)

0180lowastlowastlowast(666)

0180lowastlowastlowast(619)

WLnifa 0237lowastlowastlowast(539)

0251lowastlowastlowast(532)

0275lowastlowastlowast(704)

0260lowastlowastlowast(600)

WLnprivate 0065lowast(185)

0045(115)

0083lowastlowast(232)

0060(153)

120588

0420lowastlowastlowast(1165)

0462lowastlowastlowast(1251)

0429lowastlowastlowast(1205)

0463lowastlowastlowast(1266)

1198772 0448 0443 0441 0438

log 119871 599812 595228 595228 587624Observations 2810 2529 2810 2529Note the values in parentheses represent the 119905-statistics values of thecorresponding estimated coefficientslowastlowastlowastlowastlowast andlowast respectively representsignificance level of 1 5 and 10

local governmentsrsquo motivation of pursuing revenues underthe fiscal revenue decentralization system is not conduciveto free flowing of financial resources among regions whichdirectly affects the financial efficiency in other regions

When it comes to the estimated coefficients of othercontrolled variables we found that not all the spatial spillovereffects of variables are significant Overall various economicactivities have direct and indirect influence on financial

efficiency and promote the strategic financial competitionamong regions The specific effects are as follows the levelof open-into-inside can not only improve the local financialefficiency significantly but also increase the financial effi-ciency correspondingly in neighboring regions The strongdemand for various financial products and services of privateenterprises promotes financial innovation and improves theefficiency of resource allocation in financial markets Onthe contrary urbanization information technology develop-ment and fixed assets investment suppress the improvementof local financial efficiency but play a positive role in promot-ing the financial efficiency in neighboring regionsThe accel-erating of urbanization process the application of internettechnology and investment in fixed assets promote economicexchanges and cooperation among regions so that the spatialspillover effects of financial development and innovation areincreasingly significant Furthermore although the regionaleconomic development and industrial structure significantlycontribute to the improvement of regional financial efficiencyin general the spillover effects among regions are not obvious

52 Decomposition of Fiscal Decentralizationrsquos Effects onRegional Financial Efficiency Whether there truly existspillover effects among the variables relying solely on the spa-tial lag coefficients of SDMmodel and DSDMmodel is likelyto lead to misinterpretation of the estimated results Spatialeffects can be further decomposed into direct effect indirecteffect (spillover effects) and total effect with reference to theresearch result of LeSage and Pace [32] The indirect effectsrepresent the independent variablesrsquo influence on the depen-dent variable in all other regions through spatial interactionsTherefore we will further verify the existence of spillovereffects by using partial differential matrix analysis Table 4illustrates the direct effect indirect effect and total effect ofthe fiscal expenditure and revenue decentralization in SDMmodel The results show that the estimated coefficients ofindependent variablesrsquo direct effect have the same directionwith the estimated coefficients of SDMmodel in Table 3 andthe estimated coefficients and significance levels are also veryclose

According to the estimated results of SDM model inTable 4 the direct effect indirect effect and total effectof fiscal expenditure decentralization on regional financialefficiency are positive and pass the 10 significance leveltest The regression coefficient of direct effect is 0368 theregression coefficient of indirect effect is 0560 and theregression coefficient of total effect is 0928 indicating thatthe degree of fiscal expenditure decentralization increases by1 while regional financial efficiency increases by 0928Fiscal expenditure decentralization plays a significant posi-tive impact on financial development and financial efficiencywhich is opposite to the research result of Wang and Li [23]The opposite conclusions are possibly due to the differencesof sample selection and indexesThe rise of fiscal expendituredecentralization in the local region leads to an increaseby 0368 of the financial efficiency while the increase ofexpenditure decentralization in the neighboring region leadsto financial efficiency increasing by 0560 indicating that

10 Discrete Dynamics in Nature and Society

Table 4 Decomposition of direct effect indirect effect and total effect under SDMmodel

Independent variablesDependent variable sd Dependent variable sd

Direct effect Indirect effect Total effect Direct effect Indirect effect Total effect

Lnsd 0368lowastlowastlowast(750)

0560lowast(195)

0928lowastlowastlowast(310)

Lnrd 0146lowastlowastlowast(529)

minus0283lowast(minus195)

minus0137lowast(minus192)

Lnpgdp 0175lowastlowastlowast(435)

0132(153)

0307lowastlowastlowast(359)

0145lowastlowastlowast(346)

0187lowastlowast(218)

0332lowastlowastlowast(392)

Lnurban minus0046lowastlowast(minus207)

0127lowastlowast(200)

0081(130)

minus0045lowastlowast(minus204)

0146lowastlowast(226)

0101(158)

Lnitu 0486lowastlowastlowast(275)

2321lowastlowast(237)

2807lowastlowastlowast(292)

0559lowastlowastlowast(315)

1488(151)

2047lowastlowast(211)

Lninternet minus0291lowastlowastlowast(minus3055)

0086lowast(190)

minus0205lowastlowastlowast(minus432)

minus0288lowastlowastlowast(minus3029)

0091lowastlowast(200)

minus0197lowastlowastlowast(minus410)

Lnifa minus0441lowastlowastlowast(minus2260)

0087(149)

minus0353lowastlowastlowast(minus594)

minus0430lowastlowastlowast(minus2254)

0154lowastlowastlowast(283)

minus0276lowastlowastlowast(minus494)

Lnprivate 0041lowastlowastlowast(319)

0134lowastlowast(235)

0175lowastlowastlowast(285)

0042lowastlowastlowast(327)

0166lowastlowastlowast(283)

0208lowastlowastlowast(330)

Note lowast lowast lowast lowastlowast and lowast respectively represent the significant level of 1 5 and 10

fiscal expenditure decentralization significantly promotesthe competition of financial development and innovationamong regions The expansion of the autonomy of localgovernment expenditure strengthens the regulatory capacityof local governments which is conducive to acceleratingmarket-oriented reform processing and achieves the organicintegration between fiscal system and market mechanism inthe allocation of financial resources which is beneficial tocreating a favorable institutional environment is also con-ducive to the flowing of financial resources among regionsand greatly promotes financial efficiency

The fiscal revenue decentralizationrsquos direct effect onregional financial efficiency is significantly positive whilethe spillover effect is significantly negative indicating thatthe impact of fiscal revenue decentralization on regionalfinancial efficiency presents the feature of ldquodouble-edged-swordrdquo Fiscal revenue decentralization has a significant posi-tive impact on the local financial efficiencyHowever there aresignificant negative spillover effects on the financial efficiencyin other regions The regression coefficient of the directeffect is 0146 the regression coefficient of the indirect effectis minus0283 and the regression coefficient of the total effectis minus0137 which indicates that the degree of fiscal revenuedecentralization increasing by 1 will lead to decrease by0137 of regional financial efficiency Furthermore whilethe increase of local fiscal revenue decentralization leads tofinancial efficiencyrsquos increase by 0146 the increase of thefiscal revenue decentralization in neighboring regions resultsin the financial efficiencyrsquos decrease by 0283 Comparingthe estimated coefficient of direct effect and that of indirecteffect it can be found that the negative spillover effectsof fiscal revenue decentralization are significantly greaterthan the direct scale effect so the negative externalitiesof fiscal revenue decentralization on financial efficiencyare greater than the positive externality On one hand inorder to obtain high-quality financial resources the local

governments ignore the differences like the level of economicdevelopment factor endowments and other aspects amongregions and seek to imitate policies and instruments inother regions thus leading to chain reaction of absorbinginvestment by other regions On the other hand whenthe neighboring regions or the regions with similar levelof economic development take some measures to attractliquidity factors the local government driven by yardstickcompetitionwill adopt similar policies to avoid the outflow ofelements which results in convergence of competitive tactics[40 41] Tax incentives reduction of fee and other wayswill become important means of competition and imitationbetween local governments in different regions on the basisof the independent revenue authority However such kind ofcompetitive mode between local governments for resourceswill consume huge fiscal resources and weaken the capacityof local governmentrsquos provision of public goods and servicesUltimately it is harmful to provide healthy stable andsustainable economic development environment for financialinnovation and increase of financial efficiency

53 Robustness Test To ensure the robustness and reliabilityof research results we will calculate the spatial lags withthe geographical distance weight matrix and do econometricregression by taking financial efficiency as the dependentvariable so that we can comparatively analyze the differenceof influence with different spatial weight setting based on theabove The robustness test results are as in Table 5

As shown in Table 5 the estimated results indicate thatthe sign of regression coefficients of the fiscal expendituredecentralization and revenue decentralization is consistentwith that of corresponding variable in the model of Table 3and the regression results in themodel are basically consistentwith the conclusions in Table 3 Comparing the estimatedresults based on two different weightmatrixes it is found thatthere are some differences in the influence coefficients of two

Discrete Dynamics in Nature and Society 11

Table 5 Robustness test on Spatial Durbin Models (geographicaldistance weight matrix)

Independentvariables

Dependent variable sd Dependent variable rdSDMmodel

DSDMmodel

SDMmodel

DSDMmodel

feitminus1 mdash 0202lowastlowastlowast(1276) mdash 0200lowastlowastlowast

(1255)

Lnsd or Lnrd 0364lowastlowastlowast(636)

0348lowastlowastlowast(592)

0173lowastlowastlowast(522)

0161lowastlowastlowast(458)

Lnpgdp 0268lowastlowastlowast(735)

0225lowastlowastlowast(578)

0242lowastlowastlowast(637)

0209lowastlowastlowast(522)

Lnurban minus0026(minus112)

minus0026(minus112)

minus0028(minus120)

minus0027(minus115)

Lnitu 0465lowastlowast(246)

0478lowastlowast(238)

0527lowastlowastlowast(278)

0542lowastlowastlowast(269)

Lninternet minus0295lowastlowastlowast(minus3038)

minus0300lowastlowastlowast(minus2799)

minus0292lowastlowastlowast(minus3005)

minus0298lowastlowastlowast(minus2771)

Lnifa minus0484lowastlowastlowast(minus2479)

minus0427lowastlowastlowast(minus2024)

minus0472lowastlowastlowast(minus2448)

minus0417lowastlowastlowast(minus1988)

Lnprivate 0036lowastlowastlowast(312)

0035lowastlowastlowast(284)

0035lowastlowastlowast(307)

0035lowastlowastlowast(288)

WLnsd orWLnrd

minus0264(minus149)

minus0418lowastlowast(minus225)

minus0277lowastlowastlowast(minus370)

minus0241lowastlowastlowast(minus276)

WLnpgdp minus0269lowastlowastlowast(minus427)

minus0286lowastlowastlowast(minus428)

minus0237lowastlowastlowast(minus369)

minus0267lowastlowastlowast(minus394)

WLnurban 0053(147)

0059lowast(167)

0059(163)

0062lowast(173)

WLnitu 0104(020)

minus0085(minus016)

minus0205(minus039)

minus0290(minus051)

WLninternet 0217lowastlowastlowast(838)

0239lowastlowastlowast(853)

0214lowastlowastlowast(831)

0235lowastlowastlowast(841)

WLnifa 0434lowastlowastlowast(1052)

0464lowastlowastlowast(1054)

0433lowastlowastlowast(1165)

0450lowastlowastlowast(1129)

WLnprivate 0062lowast(187)

0038(106)

0081lowastlowast(241)

0050(139)

120588

0547lowastlowastlowast(1858)

0580lowastlowastlowast(1938)

0541lowastlowastlowast(1839)

0570lowastlowastlowast(1898)

1198772 0444 0428 0443 0430

log 119871 659741 665804 665804 659509Observations 2810 2529 2810 2529Note lowastlowastlowast lowastlowast and lowast respectively represent the significant level of 1 5and 10

kinds of fiscal decentralizations indicators which show thatthe gap between the levels of economic developmentwill havean impact on role of fiscal decentralization acting on regionalfinancial efficiency On one hand under the fiscal decen-tralization system because of the heterogeneity of regionaleconomic development level the local governmentsrsquo ability ofmaking up the gap between fiscal revenue and expenditure isdifferent and the fiscal pressure is also different therefore thedegree of interference to the regional market environment bythe local government is different which leads to the regionaldifferences in financial efficiency in a certain degree [21]On the other hand the occurrence of financial innovationdepends on the support of the economic foundation and

the different economic and social environment will lead tothe fact that the financial innovation activities are more andmore concentrated in the regions with the advantages offinancial location resulting in nonequilibrium of financialefficiency in different regions [42] Under the performanceassessment method focusing on GDP local governments willnot only consider the strategic behaviors of the neighboringregions but also tend to think about the behaviors of theregions with the similar level of economy In additioncross-regional knowledge diffusion technology spillover andhuman capital flow are more prone to occur among theregions with similar level of economic development so thatthe financial exchanges and contacts among regions are morefrequent It also shows that besides considering the impact ofgeographical distance the economic distance factor shouldbe taken into consideration while we are researching thespatial effects of fiscal decentralization on regional financialinnovation in addition to the consideration of the impact ofgeographical distance the economic distance factor shouldbe considered

6 Conclusions and Enlightenments

Based on the panel data of 281 prefecture-level cities in Chinacovering the period of 2003ndash2012 the regional financialefficiency is measured by the superefficiency SBMmodel thegeographic distance weight matrix and economic-geographymixed weight matrix which can reflect the regional similarityare introduced into the SDM and the spatial effects of thefiscal expenditure and revenue decentralization on regionalfinancial efficiency under province are investigated as wellas the strategic financial competition arising from it Basedon the results of previous empirical analysis we can draw thefollowing basic conclusions

First the geographical distribution of Chinarsquos prefecture-level citiesrsquo financial efficiency presents significant features oftime inertia and spatial dependency Financial developmentand innovation among regions particularly neighboringregions is not independent but shows significant spatialspillover effects The financial efficiency of the local regionwill be significantly affected by the financial efficiency ofadjacent regions The spatial effect is increasingly obviousin the process of financial development and innovation andthe size of the spillover effects is closely related to spatialweight (which is used to measure regional similarities) andmainly shows that there is competition for financial efficiencyamong regions with similar economic attributes and adjacentregions

Second it cannot be ignored that the institutional factorsespecially fiscal decentralization have effect on the financialefficiency and the factors have a decisive role in the allocationof financial resource and the regional distribution of financialinnovation The improvement of fiscal decentralization levelwill help to attract more financial resources thereby promot-ing an increase in financial efficiencyThe improvement of thefinancial autonomy in neighboring jurisdictions will reducethe financial resources flowing into the local region andsuppresses the occurrence of the local financial innovationIn addition there exist significant spatial spillover effects of

12 Discrete Dynamics in Nature and Society

financial development and efficiency caused by fiscal decen-tralization The scope of spatial spillovers is mainly reflectedin two aspects Thus the local governmentsrsquo competition notonly strongly increases the financial efficiency of the regionbut also accelerates interregional competition for financialefficiency

Third fiscal expenditure decentralization and revenuedecentralizationrsquos impact on regional financial efficiencyhave nonequilibrium characteristics First of all there existsasymmetry between fiscal expenditure decentralization andrevenue decentralizationrsquos impact on regional financial effi-ciency Fiscal expenditure decentralization can promoteoptimization of the fiscal expenditure structure and fiscalrevenue decentralization can increase the local governmentsrsquorevenue share so as to provide positive incentives [43] bothof them have a positive and promoting effect on the improve-ment of local financial efficiency and such effect largelydepends on the symmetry between the fiscal expenditure andrevenue decentralization If it is asymmetric between theirproperty rights and the expenditure responsibilities underfiscal decentralization system the local governments maymake excessive intervention to regional financial resourcesallocation tomeet their own demands for financial resourcesNext there exists nonconformance between fiscal expendi-ture decentralization and revenue decentralizationrsquos impacton financial efficiency in neighboring regions Comparedwith the fiscal expenditure decentralization the fiscal revenuedecentralization has more apparent inhibitory effect on thefinancial efficiency in neighboring regions

Our research conclusions have important policy implica-tions that illuminate the relationship between fiscal decen-tralization and financial efficiency

(1) The regional financial efficiency is closely relatedto the location and the economic development of the sur-rounding cities The regional financial efficiency is affectednot only by the financial development and innovation inneighboring regions but also by the impact of the struc-tural differences among regions In this case breaking theadministrativemonopoly of local protectionism and carryingon the ldquorichening neighborrdquo policy of financial developmentare the best choices for the local government to developthe regional economy When making policies of financialdevelopment local governments need to consider their owneconomic development environment and economic featuresin neighboring regions and strengthen communication andcoordination with the surrounding regions so as to avoidthe possible contradictions and conflicts on financial devel-opment policies among regions

(2) The spatial spillover effects of the regional financialefficiency should be divided into two kinds of effects includ-ing global geographic spillover effects and local geographicspillover effects In order to promote the increase of financialefficiency in a better way it should also be based on thehierarchy of spatial effect of and the degree of spillover effectto define the relationship between the government and themarket and central and local governments and among thelocal governments and their role in regional financial devel-opment and innovation so as to better promote financialefficiency

(3) There is a two-way interactive mechanism betweenthe fiscal system and financial development such effectsmay affect the behaviors of the local governments and theallocation of resources and ultimately act on macroeconomicperformance [10] Under the fiscal decentralization systemthe strategic competitive behaviors among local governmentssignificantly affect the liquidity of financial resources and theoccurrence of financial innovation the local governmentsrsquoexpansion of fiscal expenditure and revenue shortfall becomethe main driving force of each region competing for financialresources In order to better play the role of incentivesand constraints of fiscal competition mechanism under thefiscal decentralization system and avoid the negative impactbrought by the vicious interregional strategic financial com-petitive behaviors on the regional economic development thelocal governmentsrsquo short-sighted behaviors should be strictlyregulated and then the interregional financial cooperationwith mutual benefit as the core should be actively promoted

(4) Constructing a scientific and rational framework forfiscal decentralization will be an important force to accelerateeffectively the improvement of regional financial efficiencyChinarsquos future subprovincial fiscal decentralization reformshould adapt to the new normal financial development andfurther rationalize the provincial and subprovincial fiscaldecentralization system to ensure a balanced and appropriatefiscal expenditure and revenue decentralization On onehand we should weaken local governmentsrsquo negative inter-vention in financial system by reducing local governmentsrsquofinancial pressure caused by the asymmetry between thefiscal resources and expenditure responsibilities On the otherhandwe should also actively play the function of fiscal systemto adjust the financial development and innovation and fullyplay public financersquos role in promoting the flowing of finan-cial resources among regions the balanced development ofregional finance and the improvement of financial efficiency

Competing Interests

There are no competing interests in regard to the contentdiscussed in this paper

Authorsrsquo Contributions

All authors conceived and designed the research XiaomeiHucollected and processed the data Jianmin Liu analyzed thedata and interpreted the results all authors were also involvedin writing the paper and have all approved the submittedform

Acknowledgments

The research of this paper is jointly funded by the NationalSocial Science Foundation of China under Project no14BJY159 and the Hunan Philosophy and Social ScienceFund Project under Grant no 13JD09 The authors gratefullyacknowledge the support of Jinguang Wu from HunanUniversity of Finance and Economics China

Discrete Dynamics in Nature and Society 13

References

[1] J Zhang and X Jin ldquoRe-examine of the relationship betweenChinarsquos financial deepening and productivity 1987ndash2001rdquo Eco-nomic Research Journal vol 11 pp 35ndash43 2005

[2] K Chen A L Hillman and Q Y Gu ldquoFiscal centralization andchanges of local governmentsrsquo behaviorsmdashfrom lsquohands for helprsquoto lsquohands for grabrsquordquoChina EconomicQuarterly vol 1 pp 111ndash1302002

[3] L Zhou and Z M Wang ldquoEmpirical analysis of regionalfinancial development of China and economic growth 1978ndash2000rdquo Financial Research vol 10 pp 1ndash13 2002

[4] L Zhou ldquolsquoSecond financersquo of Chinarsquos financial industry andfinancial division during the reforming periodrdquo The Journal ofWorld Economy vol 6 pp 72ndash79 2003

[5] L Zhou ldquoGradual transition state capacity and fiscal functionof financerdquo Journal of Finance and Economics vol 2 pp 26ndash372005

[6] E L He and W Liu ldquoFinancial capital local governmentsrsquointervention and industrial structuremdashan empirical analysisbased on the data of central regionsrdquo Economic Issues vol 5pp 37ndash40 2011

[7] X L Dou and W J Wang ldquoThe differences and trends of Chi-nese regional financial development and local governmentsrsquobehavior and analysis of tax sharing reform on the differencesand trends of Chinese regional financial developmentrdquo Shang-hai Finance vol 2 pp 9ndash15 2011

[8] J Zhang and K R Shen ldquoLocal government interventionregional financial development and the transformation ofChinarsquos economic growth modemdashan empirical study under thefiscal decentralization systemrdquo Nankai Economic Studies vol 6pp 122ndash141 2008

[9] G Chen X G Yin and Y Pan ldquoChinarsquos financial developmenttax sharing reform and economic growthrdquo Financial Researchvol 2 pp 99ndash109 2006

[10] P P Ding and Y Fu ldquoBehaviors of local governments therelationship of fiscal and finance and macroeconomic volatilityassociated with Chinamdashan analysis under Chinese-style fiscaldecentralization systemrdquo Comparative Economic amp Social Sys-tems vol 6 pp 87ndash97 2012

[11] X G Yin G Chen and Y Pan ldquoTax sharing reform local gov-ernmentsrsquo intervention and financial development efficiencyrdquoJournal of Finance and Economics vol 10 pp 92ndash101 2006

[12] M G Yu and H B Pan ldquoGovernment intervention the ruleof law financial development and bank loans of state-ownedenterprisesrdquo Financial Research vol 9 pp 1ndash23 2008

[13] D X Wang J Liu and L Li ldquoFiscal decentralization bankcredit and total factor productivityrdquo Journal of Finance andEconomics vol 4 pp 69ndash80 2011

[14] Y X Jiang and R B Jiang ldquoRegional financial innovationefficiency evaluation environmental impact and gap analysisrdquoJournal of Zhejiang University (Humanities and Social Sciences)vol 43 no 4 pp 52ndash65 2013

[15] H W Yang L W Cheng Y Xu and Y W Qi ldquoMeasure andinfluencing factors of financial resources allocation efficiencyunder fiscal decentralization system based on super-efficiencyDEA-TOBIT two-step methodrdquo Journal of Jiangxi NormalUniversity (Natural Sciences Edition) vol 6 pp 587ndash592 2014

[16] N Shen F Z Liu and J Q Zhao ldquoFiscal decentralizationfinancial deepening and regional international poverty prone todeveloprdquo Finance amp Trade Economics vol 1 pp 41ndash46 2006

[17] Q Y Du G P Li and N Shen ldquoFinancial efficiency andoptimization of the regional industrial structure under thecurrent fiscal systemrdquo Xinjiang Social Science vol 4 pp 16ndash212006

[18] T Feng Y W Song and Y Lu ldquoFiscal decentralization localgovernmentsrsquo behavior and regional financial developmentrdquoJournal of Xirsquoan Jiaotong University vol 5 pp 20ndash27 2007

[19] Y A Zhou X J Ma and J Y Zhao ldquoGovernment behaviorsfinancial development and economic growthrdquo Henan SocialSciences vol 1 pp 55ndash60 2007

[20] C Hao ldquoDevelopment of financial intermediation and eco-nomic growth the Chinese experiencerdquo China EconomicReview vol 17 no 4 pp 347ndash362 2006

[21] J Zhang Fiscal Decentralization Regional Financial Devel-opment and Chinarsquos Economic Growth Performance NanjingUniversity Press Nanjing China 2011

[22] Y J Yao ldquoGovernment intervention the development of finan-cial intermediation and economic growthrdquo Journal of Financeand Economics vol 8 pp 52ndash58 2010

[23] X Wang and L Li ldquoFiscal decentralization and regional finan-cial development the theoretical and empirical analysis fromthe perspective of central governmentrdquo Public Finance Researchvol 4 pp 31ndash35 2012

[24] W L Zhang and C L Qin ldquoFinancial development fiscaldecentralization and regional economic differencesrdquo EconomicGeography vol 1 pp 583ndash588 2009

[25] Y Ma J Li and B Chen ldquoProvincial differences in Chinesefiscal decentralization financial development industrializationand economic growthrdquo EconomicTheory and Business Manage-ment vol 2 pp 5ndash19 2015

[26] Y X Jiang and R B Jiang ldquoThe spatial external effects ofregional financial innovation financial competition and finan-cial agglomerationrdquo Social Science Front vol 3 pp 79ndash84 2014

[27] K Tone ldquoA slacks-based measure of super-efficiency indata envelopment analysisrdquo European Journal of OperationalResearch vol 143 no 1 pp 32ndash41 2002

[28] Y Q Lu and D G Zhang ldquoChinarsquos regional financial efficiencyand differences of efficiency measurementrdquo Economic Geogra-phy vol 1 pp 96ndash101 2012

[29] A N Berger and D B Humphrey ldquoEfficiency of financialinstitutions international survey and directions for futureresearchrdquo European Journal of Operational Research vol 98 no2 pp 175ndash212 1997

[30] Y J Yao ldquoEmpirical analysis of financial development andurban-rural income gaprdquo Journal of Finance and Economics vol2 pp 49ndash59 2005

[31] Q Li J Song E Wang H Hu J Zhang and Y WangldquoEconomic growth and pollutant emissions in China a spatialeconometric analysisrdquo Stochastic Environmental Research andRisk Assessment vol 28 no 2 pp 429ndash442 2014

[32] J P LeSage and R K Pace Introduction to Spatial EconometricsCRCPress Taylor amp Francis Group Boca Raton Fla USA 2009

[33] Q W Guo and J X Jia ldquoFiscal decentralization organizationalstructure of local governments and the size of governmentexpenditurerdquoEconomic Research Journal vol 11 pp 59ndash87 2010httpwwwcnkicomcnArticleCJFDTotal-JJYJ201011008htm

[34] F Gong and X Lei ldquoThe measurement of Chinese-style fiscaldecentralizationrdquo Statistical Research vol 10 pp 47ndash55 2010

[35] G Fan X L Wang and H P Zhu Chinese Marketing IndexRegional Marketing Process in 2011 Economic Science PressBeijing China 2011

14 Discrete Dynamics in Nature and Society

[36] S S Rosenthal and W C Strange ldquoGeography industrialorganization and agglomerationrdquoTheReview of Economics andStatistics vol 85 no 2 pp 377ndash393 2003

[37] S Park Yoon International Banking and Financial CentersKluwer Academic Boston Mass USA 1989

[38] B Y Qiao J Y Fan and X Y Feng ldquoChinarsquos fiscal decentral-ization and compulsory primary educationrdquo Social Sciences inChina vol 6 pp 37ndash46 2005

[39] W JWang and C L Qin ldquoRegional differences in the behaviorsof local governments and growth effect of fiscal decentralizationbased on the analysis of judgment hypothesis and testingrdquoManagement World vol 1 pp 9ndash21 2008

[40] J K Brueckner ldquoStrategic interaction among governments anoverview of empirical studiesrdquo International Regional ScienceReview vol 26 no 2 pp 175ndash188 2003

[41] F Revelli ldquoOn spatial public finance empiricsrdquo International Taxand Public Finance vol 12 no 4 pp 475ndash492 2005

[42] N Shen and T Y He ldquoFinancial efficiency and regional eco-nomic growth from the perspective of fiscal decentralizationrdquoReform vol 1 pp 61ndash65 2006

[43] J M Liu X M Hu and J G Wu ldquoThe threshold effect ofsub-national fiscal decentralization revenue and expendituredecentralization on provincial industrial transformation andupgrading based on the data of 14 cities in Hunan ProvincerdquoPublic Finance Research vol 8 pp 49ndash52 2014

Submit your manuscripts athttpwwwhindawicom

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

MathematicsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Mathematical Problems in Engineering

Hindawi Publishing Corporationhttpwwwhindawicom

Differential EquationsInternational Journal of

Volume 2014

Applied MathematicsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Probability and StatisticsHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Mathematical PhysicsAdvances in

Complex AnalysisJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

OptimizationJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

CombinatoricsHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Operations ResearchAdvances in

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Function Spaces

Abstract and Applied AnalysisHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of Mathematics and Mathematical Sciences

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Algebra

Discrete Dynamics in Nature and Society

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Decision SciencesAdvances in

Discrete MathematicsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom

Volume 2014 Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Stochastic AnalysisInternational Journal of

Discrete Dynamics in Nature and Society 9

Table 3 Estimated results of regional financial efficiency under theSpatial Durbin Model (economic-geographic weighting matrix)

Independentvariables

Dependent variable sd Dependent variable rd

SDMmodel DSDMmodel SDMmodel DSDM

model

feitminus10211lowastlowastlowast(1283)

0212lowastlowastlowast(1283)

Lnsd 0357lowastlowastlowast(626)

0334lowastlowastlowast(570)

Lnrd 0153lowastlowastlowast(470)

0143lowastlowastlowast(408)

Lnpgdp 0169lowastlowastlowast(462)

0120lowastlowastlowast(308)

0138lowastlowastlowast(360)

0099lowastlowast(244)

Lnurban minus0050lowastlowast(minus237)

minus0044lowastlowast(minus209)

minus0050lowastlowast(minus237)

minus0045lowastlowast(minus215)

Lnitu 0433lowastlowast(227)

0438lowastlowast(215)

0525lowastlowastlowast(273)

0531lowastlowastlowast(259)

Lninternet minus0295lowastlowastlowast(minus2970)

minus0299lowastlowastlowast(minus2716)

minus0292lowastlowastlowast(minus2945)

minus0297lowastlowastlowast(minus2690)

Lnifa minus0446lowastlowastlowast(minus2298)

minus0386lowastlowastlowast(minus1832)

minus0437lowastlowastlowast(minus2321)

minus0374lowastlowastlowast(minus1806)

Lnprivate 0038lowastlowastlowast(322)

0037lowastlowastlowast(290)

0039lowastlowastlowast(325)

0038lowastlowastlowast(302)

WLnsd 0179(099)

minus0041(minus022)

WLnrd minus0228lowastlowast(minus251)

minus0186lowast(minus179)

WLnpgdp 0016(027)

0064(101)

0058(096)

0097(150)

WLnurban 0097lowastlowast(236)

0082lowastlowast(198)

0107lowastlowastlowast(259)

0089lowastlowast(213)

WLnitu 1159lowast(193)

0516(082)

0615(100)

0228(035)

WLninternet 0177lowastlowastlowast(647)

0181lowastlowastlowast(622)

0180lowastlowastlowast(666)

0180lowastlowastlowast(619)

WLnifa 0237lowastlowastlowast(539)

0251lowastlowastlowast(532)

0275lowastlowastlowast(704)

0260lowastlowastlowast(600)

WLnprivate 0065lowast(185)

0045(115)

0083lowastlowast(232)

0060(153)

120588

0420lowastlowastlowast(1165)

0462lowastlowastlowast(1251)

0429lowastlowastlowast(1205)

0463lowastlowastlowast(1266)

1198772 0448 0443 0441 0438

log 119871 599812 595228 595228 587624Observations 2810 2529 2810 2529Note the values in parentheses represent the 119905-statistics values of thecorresponding estimated coefficientslowastlowastlowastlowastlowast andlowast respectively representsignificance level of 1 5 and 10

local governmentsrsquo motivation of pursuing revenues underthe fiscal revenue decentralization system is not conduciveto free flowing of financial resources among regions whichdirectly affects the financial efficiency in other regions

When it comes to the estimated coefficients of othercontrolled variables we found that not all the spatial spillovereffects of variables are significant Overall various economicactivities have direct and indirect influence on financial

efficiency and promote the strategic financial competitionamong regions The specific effects are as follows the levelof open-into-inside can not only improve the local financialefficiency significantly but also increase the financial effi-ciency correspondingly in neighboring regions The strongdemand for various financial products and services of privateenterprises promotes financial innovation and improves theefficiency of resource allocation in financial markets Onthe contrary urbanization information technology develop-ment and fixed assets investment suppress the improvementof local financial efficiency but play a positive role in promot-ing the financial efficiency in neighboring regionsThe accel-erating of urbanization process the application of internettechnology and investment in fixed assets promote economicexchanges and cooperation among regions so that the spatialspillover effects of financial development and innovation areincreasingly significant Furthermore although the regionaleconomic development and industrial structure significantlycontribute to the improvement of regional financial efficiencyin general the spillover effects among regions are not obvious

52 Decomposition of Fiscal Decentralizationrsquos Effects onRegional Financial Efficiency Whether there truly existspillover effects among the variables relying solely on the spa-tial lag coefficients of SDMmodel and DSDMmodel is likelyto lead to misinterpretation of the estimated results Spatialeffects can be further decomposed into direct effect indirecteffect (spillover effects) and total effect with reference to theresearch result of LeSage and Pace [32] The indirect effectsrepresent the independent variablesrsquo influence on the depen-dent variable in all other regions through spatial interactionsTherefore we will further verify the existence of spillovereffects by using partial differential matrix analysis Table 4illustrates the direct effect indirect effect and total effect ofthe fiscal expenditure and revenue decentralization in SDMmodel The results show that the estimated coefficients ofindependent variablesrsquo direct effect have the same directionwith the estimated coefficients of SDMmodel in Table 3 andthe estimated coefficients and significance levels are also veryclose

According to the estimated results of SDM model inTable 4 the direct effect indirect effect and total effectof fiscal expenditure decentralization on regional financialefficiency are positive and pass the 10 significance leveltest The regression coefficient of direct effect is 0368 theregression coefficient of indirect effect is 0560 and theregression coefficient of total effect is 0928 indicating thatthe degree of fiscal expenditure decentralization increases by1 while regional financial efficiency increases by 0928Fiscal expenditure decentralization plays a significant posi-tive impact on financial development and financial efficiencywhich is opposite to the research result of Wang and Li [23]The opposite conclusions are possibly due to the differencesof sample selection and indexesThe rise of fiscal expendituredecentralization in the local region leads to an increaseby 0368 of the financial efficiency while the increase ofexpenditure decentralization in the neighboring region leadsto financial efficiency increasing by 0560 indicating that

10 Discrete Dynamics in Nature and Society

Table 4 Decomposition of direct effect indirect effect and total effect under SDMmodel

Independent variablesDependent variable sd Dependent variable sd

Direct effect Indirect effect Total effect Direct effect Indirect effect Total effect

Lnsd 0368lowastlowastlowast(750)

0560lowast(195)

0928lowastlowastlowast(310)

Lnrd 0146lowastlowastlowast(529)

minus0283lowast(minus195)

minus0137lowast(minus192)

Lnpgdp 0175lowastlowastlowast(435)

0132(153)

0307lowastlowastlowast(359)

0145lowastlowastlowast(346)

0187lowastlowast(218)

0332lowastlowastlowast(392)

Lnurban minus0046lowastlowast(minus207)

0127lowastlowast(200)

0081(130)

minus0045lowastlowast(minus204)

0146lowastlowast(226)

0101(158)

Lnitu 0486lowastlowastlowast(275)

2321lowastlowast(237)

2807lowastlowastlowast(292)

0559lowastlowastlowast(315)

1488(151)

2047lowastlowast(211)

Lninternet minus0291lowastlowastlowast(minus3055)

0086lowast(190)

minus0205lowastlowastlowast(minus432)

minus0288lowastlowastlowast(minus3029)

0091lowastlowast(200)

minus0197lowastlowastlowast(minus410)

Lnifa minus0441lowastlowastlowast(minus2260)

0087(149)

minus0353lowastlowastlowast(minus594)

minus0430lowastlowastlowast(minus2254)

0154lowastlowastlowast(283)

minus0276lowastlowastlowast(minus494)

Lnprivate 0041lowastlowastlowast(319)

0134lowastlowast(235)

0175lowastlowastlowast(285)

0042lowastlowastlowast(327)

0166lowastlowastlowast(283)

0208lowastlowastlowast(330)

Note lowast lowast lowast lowastlowast and lowast respectively represent the significant level of 1 5 and 10

fiscal expenditure decentralization significantly promotesthe competition of financial development and innovationamong regions The expansion of the autonomy of localgovernment expenditure strengthens the regulatory capacityof local governments which is conducive to acceleratingmarket-oriented reform processing and achieves the organicintegration between fiscal system and market mechanism inthe allocation of financial resources which is beneficial tocreating a favorable institutional environment is also con-ducive to the flowing of financial resources among regionsand greatly promotes financial efficiency

The fiscal revenue decentralizationrsquos direct effect onregional financial efficiency is significantly positive whilethe spillover effect is significantly negative indicating thatthe impact of fiscal revenue decentralization on regionalfinancial efficiency presents the feature of ldquodouble-edged-swordrdquo Fiscal revenue decentralization has a significant posi-tive impact on the local financial efficiencyHowever there aresignificant negative spillover effects on the financial efficiencyin other regions The regression coefficient of the directeffect is 0146 the regression coefficient of the indirect effectis minus0283 and the regression coefficient of the total effectis minus0137 which indicates that the degree of fiscal revenuedecentralization increasing by 1 will lead to decrease by0137 of regional financial efficiency Furthermore whilethe increase of local fiscal revenue decentralization leads tofinancial efficiencyrsquos increase by 0146 the increase of thefiscal revenue decentralization in neighboring regions resultsin the financial efficiencyrsquos decrease by 0283 Comparingthe estimated coefficient of direct effect and that of indirecteffect it can be found that the negative spillover effectsof fiscal revenue decentralization are significantly greaterthan the direct scale effect so the negative externalitiesof fiscal revenue decentralization on financial efficiencyare greater than the positive externality On one hand inorder to obtain high-quality financial resources the local

governments ignore the differences like the level of economicdevelopment factor endowments and other aspects amongregions and seek to imitate policies and instruments inother regions thus leading to chain reaction of absorbinginvestment by other regions On the other hand whenthe neighboring regions or the regions with similar levelof economic development take some measures to attractliquidity factors the local government driven by yardstickcompetitionwill adopt similar policies to avoid the outflow ofelements which results in convergence of competitive tactics[40 41] Tax incentives reduction of fee and other wayswill become important means of competition and imitationbetween local governments in different regions on the basisof the independent revenue authority However such kind ofcompetitive mode between local governments for resourceswill consume huge fiscal resources and weaken the capacityof local governmentrsquos provision of public goods and servicesUltimately it is harmful to provide healthy stable andsustainable economic development environment for financialinnovation and increase of financial efficiency

53 Robustness Test To ensure the robustness and reliabilityof research results we will calculate the spatial lags withthe geographical distance weight matrix and do econometricregression by taking financial efficiency as the dependentvariable so that we can comparatively analyze the differenceof influence with different spatial weight setting based on theabove The robustness test results are as in Table 5

As shown in Table 5 the estimated results indicate thatthe sign of regression coefficients of the fiscal expendituredecentralization and revenue decentralization is consistentwith that of corresponding variable in the model of Table 3and the regression results in themodel are basically consistentwith the conclusions in Table 3 Comparing the estimatedresults based on two different weightmatrixes it is found thatthere are some differences in the influence coefficients of two

Discrete Dynamics in Nature and Society 11

Table 5 Robustness test on Spatial Durbin Models (geographicaldistance weight matrix)

Independentvariables

Dependent variable sd Dependent variable rdSDMmodel

DSDMmodel

SDMmodel

DSDMmodel

feitminus1 mdash 0202lowastlowastlowast(1276) mdash 0200lowastlowastlowast

(1255)

Lnsd or Lnrd 0364lowastlowastlowast(636)

0348lowastlowastlowast(592)

0173lowastlowastlowast(522)

0161lowastlowastlowast(458)

Lnpgdp 0268lowastlowastlowast(735)

0225lowastlowastlowast(578)

0242lowastlowastlowast(637)

0209lowastlowastlowast(522)

Lnurban minus0026(minus112)

minus0026(minus112)

minus0028(minus120)

minus0027(minus115)

Lnitu 0465lowastlowast(246)

0478lowastlowast(238)

0527lowastlowastlowast(278)

0542lowastlowastlowast(269)

Lninternet minus0295lowastlowastlowast(minus3038)

minus0300lowastlowastlowast(minus2799)

minus0292lowastlowastlowast(minus3005)

minus0298lowastlowastlowast(minus2771)

Lnifa minus0484lowastlowastlowast(minus2479)

minus0427lowastlowastlowast(minus2024)

minus0472lowastlowastlowast(minus2448)

minus0417lowastlowastlowast(minus1988)

Lnprivate 0036lowastlowastlowast(312)

0035lowastlowastlowast(284)

0035lowastlowastlowast(307)

0035lowastlowastlowast(288)

WLnsd orWLnrd

minus0264(minus149)

minus0418lowastlowast(minus225)

minus0277lowastlowastlowast(minus370)

minus0241lowastlowastlowast(minus276)

WLnpgdp minus0269lowastlowastlowast(minus427)

minus0286lowastlowastlowast(minus428)

minus0237lowastlowastlowast(minus369)

minus0267lowastlowastlowast(minus394)

WLnurban 0053(147)

0059lowast(167)

0059(163)

0062lowast(173)

WLnitu 0104(020)

minus0085(minus016)

minus0205(minus039)

minus0290(minus051)

WLninternet 0217lowastlowastlowast(838)

0239lowastlowastlowast(853)

0214lowastlowastlowast(831)

0235lowastlowastlowast(841)

WLnifa 0434lowastlowastlowast(1052)

0464lowastlowastlowast(1054)

0433lowastlowastlowast(1165)

0450lowastlowastlowast(1129)

WLnprivate 0062lowast(187)

0038(106)

0081lowastlowast(241)

0050(139)

120588

0547lowastlowastlowast(1858)

0580lowastlowastlowast(1938)

0541lowastlowastlowast(1839)

0570lowastlowastlowast(1898)

1198772 0444 0428 0443 0430

log 119871 659741 665804 665804 659509Observations 2810 2529 2810 2529Note lowastlowastlowast lowastlowast and lowast respectively represent the significant level of 1 5and 10

kinds of fiscal decentralizations indicators which show thatthe gap between the levels of economic developmentwill havean impact on role of fiscal decentralization acting on regionalfinancial efficiency On one hand under the fiscal decen-tralization system because of the heterogeneity of regionaleconomic development level the local governmentsrsquo ability ofmaking up the gap between fiscal revenue and expenditure isdifferent and the fiscal pressure is also different therefore thedegree of interference to the regional market environment bythe local government is different which leads to the regionaldifferences in financial efficiency in a certain degree [21]On the other hand the occurrence of financial innovationdepends on the support of the economic foundation and

the different economic and social environment will lead tothe fact that the financial innovation activities are more andmore concentrated in the regions with the advantages offinancial location resulting in nonequilibrium of financialefficiency in different regions [42] Under the performanceassessment method focusing on GDP local governments willnot only consider the strategic behaviors of the neighboringregions but also tend to think about the behaviors of theregions with the similar level of economy In additioncross-regional knowledge diffusion technology spillover andhuman capital flow are more prone to occur among theregions with similar level of economic development so thatthe financial exchanges and contacts among regions are morefrequent It also shows that besides considering the impact ofgeographical distance the economic distance factor shouldbe taken into consideration while we are researching thespatial effects of fiscal decentralization on regional financialinnovation in addition to the consideration of the impact ofgeographical distance the economic distance factor shouldbe considered

6 Conclusions and Enlightenments

Based on the panel data of 281 prefecture-level cities in Chinacovering the period of 2003ndash2012 the regional financialefficiency is measured by the superefficiency SBMmodel thegeographic distance weight matrix and economic-geographymixed weight matrix which can reflect the regional similarityare introduced into the SDM and the spatial effects of thefiscal expenditure and revenue decentralization on regionalfinancial efficiency under province are investigated as wellas the strategic financial competition arising from it Basedon the results of previous empirical analysis we can draw thefollowing basic conclusions

First the geographical distribution of Chinarsquos prefecture-level citiesrsquo financial efficiency presents significant features oftime inertia and spatial dependency Financial developmentand innovation among regions particularly neighboringregions is not independent but shows significant spatialspillover effects The financial efficiency of the local regionwill be significantly affected by the financial efficiency ofadjacent regions The spatial effect is increasingly obviousin the process of financial development and innovation andthe size of the spillover effects is closely related to spatialweight (which is used to measure regional similarities) andmainly shows that there is competition for financial efficiencyamong regions with similar economic attributes and adjacentregions

Second it cannot be ignored that the institutional factorsespecially fiscal decentralization have effect on the financialefficiency and the factors have a decisive role in the allocationof financial resource and the regional distribution of financialinnovation The improvement of fiscal decentralization levelwill help to attract more financial resources thereby promot-ing an increase in financial efficiencyThe improvement of thefinancial autonomy in neighboring jurisdictions will reducethe financial resources flowing into the local region andsuppresses the occurrence of the local financial innovationIn addition there exist significant spatial spillover effects of

12 Discrete Dynamics in Nature and Society

financial development and efficiency caused by fiscal decen-tralization The scope of spatial spillovers is mainly reflectedin two aspects Thus the local governmentsrsquo competition notonly strongly increases the financial efficiency of the regionbut also accelerates interregional competition for financialefficiency

Third fiscal expenditure decentralization and revenuedecentralizationrsquos impact on regional financial efficiencyhave nonequilibrium characteristics First of all there existsasymmetry between fiscal expenditure decentralization andrevenue decentralizationrsquos impact on regional financial effi-ciency Fiscal expenditure decentralization can promoteoptimization of the fiscal expenditure structure and fiscalrevenue decentralization can increase the local governmentsrsquorevenue share so as to provide positive incentives [43] bothof them have a positive and promoting effect on the improve-ment of local financial efficiency and such effect largelydepends on the symmetry between the fiscal expenditure andrevenue decentralization If it is asymmetric between theirproperty rights and the expenditure responsibilities underfiscal decentralization system the local governments maymake excessive intervention to regional financial resourcesallocation tomeet their own demands for financial resourcesNext there exists nonconformance between fiscal expendi-ture decentralization and revenue decentralizationrsquos impacton financial efficiency in neighboring regions Comparedwith the fiscal expenditure decentralization the fiscal revenuedecentralization has more apparent inhibitory effect on thefinancial efficiency in neighboring regions

Our research conclusions have important policy implica-tions that illuminate the relationship between fiscal decen-tralization and financial efficiency

(1) The regional financial efficiency is closely relatedto the location and the economic development of the sur-rounding cities The regional financial efficiency is affectednot only by the financial development and innovation inneighboring regions but also by the impact of the struc-tural differences among regions In this case breaking theadministrativemonopoly of local protectionism and carryingon the ldquorichening neighborrdquo policy of financial developmentare the best choices for the local government to developthe regional economy When making policies of financialdevelopment local governments need to consider their owneconomic development environment and economic featuresin neighboring regions and strengthen communication andcoordination with the surrounding regions so as to avoidthe possible contradictions and conflicts on financial devel-opment policies among regions

(2) The spatial spillover effects of the regional financialefficiency should be divided into two kinds of effects includ-ing global geographic spillover effects and local geographicspillover effects In order to promote the increase of financialefficiency in a better way it should also be based on thehierarchy of spatial effect of and the degree of spillover effectto define the relationship between the government and themarket and central and local governments and among thelocal governments and their role in regional financial devel-opment and innovation so as to better promote financialefficiency

(3) There is a two-way interactive mechanism betweenthe fiscal system and financial development such effectsmay affect the behaviors of the local governments and theallocation of resources and ultimately act on macroeconomicperformance [10] Under the fiscal decentralization systemthe strategic competitive behaviors among local governmentssignificantly affect the liquidity of financial resources and theoccurrence of financial innovation the local governmentsrsquoexpansion of fiscal expenditure and revenue shortfall becomethe main driving force of each region competing for financialresources In order to better play the role of incentivesand constraints of fiscal competition mechanism under thefiscal decentralization system and avoid the negative impactbrought by the vicious interregional strategic financial com-petitive behaviors on the regional economic development thelocal governmentsrsquo short-sighted behaviors should be strictlyregulated and then the interregional financial cooperationwith mutual benefit as the core should be actively promoted

(4) Constructing a scientific and rational framework forfiscal decentralization will be an important force to accelerateeffectively the improvement of regional financial efficiencyChinarsquos future subprovincial fiscal decentralization reformshould adapt to the new normal financial development andfurther rationalize the provincial and subprovincial fiscaldecentralization system to ensure a balanced and appropriatefiscal expenditure and revenue decentralization On onehand we should weaken local governmentsrsquo negative inter-vention in financial system by reducing local governmentsrsquofinancial pressure caused by the asymmetry between thefiscal resources and expenditure responsibilities On the otherhandwe should also actively play the function of fiscal systemto adjust the financial development and innovation and fullyplay public financersquos role in promoting the flowing of finan-cial resources among regions the balanced development ofregional finance and the improvement of financial efficiency

Competing Interests

There are no competing interests in regard to the contentdiscussed in this paper

Authorsrsquo Contributions

All authors conceived and designed the research XiaomeiHucollected and processed the data Jianmin Liu analyzed thedata and interpreted the results all authors were also involvedin writing the paper and have all approved the submittedform

Acknowledgments

The research of this paper is jointly funded by the NationalSocial Science Foundation of China under Project no14BJY159 and the Hunan Philosophy and Social ScienceFund Project under Grant no 13JD09 The authors gratefullyacknowledge the support of Jinguang Wu from HunanUniversity of Finance and Economics China

Discrete Dynamics in Nature and Society 13

References

[1] J Zhang and X Jin ldquoRe-examine of the relationship betweenChinarsquos financial deepening and productivity 1987ndash2001rdquo Eco-nomic Research Journal vol 11 pp 35ndash43 2005

[2] K Chen A L Hillman and Q Y Gu ldquoFiscal centralization andchanges of local governmentsrsquo behaviorsmdashfrom lsquohands for helprsquoto lsquohands for grabrsquordquoChina EconomicQuarterly vol 1 pp 111ndash1302002

[3] L Zhou and Z M Wang ldquoEmpirical analysis of regionalfinancial development of China and economic growth 1978ndash2000rdquo Financial Research vol 10 pp 1ndash13 2002

[4] L Zhou ldquolsquoSecond financersquo of Chinarsquos financial industry andfinancial division during the reforming periodrdquo The Journal ofWorld Economy vol 6 pp 72ndash79 2003

[5] L Zhou ldquoGradual transition state capacity and fiscal functionof financerdquo Journal of Finance and Economics vol 2 pp 26ndash372005

[6] E L He and W Liu ldquoFinancial capital local governmentsrsquointervention and industrial structuremdashan empirical analysisbased on the data of central regionsrdquo Economic Issues vol 5pp 37ndash40 2011

[7] X L Dou and W J Wang ldquoThe differences and trends of Chi-nese regional financial development and local governmentsrsquobehavior and analysis of tax sharing reform on the differencesand trends of Chinese regional financial developmentrdquo Shang-hai Finance vol 2 pp 9ndash15 2011

[8] J Zhang and K R Shen ldquoLocal government interventionregional financial development and the transformation ofChinarsquos economic growth modemdashan empirical study under thefiscal decentralization systemrdquo Nankai Economic Studies vol 6pp 122ndash141 2008

[9] G Chen X G Yin and Y Pan ldquoChinarsquos financial developmenttax sharing reform and economic growthrdquo Financial Researchvol 2 pp 99ndash109 2006

[10] P P Ding and Y Fu ldquoBehaviors of local governments therelationship of fiscal and finance and macroeconomic volatilityassociated with Chinamdashan analysis under Chinese-style fiscaldecentralization systemrdquo Comparative Economic amp Social Sys-tems vol 6 pp 87ndash97 2012

[11] X G Yin G Chen and Y Pan ldquoTax sharing reform local gov-ernmentsrsquo intervention and financial development efficiencyrdquoJournal of Finance and Economics vol 10 pp 92ndash101 2006

[12] M G Yu and H B Pan ldquoGovernment intervention the ruleof law financial development and bank loans of state-ownedenterprisesrdquo Financial Research vol 9 pp 1ndash23 2008

[13] D X Wang J Liu and L Li ldquoFiscal decentralization bankcredit and total factor productivityrdquo Journal of Finance andEconomics vol 4 pp 69ndash80 2011

[14] Y X Jiang and R B Jiang ldquoRegional financial innovationefficiency evaluation environmental impact and gap analysisrdquoJournal of Zhejiang University (Humanities and Social Sciences)vol 43 no 4 pp 52ndash65 2013

[15] H W Yang L W Cheng Y Xu and Y W Qi ldquoMeasure andinfluencing factors of financial resources allocation efficiencyunder fiscal decentralization system based on super-efficiencyDEA-TOBIT two-step methodrdquo Journal of Jiangxi NormalUniversity (Natural Sciences Edition) vol 6 pp 587ndash592 2014

[16] N Shen F Z Liu and J Q Zhao ldquoFiscal decentralizationfinancial deepening and regional international poverty prone todeveloprdquo Finance amp Trade Economics vol 1 pp 41ndash46 2006

[17] Q Y Du G P Li and N Shen ldquoFinancial efficiency andoptimization of the regional industrial structure under thecurrent fiscal systemrdquo Xinjiang Social Science vol 4 pp 16ndash212006

[18] T Feng Y W Song and Y Lu ldquoFiscal decentralization localgovernmentsrsquo behavior and regional financial developmentrdquoJournal of Xirsquoan Jiaotong University vol 5 pp 20ndash27 2007

[19] Y A Zhou X J Ma and J Y Zhao ldquoGovernment behaviorsfinancial development and economic growthrdquo Henan SocialSciences vol 1 pp 55ndash60 2007

[20] C Hao ldquoDevelopment of financial intermediation and eco-nomic growth the Chinese experiencerdquo China EconomicReview vol 17 no 4 pp 347ndash362 2006

[21] J Zhang Fiscal Decentralization Regional Financial Devel-opment and Chinarsquos Economic Growth Performance NanjingUniversity Press Nanjing China 2011

[22] Y J Yao ldquoGovernment intervention the development of finan-cial intermediation and economic growthrdquo Journal of Financeand Economics vol 8 pp 52ndash58 2010

[23] X Wang and L Li ldquoFiscal decentralization and regional finan-cial development the theoretical and empirical analysis fromthe perspective of central governmentrdquo Public Finance Researchvol 4 pp 31ndash35 2012

[24] W L Zhang and C L Qin ldquoFinancial development fiscaldecentralization and regional economic differencesrdquo EconomicGeography vol 1 pp 583ndash588 2009

[25] Y Ma J Li and B Chen ldquoProvincial differences in Chinesefiscal decentralization financial development industrializationand economic growthrdquo EconomicTheory and Business Manage-ment vol 2 pp 5ndash19 2015

[26] Y X Jiang and R B Jiang ldquoThe spatial external effects ofregional financial innovation financial competition and finan-cial agglomerationrdquo Social Science Front vol 3 pp 79ndash84 2014

[27] K Tone ldquoA slacks-based measure of super-efficiency indata envelopment analysisrdquo European Journal of OperationalResearch vol 143 no 1 pp 32ndash41 2002

[28] Y Q Lu and D G Zhang ldquoChinarsquos regional financial efficiencyand differences of efficiency measurementrdquo Economic Geogra-phy vol 1 pp 96ndash101 2012

[29] A N Berger and D B Humphrey ldquoEfficiency of financialinstitutions international survey and directions for futureresearchrdquo European Journal of Operational Research vol 98 no2 pp 175ndash212 1997

[30] Y J Yao ldquoEmpirical analysis of financial development andurban-rural income gaprdquo Journal of Finance and Economics vol2 pp 49ndash59 2005

[31] Q Li J Song E Wang H Hu J Zhang and Y WangldquoEconomic growth and pollutant emissions in China a spatialeconometric analysisrdquo Stochastic Environmental Research andRisk Assessment vol 28 no 2 pp 429ndash442 2014

[32] J P LeSage and R K Pace Introduction to Spatial EconometricsCRCPress Taylor amp Francis Group Boca Raton Fla USA 2009

[33] Q W Guo and J X Jia ldquoFiscal decentralization organizationalstructure of local governments and the size of governmentexpenditurerdquoEconomic Research Journal vol 11 pp 59ndash87 2010httpwwwcnkicomcnArticleCJFDTotal-JJYJ201011008htm

[34] F Gong and X Lei ldquoThe measurement of Chinese-style fiscaldecentralizationrdquo Statistical Research vol 10 pp 47ndash55 2010

[35] G Fan X L Wang and H P Zhu Chinese Marketing IndexRegional Marketing Process in 2011 Economic Science PressBeijing China 2011

14 Discrete Dynamics in Nature and Society

[36] S S Rosenthal and W C Strange ldquoGeography industrialorganization and agglomerationrdquoTheReview of Economics andStatistics vol 85 no 2 pp 377ndash393 2003

[37] S Park Yoon International Banking and Financial CentersKluwer Academic Boston Mass USA 1989

[38] B Y Qiao J Y Fan and X Y Feng ldquoChinarsquos fiscal decentral-ization and compulsory primary educationrdquo Social Sciences inChina vol 6 pp 37ndash46 2005

[39] W JWang and C L Qin ldquoRegional differences in the behaviorsof local governments and growth effect of fiscal decentralizationbased on the analysis of judgment hypothesis and testingrdquoManagement World vol 1 pp 9ndash21 2008

[40] J K Brueckner ldquoStrategic interaction among governments anoverview of empirical studiesrdquo International Regional ScienceReview vol 26 no 2 pp 175ndash188 2003

[41] F Revelli ldquoOn spatial public finance empiricsrdquo International Taxand Public Finance vol 12 no 4 pp 475ndash492 2005

[42] N Shen and T Y He ldquoFinancial efficiency and regional eco-nomic growth from the perspective of fiscal decentralizationrdquoReform vol 1 pp 61ndash65 2006

[43] J M Liu X M Hu and J G Wu ldquoThe threshold effect ofsub-national fiscal decentralization revenue and expendituredecentralization on provincial industrial transformation andupgrading based on the data of 14 cities in Hunan ProvincerdquoPublic Finance Research vol 8 pp 49ndash52 2014

Submit your manuscripts athttpwwwhindawicom

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

MathematicsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Mathematical Problems in Engineering

Hindawi Publishing Corporationhttpwwwhindawicom

Differential EquationsInternational Journal of

Volume 2014

Applied MathematicsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Probability and StatisticsHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Mathematical PhysicsAdvances in

Complex AnalysisJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

OptimizationJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

CombinatoricsHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Operations ResearchAdvances in

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Function Spaces

Abstract and Applied AnalysisHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of Mathematics and Mathematical Sciences

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Algebra

Discrete Dynamics in Nature and Society

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Decision SciencesAdvances in

Discrete MathematicsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom

Volume 2014 Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Stochastic AnalysisInternational Journal of

10 Discrete Dynamics in Nature and Society

Table 4 Decomposition of direct effect indirect effect and total effect under SDMmodel

Independent variablesDependent variable sd Dependent variable sd

Direct effect Indirect effect Total effect Direct effect Indirect effect Total effect

Lnsd 0368lowastlowastlowast(750)

0560lowast(195)

0928lowastlowastlowast(310)

Lnrd 0146lowastlowastlowast(529)

minus0283lowast(minus195)

minus0137lowast(minus192)

Lnpgdp 0175lowastlowastlowast(435)

0132(153)

0307lowastlowastlowast(359)

0145lowastlowastlowast(346)

0187lowastlowast(218)

0332lowastlowastlowast(392)

Lnurban minus0046lowastlowast(minus207)

0127lowastlowast(200)

0081(130)

minus0045lowastlowast(minus204)

0146lowastlowast(226)

0101(158)

Lnitu 0486lowastlowastlowast(275)

2321lowastlowast(237)

2807lowastlowastlowast(292)

0559lowastlowastlowast(315)

1488(151)

2047lowastlowast(211)

Lninternet minus0291lowastlowastlowast(minus3055)

0086lowast(190)

minus0205lowastlowastlowast(minus432)

minus0288lowastlowastlowast(minus3029)

0091lowastlowast(200)

minus0197lowastlowastlowast(minus410)

Lnifa minus0441lowastlowastlowast(minus2260)

0087(149)

minus0353lowastlowastlowast(minus594)

minus0430lowastlowastlowast(minus2254)

0154lowastlowastlowast(283)

minus0276lowastlowastlowast(minus494)

Lnprivate 0041lowastlowastlowast(319)

0134lowastlowast(235)

0175lowastlowastlowast(285)

0042lowastlowastlowast(327)

0166lowastlowastlowast(283)

0208lowastlowastlowast(330)

Note lowast lowast lowast lowastlowast and lowast respectively represent the significant level of 1 5 and 10

fiscal expenditure decentralization significantly promotesthe competition of financial development and innovationamong regions The expansion of the autonomy of localgovernment expenditure strengthens the regulatory capacityof local governments which is conducive to acceleratingmarket-oriented reform processing and achieves the organicintegration between fiscal system and market mechanism inthe allocation of financial resources which is beneficial tocreating a favorable institutional environment is also con-ducive to the flowing of financial resources among regionsand greatly promotes financial efficiency

The fiscal revenue decentralizationrsquos direct effect onregional financial efficiency is significantly positive whilethe spillover effect is significantly negative indicating thatthe impact of fiscal revenue decentralization on regionalfinancial efficiency presents the feature of ldquodouble-edged-swordrdquo Fiscal revenue decentralization has a significant posi-tive impact on the local financial efficiencyHowever there aresignificant negative spillover effects on the financial efficiencyin other regions The regression coefficient of the directeffect is 0146 the regression coefficient of the indirect effectis minus0283 and the regression coefficient of the total effectis minus0137 which indicates that the degree of fiscal revenuedecentralization increasing by 1 will lead to decrease by0137 of regional financial efficiency Furthermore whilethe increase of local fiscal revenue decentralization leads tofinancial efficiencyrsquos increase by 0146 the increase of thefiscal revenue decentralization in neighboring regions resultsin the financial efficiencyrsquos decrease by 0283 Comparingthe estimated coefficient of direct effect and that of indirecteffect it can be found that the negative spillover effectsof fiscal revenue decentralization are significantly greaterthan the direct scale effect so the negative externalitiesof fiscal revenue decentralization on financial efficiencyare greater than the positive externality On one hand inorder to obtain high-quality financial resources the local

governments ignore the differences like the level of economicdevelopment factor endowments and other aspects amongregions and seek to imitate policies and instruments inother regions thus leading to chain reaction of absorbinginvestment by other regions On the other hand whenthe neighboring regions or the regions with similar levelof economic development take some measures to attractliquidity factors the local government driven by yardstickcompetitionwill adopt similar policies to avoid the outflow ofelements which results in convergence of competitive tactics[40 41] Tax incentives reduction of fee and other wayswill become important means of competition and imitationbetween local governments in different regions on the basisof the independent revenue authority However such kind ofcompetitive mode between local governments for resourceswill consume huge fiscal resources and weaken the capacityof local governmentrsquos provision of public goods and servicesUltimately it is harmful to provide healthy stable andsustainable economic development environment for financialinnovation and increase of financial efficiency

53 Robustness Test To ensure the robustness and reliabilityof research results we will calculate the spatial lags withthe geographical distance weight matrix and do econometricregression by taking financial efficiency as the dependentvariable so that we can comparatively analyze the differenceof influence with different spatial weight setting based on theabove The robustness test results are as in Table 5

As shown in Table 5 the estimated results indicate thatthe sign of regression coefficients of the fiscal expendituredecentralization and revenue decentralization is consistentwith that of corresponding variable in the model of Table 3and the regression results in themodel are basically consistentwith the conclusions in Table 3 Comparing the estimatedresults based on two different weightmatrixes it is found thatthere are some differences in the influence coefficients of two

Discrete Dynamics in Nature and Society 11

Table 5 Robustness test on Spatial Durbin Models (geographicaldistance weight matrix)

Independentvariables

Dependent variable sd Dependent variable rdSDMmodel

DSDMmodel

SDMmodel

DSDMmodel

feitminus1 mdash 0202lowastlowastlowast(1276) mdash 0200lowastlowastlowast

(1255)

Lnsd or Lnrd 0364lowastlowastlowast(636)

0348lowastlowastlowast(592)

0173lowastlowastlowast(522)

0161lowastlowastlowast(458)

Lnpgdp 0268lowastlowastlowast(735)

0225lowastlowastlowast(578)

0242lowastlowastlowast(637)

0209lowastlowastlowast(522)

Lnurban minus0026(minus112)

minus0026(minus112)

minus0028(minus120)

minus0027(minus115)

Lnitu 0465lowastlowast(246)

0478lowastlowast(238)

0527lowastlowastlowast(278)

0542lowastlowastlowast(269)

Lninternet minus0295lowastlowastlowast(minus3038)

minus0300lowastlowastlowast(minus2799)

minus0292lowastlowastlowast(minus3005)

minus0298lowastlowastlowast(minus2771)

Lnifa minus0484lowastlowastlowast(minus2479)

minus0427lowastlowastlowast(minus2024)

minus0472lowastlowastlowast(minus2448)

minus0417lowastlowastlowast(minus1988)

Lnprivate 0036lowastlowastlowast(312)

0035lowastlowastlowast(284)

0035lowastlowastlowast(307)

0035lowastlowastlowast(288)

WLnsd orWLnrd

minus0264(minus149)

minus0418lowastlowast(minus225)

minus0277lowastlowastlowast(minus370)

minus0241lowastlowastlowast(minus276)

WLnpgdp minus0269lowastlowastlowast(minus427)

minus0286lowastlowastlowast(minus428)

minus0237lowastlowastlowast(minus369)

minus0267lowastlowastlowast(minus394)

WLnurban 0053(147)

0059lowast(167)

0059(163)

0062lowast(173)

WLnitu 0104(020)

minus0085(minus016)

minus0205(minus039)

minus0290(minus051)

WLninternet 0217lowastlowastlowast(838)

0239lowastlowastlowast(853)

0214lowastlowastlowast(831)

0235lowastlowastlowast(841)

WLnifa 0434lowastlowastlowast(1052)

0464lowastlowastlowast(1054)

0433lowastlowastlowast(1165)

0450lowastlowastlowast(1129)

WLnprivate 0062lowast(187)

0038(106)

0081lowastlowast(241)

0050(139)

120588

0547lowastlowastlowast(1858)

0580lowastlowastlowast(1938)

0541lowastlowastlowast(1839)

0570lowastlowastlowast(1898)

1198772 0444 0428 0443 0430

log 119871 659741 665804 665804 659509Observations 2810 2529 2810 2529Note lowastlowastlowast lowastlowast and lowast respectively represent the significant level of 1 5and 10

kinds of fiscal decentralizations indicators which show thatthe gap between the levels of economic developmentwill havean impact on role of fiscal decentralization acting on regionalfinancial efficiency On one hand under the fiscal decen-tralization system because of the heterogeneity of regionaleconomic development level the local governmentsrsquo ability ofmaking up the gap between fiscal revenue and expenditure isdifferent and the fiscal pressure is also different therefore thedegree of interference to the regional market environment bythe local government is different which leads to the regionaldifferences in financial efficiency in a certain degree [21]On the other hand the occurrence of financial innovationdepends on the support of the economic foundation and

the different economic and social environment will lead tothe fact that the financial innovation activities are more andmore concentrated in the regions with the advantages offinancial location resulting in nonequilibrium of financialefficiency in different regions [42] Under the performanceassessment method focusing on GDP local governments willnot only consider the strategic behaviors of the neighboringregions but also tend to think about the behaviors of theregions with the similar level of economy In additioncross-regional knowledge diffusion technology spillover andhuman capital flow are more prone to occur among theregions with similar level of economic development so thatthe financial exchanges and contacts among regions are morefrequent It also shows that besides considering the impact ofgeographical distance the economic distance factor shouldbe taken into consideration while we are researching thespatial effects of fiscal decentralization on regional financialinnovation in addition to the consideration of the impact ofgeographical distance the economic distance factor shouldbe considered

6 Conclusions and Enlightenments

Based on the panel data of 281 prefecture-level cities in Chinacovering the period of 2003ndash2012 the regional financialefficiency is measured by the superefficiency SBMmodel thegeographic distance weight matrix and economic-geographymixed weight matrix which can reflect the regional similarityare introduced into the SDM and the spatial effects of thefiscal expenditure and revenue decentralization on regionalfinancial efficiency under province are investigated as wellas the strategic financial competition arising from it Basedon the results of previous empirical analysis we can draw thefollowing basic conclusions

First the geographical distribution of Chinarsquos prefecture-level citiesrsquo financial efficiency presents significant features oftime inertia and spatial dependency Financial developmentand innovation among regions particularly neighboringregions is not independent but shows significant spatialspillover effects The financial efficiency of the local regionwill be significantly affected by the financial efficiency ofadjacent regions The spatial effect is increasingly obviousin the process of financial development and innovation andthe size of the spillover effects is closely related to spatialweight (which is used to measure regional similarities) andmainly shows that there is competition for financial efficiencyamong regions with similar economic attributes and adjacentregions

Second it cannot be ignored that the institutional factorsespecially fiscal decentralization have effect on the financialefficiency and the factors have a decisive role in the allocationof financial resource and the regional distribution of financialinnovation The improvement of fiscal decentralization levelwill help to attract more financial resources thereby promot-ing an increase in financial efficiencyThe improvement of thefinancial autonomy in neighboring jurisdictions will reducethe financial resources flowing into the local region andsuppresses the occurrence of the local financial innovationIn addition there exist significant spatial spillover effects of

12 Discrete Dynamics in Nature and Society

financial development and efficiency caused by fiscal decen-tralization The scope of spatial spillovers is mainly reflectedin two aspects Thus the local governmentsrsquo competition notonly strongly increases the financial efficiency of the regionbut also accelerates interregional competition for financialefficiency

Third fiscal expenditure decentralization and revenuedecentralizationrsquos impact on regional financial efficiencyhave nonequilibrium characteristics First of all there existsasymmetry between fiscal expenditure decentralization andrevenue decentralizationrsquos impact on regional financial effi-ciency Fiscal expenditure decentralization can promoteoptimization of the fiscal expenditure structure and fiscalrevenue decentralization can increase the local governmentsrsquorevenue share so as to provide positive incentives [43] bothof them have a positive and promoting effect on the improve-ment of local financial efficiency and such effect largelydepends on the symmetry between the fiscal expenditure andrevenue decentralization If it is asymmetric between theirproperty rights and the expenditure responsibilities underfiscal decentralization system the local governments maymake excessive intervention to regional financial resourcesallocation tomeet their own demands for financial resourcesNext there exists nonconformance between fiscal expendi-ture decentralization and revenue decentralizationrsquos impacton financial efficiency in neighboring regions Comparedwith the fiscal expenditure decentralization the fiscal revenuedecentralization has more apparent inhibitory effect on thefinancial efficiency in neighboring regions

Our research conclusions have important policy implica-tions that illuminate the relationship between fiscal decen-tralization and financial efficiency

(1) The regional financial efficiency is closely relatedto the location and the economic development of the sur-rounding cities The regional financial efficiency is affectednot only by the financial development and innovation inneighboring regions but also by the impact of the struc-tural differences among regions In this case breaking theadministrativemonopoly of local protectionism and carryingon the ldquorichening neighborrdquo policy of financial developmentare the best choices for the local government to developthe regional economy When making policies of financialdevelopment local governments need to consider their owneconomic development environment and economic featuresin neighboring regions and strengthen communication andcoordination with the surrounding regions so as to avoidthe possible contradictions and conflicts on financial devel-opment policies among regions

(2) The spatial spillover effects of the regional financialefficiency should be divided into two kinds of effects includ-ing global geographic spillover effects and local geographicspillover effects In order to promote the increase of financialefficiency in a better way it should also be based on thehierarchy of spatial effect of and the degree of spillover effectto define the relationship between the government and themarket and central and local governments and among thelocal governments and their role in regional financial devel-opment and innovation so as to better promote financialefficiency

(3) There is a two-way interactive mechanism betweenthe fiscal system and financial development such effectsmay affect the behaviors of the local governments and theallocation of resources and ultimately act on macroeconomicperformance [10] Under the fiscal decentralization systemthe strategic competitive behaviors among local governmentssignificantly affect the liquidity of financial resources and theoccurrence of financial innovation the local governmentsrsquoexpansion of fiscal expenditure and revenue shortfall becomethe main driving force of each region competing for financialresources In order to better play the role of incentivesand constraints of fiscal competition mechanism under thefiscal decentralization system and avoid the negative impactbrought by the vicious interregional strategic financial com-petitive behaviors on the regional economic development thelocal governmentsrsquo short-sighted behaviors should be strictlyregulated and then the interregional financial cooperationwith mutual benefit as the core should be actively promoted

(4) Constructing a scientific and rational framework forfiscal decentralization will be an important force to accelerateeffectively the improvement of regional financial efficiencyChinarsquos future subprovincial fiscal decentralization reformshould adapt to the new normal financial development andfurther rationalize the provincial and subprovincial fiscaldecentralization system to ensure a balanced and appropriatefiscal expenditure and revenue decentralization On onehand we should weaken local governmentsrsquo negative inter-vention in financial system by reducing local governmentsrsquofinancial pressure caused by the asymmetry between thefiscal resources and expenditure responsibilities On the otherhandwe should also actively play the function of fiscal systemto adjust the financial development and innovation and fullyplay public financersquos role in promoting the flowing of finan-cial resources among regions the balanced development ofregional finance and the improvement of financial efficiency

Competing Interests

There are no competing interests in regard to the contentdiscussed in this paper

Authorsrsquo Contributions

All authors conceived and designed the research XiaomeiHucollected and processed the data Jianmin Liu analyzed thedata and interpreted the results all authors were also involvedin writing the paper and have all approved the submittedform

Acknowledgments

The research of this paper is jointly funded by the NationalSocial Science Foundation of China under Project no14BJY159 and the Hunan Philosophy and Social ScienceFund Project under Grant no 13JD09 The authors gratefullyacknowledge the support of Jinguang Wu from HunanUniversity of Finance and Economics China

Discrete Dynamics in Nature and Society 13

References

[1] J Zhang and X Jin ldquoRe-examine of the relationship betweenChinarsquos financial deepening and productivity 1987ndash2001rdquo Eco-nomic Research Journal vol 11 pp 35ndash43 2005

[2] K Chen A L Hillman and Q Y Gu ldquoFiscal centralization andchanges of local governmentsrsquo behaviorsmdashfrom lsquohands for helprsquoto lsquohands for grabrsquordquoChina EconomicQuarterly vol 1 pp 111ndash1302002

[3] L Zhou and Z M Wang ldquoEmpirical analysis of regionalfinancial development of China and economic growth 1978ndash2000rdquo Financial Research vol 10 pp 1ndash13 2002

[4] L Zhou ldquolsquoSecond financersquo of Chinarsquos financial industry andfinancial division during the reforming periodrdquo The Journal ofWorld Economy vol 6 pp 72ndash79 2003

[5] L Zhou ldquoGradual transition state capacity and fiscal functionof financerdquo Journal of Finance and Economics vol 2 pp 26ndash372005

[6] E L He and W Liu ldquoFinancial capital local governmentsrsquointervention and industrial structuremdashan empirical analysisbased on the data of central regionsrdquo Economic Issues vol 5pp 37ndash40 2011

[7] X L Dou and W J Wang ldquoThe differences and trends of Chi-nese regional financial development and local governmentsrsquobehavior and analysis of tax sharing reform on the differencesand trends of Chinese regional financial developmentrdquo Shang-hai Finance vol 2 pp 9ndash15 2011

[8] J Zhang and K R Shen ldquoLocal government interventionregional financial development and the transformation ofChinarsquos economic growth modemdashan empirical study under thefiscal decentralization systemrdquo Nankai Economic Studies vol 6pp 122ndash141 2008

[9] G Chen X G Yin and Y Pan ldquoChinarsquos financial developmenttax sharing reform and economic growthrdquo Financial Researchvol 2 pp 99ndash109 2006

[10] P P Ding and Y Fu ldquoBehaviors of local governments therelationship of fiscal and finance and macroeconomic volatilityassociated with Chinamdashan analysis under Chinese-style fiscaldecentralization systemrdquo Comparative Economic amp Social Sys-tems vol 6 pp 87ndash97 2012

[11] X G Yin G Chen and Y Pan ldquoTax sharing reform local gov-ernmentsrsquo intervention and financial development efficiencyrdquoJournal of Finance and Economics vol 10 pp 92ndash101 2006

[12] M G Yu and H B Pan ldquoGovernment intervention the ruleof law financial development and bank loans of state-ownedenterprisesrdquo Financial Research vol 9 pp 1ndash23 2008

[13] D X Wang J Liu and L Li ldquoFiscal decentralization bankcredit and total factor productivityrdquo Journal of Finance andEconomics vol 4 pp 69ndash80 2011

[14] Y X Jiang and R B Jiang ldquoRegional financial innovationefficiency evaluation environmental impact and gap analysisrdquoJournal of Zhejiang University (Humanities and Social Sciences)vol 43 no 4 pp 52ndash65 2013

[15] H W Yang L W Cheng Y Xu and Y W Qi ldquoMeasure andinfluencing factors of financial resources allocation efficiencyunder fiscal decentralization system based on super-efficiencyDEA-TOBIT two-step methodrdquo Journal of Jiangxi NormalUniversity (Natural Sciences Edition) vol 6 pp 587ndash592 2014

[16] N Shen F Z Liu and J Q Zhao ldquoFiscal decentralizationfinancial deepening and regional international poverty prone todeveloprdquo Finance amp Trade Economics vol 1 pp 41ndash46 2006

[17] Q Y Du G P Li and N Shen ldquoFinancial efficiency andoptimization of the regional industrial structure under thecurrent fiscal systemrdquo Xinjiang Social Science vol 4 pp 16ndash212006

[18] T Feng Y W Song and Y Lu ldquoFiscal decentralization localgovernmentsrsquo behavior and regional financial developmentrdquoJournal of Xirsquoan Jiaotong University vol 5 pp 20ndash27 2007

[19] Y A Zhou X J Ma and J Y Zhao ldquoGovernment behaviorsfinancial development and economic growthrdquo Henan SocialSciences vol 1 pp 55ndash60 2007

[20] C Hao ldquoDevelopment of financial intermediation and eco-nomic growth the Chinese experiencerdquo China EconomicReview vol 17 no 4 pp 347ndash362 2006

[21] J Zhang Fiscal Decentralization Regional Financial Devel-opment and Chinarsquos Economic Growth Performance NanjingUniversity Press Nanjing China 2011

[22] Y J Yao ldquoGovernment intervention the development of finan-cial intermediation and economic growthrdquo Journal of Financeand Economics vol 8 pp 52ndash58 2010

[23] X Wang and L Li ldquoFiscal decentralization and regional finan-cial development the theoretical and empirical analysis fromthe perspective of central governmentrdquo Public Finance Researchvol 4 pp 31ndash35 2012

[24] W L Zhang and C L Qin ldquoFinancial development fiscaldecentralization and regional economic differencesrdquo EconomicGeography vol 1 pp 583ndash588 2009

[25] Y Ma J Li and B Chen ldquoProvincial differences in Chinesefiscal decentralization financial development industrializationand economic growthrdquo EconomicTheory and Business Manage-ment vol 2 pp 5ndash19 2015

[26] Y X Jiang and R B Jiang ldquoThe spatial external effects ofregional financial innovation financial competition and finan-cial agglomerationrdquo Social Science Front vol 3 pp 79ndash84 2014

[27] K Tone ldquoA slacks-based measure of super-efficiency indata envelopment analysisrdquo European Journal of OperationalResearch vol 143 no 1 pp 32ndash41 2002

[28] Y Q Lu and D G Zhang ldquoChinarsquos regional financial efficiencyand differences of efficiency measurementrdquo Economic Geogra-phy vol 1 pp 96ndash101 2012

[29] A N Berger and D B Humphrey ldquoEfficiency of financialinstitutions international survey and directions for futureresearchrdquo European Journal of Operational Research vol 98 no2 pp 175ndash212 1997

[30] Y J Yao ldquoEmpirical analysis of financial development andurban-rural income gaprdquo Journal of Finance and Economics vol2 pp 49ndash59 2005

[31] Q Li J Song E Wang H Hu J Zhang and Y WangldquoEconomic growth and pollutant emissions in China a spatialeconometric analysisrdquo Stochastic Environmental Research andRisk Assessment vol 28 no 2 pp 429ndash442 2014

[32] J P LeSage and R K Pace Introduction to Spatial EconometricsCRCPress Taylor amp Francis Group Boca Raton Fla USA 2009

[33] Q W Guo and J X Jia ldquoFiscal decentralization organizationalstructure of local governments and the size of governmentexpenditurerdquoEconomic Research Journal vol 11 pp 59ndash87 2010httpwwwcnkicomcnArticleCJFDTotal-JJYJ201011008htm

[34] F Gong and X Lei ldquoThe measurement of Chinese-style fiscaldecentralizationrdquo Statistical Research vol 10 pp 47ndash55 2010

[35] G Fan X L Wang and H P Zhu Chinese Marketing IndexRegional Marketing Process in 2011 Economic Science PressBeijing China 2011

14 Discrete Dynamics in Nature and Society

[36] S S Rosenthal and W C Strange ldquoGeography industrialorganization and agglomerationrdquoTheReview of Economics andStatistics vol 85 no 2 pp 377ndash393 2003

[37] S Park Yoon International Banking and Financial CentersKluwer Academic Boston Mass USA 1989

[38] B Y Qiao J Y Fan and X Y Feng ldquoChinarsquos fiscal decentral-ization and compulsory primary educationrdquo Social Sciences inChina vol 6 pp 37ndash46 2005

[39] W JWang and C L Qin ldquoRegional differences in the behaviorsof local governments and growth effect of fiscal decentralizationbased on the analysis of judgment hypothesis and testingrdquoManagement World vol 1 pp 9ndash21 2008

[40] J K Brueckner ldquoStrategic interaction among governments anoverview of empirical studiesrdquo International Regional ScienceReview vol 26 no 2 pp 175ndash188 2003

[41] F Revelli ldquoOn spatial public finance empiricsrdquo International Taxand Public Finance vol 12 no 4 pp 475ndash492 2005

[42] N Shen and T Y He ldquoFinancial efficiency and regional eco-nomic growth from the perspective of fiscal decentralizationrdquoReform vol 1 pp 61ndash65 2006

[43] J M Liu X M Hu and J G Wu ldquoThe threshold effect ofsub-national fiscal decentralization revenue and expendituredecentralization on provincial industrial transformation andupgrading based on the data of 14 cities in Hunan ProvincerdquoPublic Finance Research vol 8 pp 49ndash52 2014

Submit your manuscripts athttpwwwhindawicom

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

MathematicsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Mathematical Problems in Engineering

Hindawi Publishing Corporationhttpwwwhindawicom

Differential EquationsInternational Journal of

Volume 2014

Applied MathematicsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Probability and StatisticsHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Mathematical PhysicsAdvances in

Complex AnalysisJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

OptimizationJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

CombinatoricsHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Operations ResearchAdvances in

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Function Spaces

Abstract and Applied AnalysisHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of Mathematics and Mathematical Sciences

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Algebra

Discrete Dynamics in Nature and Society

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Decision SciencesAdvances in

Discrete MathematicsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom

Volume 2014 Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Stochastic AnalysisInternational Journal of

Discrete Dynamics in Nature and Society 11

Table 5 Robustness test on Spatial Durbin Models (geographicaldistance weight matrix)

Independentvariables

Dependent variable sd Dependent variable rdSDMmodel

DSDMmodel

SDMmodel

DSDMmodel

feitminus1 mdash 0202lowastlowastlowast(1276) mdash 0200lowastlowastlowast

(1255)

Lnsd or Lnrd 0364lowastlowastlowast(636)

0348lowastlowastlowast(592)

0173lowastlowastlowast(522)

0161lowastlowastlowast(458)

Lnpgdp 0268lowastlowastlowast(735)

0225lowastlowastlowast(578)

0242lowastlowastlowast(637)

0209lowastlowastlowast(522)

Lnurban minus0026(minus112)

minus0026(minus112)

minus0028(minus120)

minus0027(minus115)

Lnitu 0465lowastlowast(246)

0478lowastlowast(238)

0527lowastlowastlowast(278)

0542lowastlowastlowast(269)

Lninternet minus0295lowastlowastlowast(minus3038)

minus0300lowastlowastlowast(minus2799)

minus0292lowastlowastlowast(minus3005)

minus0298lowastlowastlowast(minus2771)

Lnifa minus0484lowastlowastlowast(minus2479)

minus0427lowastlowastlowast(minus2024)

minus0472lowastlowastlowast(minus2448)

minus0417lowastlowastlowast(minus1988)

Lnprivate 0036lowastlowastlowast(312)

0035lowastlowastlowast(284)

0035lowastlowastlowast(307)

0035lowastlowastlowast(288)

WLnsd orWLnrd

minus0264(minus149)

minus0418lowastlowast(minus225)

minus0277lowastlowastlowast(minus370)

minus0241lowastlowastlowast(minus276)

WLnpgdp minus0269lowastlowastlowast(minus427)

minus0286lowastlowastlowast(minus428)

minus0237lowastlowastlowast(minus369)

minus0267lowastlowastlowast(minus394)

WLnurban 0053(147)

0059lowast(167)

0059(163)

0062lowast(173)

WLnitu 0104(020)

minus0085(minus016)

minus0205(minus039)

minus0290(minus051)

WLninternet 0217lowastlowastlowast(838)

0239lowastlowastlowast(853)

0214lowastlowastlowast(831)

0235lowastlowastlowast(841)

WLnifa 0434lowastlowastlowast(1052)

0464lowastlowastlowast(1054)

0433lowastlowastlowast(1165)

0450lowastlowastlowast(1129)

WLnprivate 0062lowast(187)

0038(106)

0081lowastlowast(241)

0050(139)

120588

0547lowastlowastlowast(1858)

0580lowastlowastlowast(1938)

0541lowastlowastlowast(1839)

0570lowastlowastlowast(1898)

1198772 0444 0428 0443 0430

log 119871 659741 665804 665804 659509Observations 2810 2529 2810 2529Note lowastlowastlowast lowastlowast and lowast respectively represent the significant level of 1 5and 10

kinds of fiscal decentralizations indicators which show thatthe gap between the levels of economic developmentwill havean impact on role of fiscal decentralization acting on regionalfinancial efficiency On one hand under the fiscal decen-tralization system because of the heterogeneity of regionaleconomic development level the local governmentsrsquo ability ofmaking up the gap between fiscal revenue and expenditure isdifferent and the fiscal pressure is also different therefore thedegree of interference to the regional market environment bythe local government is different which leads to the regionaldifferences in financial efficiency in a certain degree [21]On the other hand the occurrence of financial innovationdepends on the support of the economic foundation and

the different economic and social environment will lead tothe fact that the financial innovation activities are more andmore concentrated in the regions with the advantages offinancial location resulting in nonequilibrium of financialefficiency in different regions [42] Under the performanceassessment method focusing on GDP local governments willnot only consider the strategic behaviors of the neighboringregions but also tend to think about the behaviors of theregions with the similar level of economy In additioncross-regional knowledge diffusion technology spillover andhuman capital flow are more prone to occur among theregions with similar level of economic development so thatthe financial exchanges and contacts among regions are morefrequent It also shows that besides considering the impact ofgeographical distance the economic distance factor shouldbe taken into consideration while we are researching thespatial effects of fiscal decentralization on regional financialinnovation in addition to the consideration of the impact ofgeographical distance the economic distance factor shouldbe considered

6 Conclusions and Enlightenments

Based on the panel data of 281 prefecture-level cities in Chinacovering the period of 2003ndash2012 the regional financialefficiency is measured by the superefficiency SBMmodel thegeographic distance weight matrix and economic-geographymixed weight matrix which can reflect the regional similarityare introduced into the SDM and the spatial effects of thefiscal expenditure and revenue decentralization on regionalfinancial efficiency under province are investigated as wellas the strategic financial competition arising from it Basedon the results of previous empirical analysis we can draw thefollowing basic conclusions

First the geographical distribution of Chinarsquos prefecture-level citiesrsquo financial efficiency presents significant features oftime inertia and spatial dependency Financial developmentand innovation among regions particularly neighboringregions is not independent but shows significant spatialspillover effects The financial efficiency of the local regionwill be significantly affected by the financial efficiency ofadjacent regions The spatial effect is increasingly obviousin the process of financial development and innovation andthe size of the spillover effects is closely related to spatialweight (which is used to measure regional similarities) andmainly shows that there is competition for financial efficiencyamong regions with similar economic attributes and adjacentregions

Second it cannot be ignored that the institutional factorsespecially fiscal decentralization have effect on the financialefficiency and the factors have a decisive role in the allocationof financial resource and the regional distribution of financialinnovation The improvement of fiscal decentralization levelwill help to attract more financial resources thereby promot-ing an increase in financial efficiencyThe improvement of thefinancial autonomy in neighboring jurisdictions will reducethe financial resources flowing into the local region andsuppresses the occurrence of the local financial innovationIn addition there exist significant spatial spillover effects of

12 Discrete Dynamics in Nature and Society

financial development and efficiency caused by fiscal decen-tralization The scope of spatial spillovers is mainly reflectedin two aspects Thus the local governmentsrsquo competition notonly strongly increases the financial efficiency of the regionbut also accelerates interregional competition for financialefficiency

Third fiscal expenditure decentralization and revenuedecentralizationrsquos impact on regional financial efficiencyhave nonequilibrium characteristics First of all there existsasymmetry between fiscal expenditure decentralization andrevenue decentralizationrsquos impact on regional financial effi-ciency Fiscal expenditure decentralization can promoteoptimization of the fiscal expenditure structure and fiscalrevenue decentralization can increase the local governmentsrsquorevenue share so as to provide positive incentives [43] bothof them have a positive and promoting effect on the improve-ment of local financial efficiency and such effect largelydepends on the symmetry between the fiscal expenditure andrevenue decentralization If it is asymmetric between theirproperty rights and the expenditure responsibilities underfiscal decentralization system the local governments maymake excessive intervention to regional financial resourcesallocation tomeet their own demands for financial resourcesNext there exists nonconformance between fiscal expendi-ture decentralization and revenue decentralizationrsquos impacton financial efficiency in neighboring regions Comparedwith the fiscal expenditure decentralization the fiscal revenuedecentralization has more apparent inhibitory effect on thefinancial efficiency in neighboring regions

Our research conclusions have important policy implica-tions that illuminate the relationship between fiscal decen-tralization and financial efficiency

(1) The regional financial efficiency is closely relatedto the location and the economic development of the sur-rounding cities The regional financial efficiency is affectednot only by the financial development and innovation inneighboring regions but also by the impact of the struc-tural differences among regions In this case breaking theadministrativemonopoly of local protectionism and carryingon the ldquorichening neighborrdquo policy of financial developmentare the best choices for the local government to developthe regional economy When making policies of financialdevelopment local governments need to consider their owneconomic development environment and economic featuresin neighboring regions and strengthen communication andcoordination with the surrounding regions so as to avoidthe possible contradictions and conflicts on financial devel-opment policies among regions

(2) The spatial spillover effects of the regional financialefficiency should be divided into two kinds of effects includ-ing global geographic spillover effects and local geographicspillover effects In order to promote the increase of financialefficiency in a better way it should also be based on thehierarchy of spatial effect of and the degree of spillover effectto define the relationship between the government and themarket and central and local governments and among thelocal governments and their role in regional financial devel-opment and innovation so as to better promote financialefficiency

(3) There is a two-way interactive mechanism betweenthe fiscal system and financial development such effectsmay affect the behaviors of the local governments and theallocation of resources and ultimately act on macroeconomicperformance [10] Under the fiscal decentralization systemthe strategic competitive behaviors among local governmentssignificantly affect the liquidity of financial resources and theoccurrence of financial innovation the local governmentsrsquoexpansion of fiscal expenditure and revenue shortfall becomethe main driving force of each region competing for financialresources In order to better play the role of incentivesand constraints of fiscal competition mechanism under thefiscal decentralization system and avoid the negative impactbrought by the vicious interregional strategic financial com-petitive behaviors on the regional economic development thelocal governmentsrsquo short-sighted behaviors should be strictlyregulated and then the interregional financial cooperationwith mutual benefit as the core should be actively promoted

(4) Constructing a scientific and rational framework forfiscal decentralization will be an important force to accelerateeffectively the improvement of regional financial efficiencyChinarsquos future subprovincial fiscal decentralization reformshould adapt to the new normal financial development andfurther rationalize the provincial and subprovincial fiscaldecentralization system to ensure a balanced and appropriatefiscal expenditure and revenue decentralization On onehand we should weaken local governmentsrsquo negative inter-vention in financial system by reducing local governmentsrsquofinancial pressure caused by the asymmetry between thefiscal resources and expenditure responsibilities On the otherhandwe should also actively play the function of fiscal systemto adjust the financial development and innovation and fullyplay public financersquos role in promoting the flowing of finan-cial resources among regions the balanced development ofregional finance and the improvement of financial efficiency

Competing Interests

There are no competing interests in regard to the contentdiscussed in this paper

Authorsrsquo Contributions

All authors conceived and designed the research XiaomeiHucollected and processed the data Jianmin Liu analyzed thedata and interpreted the results all authors were also involvedin writing the paper and have all approved the submittedform

Acknowledgments

The research of this paper is jointly funded by the NationalSocial Science Foundation of China under Project no14BJY159 and the Hunan Philosophy and Social ScienceFund Project under Grant no 13JD09 The authors gratefullyacknowledge the support of Jinguang Wu from HunanUniversity of Finance and Economics China

Discrete Dynamics in Nature and Society 13

References

[1] J Zhang and X Jin ldquoRe-examine of the relationship betweenChinarsquos financial deepening and productivity 1987ndash2001rdquo Eco-nomic Research Journal vol 11 pp 35ndash43 2005

[2] K Chen A L Hillman and Q Y Gu ldquoFiscal centralization andchanges of local governmentsrsquo behaviorsmdashfrom lsquohands for helprsquoto lsquohands for grabrsquordquoChina EconomicQuarterly vol 1 pp 111ndash1302002

[3] L Zhou and Z M Wang ldquoEmpirical analysis of regionalfinancial development of China and economic growth 1978ndash2000rdquo Financial Research vol 10 pp 1ndash13 2002

[4] L Zhou ldquolsquoSecond financersquo of Chinarsquos financial industry andfinancial division during the reforming periodrdquo The Journal ofWorld Economy vol 6 pp 72ndash79 2003

[5] L Zhou ldquoGradual transition state capacity and fiscal functionof financerdquo Journal of Finance and Economics vol 2 pp 26ndash372005

[6] E L He and W Liu ldquoFinancial capital local governmentsrsquointervention and industrial structuremdashan empirical analysisbased on the data of central regionsrdquo Economic Issues vol 5pp 37ndash40 2011

[7] X L Dou and W J Wang ldquoThe differences and trends of Chi-nese regional financial development and local governmentsrsquobehavior and analysis of tax sharing reform on the differencesand trends of Chinese regional financial developmentrdquo Shang-hai Finance vol 2 pp 9ndash15 2011

[8] J Zhang and K R Shen ldquoLocal government interventionregional financial development and the transformation ofChinarsquos economic growth modemdashan empirical study under thefiscal decentralization systemrdquo Nankai Economic Studies vol 6pp 122ndash141 2008

[9] G Chen X G Yin and Y Pan ldquoChinarsquos financial developmenttax sharing reform and economic growthrdquo Financial Researchvol 2 pp 99ndash109 2006

[10] P P Ding and Y Fu ldquoBehaviors of local governments therelationship of fiscal and finance and macroeconomic volatilityassociated with Chinamdashan analysis under Chinese-style fiscaldecentralization systemrdquo Comparative Economic amp Social Sys-tems vol 6 pp 87ndash97 2012

[11] X G Yin G Chen and Y Pan ldquoTax sharing reform local gov-ernmentsrsquo intervention and financial development efficiencyrdquoJournal of Finance and Economics vol 10 pp 92ndash101 2006

[12] M G Yu and H B Pan ldquoGovernment intervention the ruleof law financial development and bank loans of state-ownedenterprisesrdquo Financial Research vol 9 pp 1ndash23 2008

[13] D X Wang J Liu and L Li ldquoFiscal decentralization bankcredit and total factor productivityrdquo Journal of Finance andEconomics vol 4 pp 69ndash80 2011

[14] Y X Jiang and R B Jiang ldquoRegional financial innovationefficiency evaluation environmental impact and gap analysisrdquoJournal of Zhejiang University (Humanities and Social Sciences)vol 43 no 4 pp 52ndash65 2013

[15] H W Yang L W Cheng Y Xu and Y W Qi ldquoMeasure andinfluencing factors of financial resources allocation efficiencyunder fiscal decentralization system based on super-efficiencyDEA-TOBIT two-step methodrdquo Journal of Jiangxi NormalUniversity (Natural Sciences Edition) vol 6 pp 587ndash592 2014

[16] N Shen F Z Liu and J Q Zhao ldquoFiscal decentralizationfinancial deepening and regional international poverty prone todeveloprdquo Finance amp Trade Economics vol 1 pp 41ndash46 2006

[17] Q Y Du G P Li and N Shen ldquoFinancial efficiency andoptimization of the regional industrial structure under thecurrent fiscal systemrdquo Xinjiang Social Science vol 4 pp 16ndash212006

[18] T Feng Y W Song and Y Lu ldquoFiscal decentralization localgovernmentsrsquo behavior and regional financial developmentrdquoJournal of Xirsquoan Jiaotong University vol 5 pp 20ndash27 2007

[19] Y A Zhou X J Ma and J Y Zhao ldquoGovernment behaviorsfinancial development and economic growthrdquo Henan SocialSciences vol 1 pp 55ndash60 2007

[20] C Hao ldquoDevelopment of financial intermediation and eco-nomic growth the Chinese experiencerdquo China EconomicReview vol 17 no 4 pp 347ndash362 2006

[21] J Zhang Fiscal Decentralization Regional Financial Devel-opment and Chinarsquos Economic Growth Performance NanjingUniversity Press Nanjing China 2011

[22] Y J Yao ldquoGovernment intervention the development of finan-cial intermediation and economic growthrdquo Journal of Financeand Economics vol 8 pp 52ndash58 2010

[23] X Wang and L Li ldquoFiscal decentralization and regional finan-cial development the theoretical and empirical analysis fromthe perspective of central governmentrdquo Public Finance Researchvol 4 pp 31ndash35 2012

[24] W L Zhang and C L Qin ldquoFinancial development fiscaldecentralization and regional economic differencesrdquo EconomicGeography vol 1 pp 583ndash588 2009

[25] Y Ma J Li and B Chen ldquoProvincial differences in Chinesefiscal decentralization financial development industrializationand economic growthrdquo EconomicTheory and Business Manage-ment vol 2 pp 5ndash19 2015

[26] Y X Jiang and R B Jiang ldquoThe spatial external effects ofregional financial innovation financial competition and finan-cial agglomerationrdquo Social Science Front vol 3 pp 79ndash84 2014

[27] K Tone ldquoA slacks-based measure of super-efficiency indata envelopment analysisrdquo European Journal of OperationalResearch vol 143 no 1 pp 32ndash41 2002

[28] Y Q Lu and D G Zhang ldquoChinarsquos regional financial efficiencyand differences of efficiency measurementrdquo Economic Geogra-phy vol 1 pp 96ndash101 2012

[29] A N Berger and D B Humphrey ldquoEfficiency of financialinstitutions international survey and directions for futureresearchrdquo European Journal of Operational Research vol 98 no2 pp 175ndash212 1997

[30] Y J Yao ldquoEmpirical analysis of financial development andurban-rural income gaprdquo Journal of Finance and Economics vol2 pp 49ndash59 2005

[31] Q Li J Song E Wang H Hu J Zhang and Y WangldquoEconomic growth and pollutant emissions in China a spatialeconometric analysisrdquo Stochastic Environmental Research andRisk Assessment vol 28 no 2 pp 429ndash442 2014

[32] J P LeSage and R K Pace Introduction to Spatial EconometricsCRCPress Taylor amp Francis Group Boca Raton Fla USA 2009

[33] Q W Guo and J X Jia ldquoFiscal decentralization organizationalstructure of local governments and the size of governmentexpenditurerdquoEconomic Research Journal vol 11 pp 59ndash87 2010httpwwwcnkicomcnArticleCJFDTotal-JJYJ201011008htm

[34] F Gong and X Lei ldquoThe measurement of Chinese-style fiscaldecentralizationrdquo Statistical Research vol 10 pp 47ndash55 2010

[35] G Fan X L Wang and H P Zhu Chinese Marketing IndexRegional Marketing Process in 2011 Economic Science PressBeijing China 2011

14 Discrete Dynamics in Nature and Society

[36] S S Rosenthal and W C Strange ldquoGeography industrialorganization and agglomerationrdquoTheReview of Economics andStatistics vol 85 no 2 pp 377ndash393 2003

[37] S Park Yoon International Banking and Financial CentersKluwer Academic Boston Mass USA 1989

[38] B Y Qiao J Y Fan and X Y Feng ldquoChinarsquos fiscal decentral-ization and compulsory primary educationrdquo Social Sciences inChina vol 6 pp 37ndash46 2005

[39] W JWang and C L Qin ldquoRegional differences in the behaviorsof local governments and growth effect of fiscal decentralizationbased on the analysis of judgment hypothesis and testingrdquoManagement World vol 1 pp 9ndash21 2008

[40] J K Brueckner ldquoStrategic interaction among governments anoverview of empirical studiesrdquo International Regional ScienceReview vol 26 no 2 pp 175ndash188 2003

[41] F Revelli ldquoOn spatial public finance empiricsrdquo International Taxand Public Finance vol 12 no 4 pp 475ndash492 2005

[42] N Shen and T Y He ldquoFinancial efficiency and regional eco-nomic growth from the perspective of fiscal decentralizationrdquoReform vol 1 pp 61ndash65 2006

[43] J M Liu X M Hu and J G Wu ldquoThe threshold effect ofsub-national fiscal decentralization revenue and expendituredecentralization on provincial industrial transformation andupgrading based on the data of 14 cities in Hunan ProvincerdquoPublic Finance Research vol 8 pp 49ndash52 2014

Submit your manuscripts athttpwwwhindawicom

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

MathematicsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Mathematical Problems in Engineering

Hindawi Publishing Corporationhttpwwwhindawicom

Differential EquationsInternational Journal of

Volume 2014

Applied MathematicsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Probability and StatisticsHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Mathematical PhysicsAdvances in

Complex AnalysisJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

OptimizationJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

CombinatoricsHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Operations ResearchAdvances in

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Function Spaces

Abstract and Applied AnalysisHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of Mathematics and Mathematical Sciences

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Algebra

Discrete Dynamics in Nature and Society

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Decision SciencesAdvances in

Discrete MathematicsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom

Volume 2014 Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Stochastic AnalysisInternational Journal of

12 Discrete Dynamics in Nature and Society

financial development and efficiency caused by fiscal decen-tralization The scope of spatial spillovers is mainly reflectedin two aspects Thus the local governmentsrsquo competition notonly strongly increases the financial efficiency of the regionbut also accelerates interregional competition for financialefficiency

Third fiscal expenditure decentralization and revenuedecentralizationrsquos impact on regional financial efficiencyhave nonequilibrium characteristics First of all there existsasymmetry between fiscal expenditure decentralization andrevenue decentralizationrsquos impact on regional financial effi-ciency Fiscal expenditure decentralization can promoteoptimization of the fiscal expenditure structure and fiscalrevenue decentralization can increase the local governmentsrsquorevenue share so as to provide positive incentives [43] bothof them have a positive and promoting effect on the improve-ment of local financial efficiency and such effect largelydepends on the symmetry between the fiscal expenditure andrevenue decentralization If it is asymmetric between theirproperty rights and the expenditure responsibilities underfiscal decentralization system the local governments maymake excessive intervention to regional financial resourcesallocation tomeet their own demands for financial resourcesNext there exists nonconformance between fiscal expendi-ture decentralization and revenue decentralizationrsquos impacton financial efficiency in neighboring regions Comparedwith the fiscal expenditure decentralization the fiscal revenuedecentralization has more apparent inhibitory effect on thefinancial efficiency in neighboring regions

Our research conclusions have important policy implica-tions that illuminate the relationship between fiscal decen-tralization and financial efficiency

(1) The regional financial efficiency is closely relatedto the location and the economic development of the sur-rounding cities The regional financial efficiency is affectednot only by the financial development and innovation inneighboring regions but also by the impact of the struc-tural differences among regions In this case breaking theadministrativemonopoly of local protectionism and carryingon the ldquorichening neighborrdquo policy of financial developmentare the best choices for the local government to developthe regional economy When making policies of financialdevelopment local governments need to consider their owneconomic development environment and economic featuresin neighboring regions and strengthen communication andcoordination with the surrounding regions so as to avoidthe possible contradictions and conflicts on financial devel-opment policies among regions

(2) The spatial spillover effects of the regional financialefficiency should be divided into two kinds of effects includ-ing global geographic spillover effects and local geographicspillover effects In order to promote the increase of financialefficiency in a better way it should also be based on thehierarchy of spatial effect of and the degree of spillover effectto define the relationship between the government and themarket and central and local governments and among thelocal governments and their role in regional financial devel-opment and innovation so as to better promote financialefficiency

(3) There is a two-way interactive mechanism betweenthe fiscal system and financial development such effectsmay affect the behaviors of the local governments and theallocation of resources and ultimately act on macroeconomicperformance [10] Under the fiscal decentralization systemthe strategic competitive behaviors among local governmentssignificantly affect the liquidity of financial resources and theoccurrence of financial innovation the local governmentsrsquoexpansion of fiscal expenditure and revenue shortfall becomethe main driving force of each region competing for financialresources In order to better play the role of incentivesand constraints of fiscal competition mechanism under thefiscal decentralization system and avoid the negative impactbrought by the vicious interregional strategic financial com-petitive behaviors on the regional economic development thelocal governmentsrsquo short-sighted behaviors should be strictlyregulated and then the interregional financial cooperationwith mutual benefit as the core should be actively promoted

(4) Constructing a scientific and rational framework forfiscal decentralization will be an important force to accelerateeffectively the improvement of regional financial efficiencyChinarsquos future subprovincial fiscal decentralization reformshould adapt to the new normal financial development andfurther rationalize the provincial and subprovincial fiscaldecentralization system to ensure a balanced and appropriatefiscal expenditure and revenue decentralization On onehand we should weaken local governmentsrsquo negative inter-vention in financial system by reducing local governmentsrsquofinancial pressure caused by the asymmetry between thefiscal resources and expenditure responsibilities On the otherhandwe should also actively play the function of fiscal systemto adjust the financial development and innovation and fullyplay public financersquos role in promoting the flowing of finan-cial resources among regions the balanced development ofregional finance and the improvement of financial efficiency

Competing Interests

There are no competing interests in regard to the contentdiscussed in this paper

Authorsrsquo Contributions

All authors conceived and designed the research XiaomeiHucollected and processed the data Jianmin Liu analyzed thedata and interpreted the results all authors were also involvedin writing the paper and have all approved the submittedform

Acknowledgments

The research of this paper is jointly funded by the NationalSocial Science Foundation of China under Project no14BJY159 and the Hunan Philosophy and Social ScienceFund Project under Grant no 13JD09 The authors gratefullyacknowledge the support of Jinguang Wu from HunanUniversity of Finance and Economics China

Discrete Dynamics in Nature and Society 13

References

[1] J Zhang and X Jin ldquoRe-examine of the relationship betweenChinarsquos financial deepening and productivity 1987ndash2001rdquo Eco-nomic Research Journal vol 11 pp 35ndash43 2005

[2] K Chen A L Hillman and Q Y Gu ldquoFiscal centralization andchanges of local governmentsrsquo behaviorsmdashfrom lsquohands for helprsquoto lsquohands for grabrsquordquoChina EconomicQuarterly vol 1 pp 111ndash1302002

[3] L Zhou and Z M Wang ldquoEmpirical analysis of regionalfinancial development of China and economic growth 1978ndash2000rdquo Financial Research vol 10 pp 1ndash13 2002

[4] L Zhou ldquolsquoSecond financersquo of Chinarsquos financial industry andfinancial division during the reforming periodrdquo The Journal ofWorld Economy vol 6 pp 72ndash79 2003

[5] L Zhou ldquoGradual transition state capacity and fiscal functionof financerdquo Journal of Finance and Economics vol 2 pp 26ndash372005

[6] E L He and W Liu ldquoFinancial capital local governmentsrsquointervention and industrial structuremdashan empirical analysisbased on the data of central regionsrdquo Economic Issues vol 5pp 37ndash40 2011

[7] X L Dou and W J Wang ldquoThe differences and trends of Chi-nese regional financial development and local governmentsrsquobehavior and analysis of tax sharing reform on the differencesand trends of Chinese regional financial developmentrdquo Shang-hai Finance vol 2 pp 9ndash15 2011

[8] J Zhang and K R Shen ldquoLocal government interventionregional financial development and the transformation ofChinarsquos economic growth modemdashan empirical study under thefiscal decentralization systemrdquo Nankai Economic Studies vol 6pp 122ndash141 2008

[9] G Chen X G Yin and Y Pan ldquoChinarsquos financial developmenttax sharing reform and economic growthrdquo Financial Researchvol 2 pp 99ndash109 2006

[10] P P Ding and Y Fu ldquoBehaviors of local governments therelationship of fiscal and finance and macroeconomic volatilityassociated with Chinamdashan analysis under Chinese-style fiscaldecentralization systemrdquo Comparative Economic amp Social Sys-tems vol 6 pp 87ndash97 2012

[11] X G Yin G Chen and Y Pan ldquoTax sharing reform local gov-ernmentsrsquo intervention and financial development efficiencyrdquoJournal of Finance and Economics vol 10 pp 92ndash101 2006

[12] M G Yu and H B Pan ldquoGovernment intervention the ruleof law financial development and bank loans of state-ownedenterprisesrdquo Financial Research vol 9 pp 1ndash23 2008

[13] D X Wang J Liu and L Li ldquoFiscal decentralization bankcredit and total factor productivityrdquo Journal of Finance andEconomics vol 4 pp 69ndash80 2011

[14] Y X Jiang and R B Jiang ldquoRegional financial innovationefficiency evaluation environmental impact and gap analysisrdquoJournal of Zhejiang University (Humanities and Social Sciences)vol 43 no 4 pp 52ndash65 2013

[15] H W Yang L W Cheng Y Xu and Y W Qi ldquoMeasure andinfluencing factors of financial resources allocation efficiencyunder fiscal decentralization system based on super-efficiencyDEA-TOBIT two-step methodrdquo Journal of Jiangxi NormalUniversity (Natural Sciences Edition) vol 6 pp 587ndash592 2014

[16] N Shen F Z Liu and J Q Zhao ldquoFiscal decentralizationfinancial deepening and regional international poverty prone todeveloprdquo Finance amp Trade Economics vol 1 pp 41ndash46 2006

[17] Q Y Du G P Li and N Shen ldquoFinancial efficiency andoptimization of the regional industrial structure under thecurrent fiscal systemrdquo Xinjiang Social Science vol 4 pp 16ndash212006

[18] T Feng Y W Song and Y Lu ldquoFiscal decentralization localgovernmentsrsquo behavior and regional financial developmentrdquoJournal of Xirsquoan Jiaotong University vol 5 pp 20ndash27 2007

[19] Y A Zhou X J Ma and J Y Zhao ldquoGovernment behaviorsfinancial development and economic growthrdquo Henan SocialSciences vol 1 pp 55ndash60 2007

[20] C Hao ldquoDevelopment of financial intermediation and eco-nomic growth the Chinese experiencerdquo China EconomicReview vol 17 no 4 pp 347ndash362 2006

[21] J Zhang Fiscal Decentralization Regional Financial Devel-opment and Chinarsquos Economic Growth Performance NanjingUniversity Press Nanjing China 2011

[22] Y J Yao ldquoGovernment intervention the development of finan-cial intermediation and economic growthrdquo Journal of Financeand Economics vol 8 pp 52ndash58 2010

[23] X Wang and L Li ldquoFiscal decentralization and regional finan-cial development the theoretical and empirical analysis fromthe perspective of central governmentrdquo Public Finance Researchvol 4 pp 31ndash35 2012

[24] W L Zhang and C L Qin ldquoFinancial development fiscaldecentralization and regional economic differencesrdquo EconomicGeography vol 1 pp 583ndash588 2009

[25] Y Ma J Li and B Chen ldquoProvincial differences in Chinesefiscal decentralization financial development industrializationand economic growthrdquo EconomicTheory and Business Manage-ment vol 2 pp 5ndash19 2015

[26] Y X Jiang and R B Jiang ldquoThe spatial external effects ofregional financial innovation financial competition and finan-cial agglomerationrdquo Social Science Front vol 3 pp 79ndash84 2014

[27] K Tone ldquoA slacks-based measure of super-efficiency indata envelopment analysisrdquo European Journal of OperationalResearch vol 143 no 1 pp 32ndash41 2002

[28] Y Q Lu and D G Zhang ldquoChinarsquos regional financial efficiencyand differences of efficiency measurementrdquo Economic Geogra-phy vol 1 pp 96ndash101 2012

[29] A N Berger and D B Humphrey ldquoEfficiency of financialinstitutions international survey and directions for futureresearchrdquo European Journal of Operational Research vol 98 no2 pp 175ndash212 1997

[30] Y J Yao ldquoEmpirical analysis of financial development andurban-rural income gaprdquo Journal of Finance and Economics vol2 pp 49ndash59 2005

[31] Q Li J Song E Wang H Hu J Zhang and Y WangldquoEconomic growth and pollutant emissions in China a spatialeconometric analysisrdquo Stochastic Environmental Research andRisk Assessment vol 28 no 2 pp 429ndash442 2014

[32] J P LeSage and R K Pace Introduction to Spatial EconometricsCRCPress Taylor amp Francis Group Boca Raton Fla USA 2009

[33] Q W Guo and J X Jia ldquoFiscal decentralization organizationalstructure of local governments and the size of governmentexpenditurerdquoEconomic Research Journal vol 11 pp 59ndash87 2010httpwwwcnkicomcnArticleCJFDTotal-JJYJ201011008htm

[34] F Gong and X Lei ldquoThe measurement of Chinese-style fiscaldecentralizationrdquo Statistical Research vol 10 pp 47ndash55 2010

[35] G Fan X L Wang and H P Zhu Chinese Marketing IndexRegional Marketing Process in 2011 Economic Science PressBeijing China 2011

14 Discrete Dynamics in Nature and Society

[36] S S Rosenthal and W C Strange ldquoGeography industrialorganization and agglomerationrdquoTheReview of Economics andStatistics vol 85 no 2 pp 377ndash393 2003

[37] S Park Yoon International Banking and Financial CentersKluwer Academic Boston Mass USA 1989

[38] B Y Qiao J Y Fan and X Y Feng ldquoChinarsquos fiscal decentral-ization and compulsory primary educationrdquo Social Sciences inChina vol 6 pp 37ndash46 2005

[39] W JWang and C L Qin ldquoRegional differences in the behaviorsof local governments and growth effect of fiscal decentralizationbased on the analysis of judgment hypothesis and testingrdquoManagement World vol 1 pp 9ndash21 2008

[40] J K Brueckner ldquoStrategic interaction among governments anoverview of empirical studiesrdquo International Regional ScienceReview vol 26 no 2 pp 175ndash188 2003

[41] F Revelli ldquoOn spatial public finance empiricsrdquo International Taxand Public Finance vol 12 no 4 pp 475ndash492 2005

[42] N Shen and T Y He ldquoFinancial efficiency and regional eco-nomic growth from the perspective of fiscal decentralizationrdquoReform vol 1 pp 61ndash65 2006

[43] J M Liu X M Hu and J G Wu ldquoThe threshold effect ofsub-national fiscal decentralization revenue and expendituredecentralization on provincial industrial transformation andupgrading based on the data of 14 cities in Hunan ProvincerdquoPublic Finance Research vol 8 pp 49ndash52 2014

Submit your manuscripts athttpwwwhindawicom

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

MathematicsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Mathematical Problems in Engineering

Hindawi Publishing Corporationhttpwwwhindawicom

Differential EquationsInternational Journal of

Volume 2014

Applied MathematicsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Probability and StatisticsHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Mathematical PhysicsAdvances in

Complex AnalysisJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

OptimizationJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

CombinatoricsHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Operations ResearchAdvances in

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Function Spaces

Abstract and Applied AnalysisHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of Mathematics and Mathematical Sciences

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Algebra

Discrete Dynamics in Nature and Society

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Decision SciencesAdvances in

Discrete MathematicsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom

Volume 2014 Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Stochastic AnalysisInternational Journal of

Discrete Dynamics in Nature and Society 13

References

[1] J Zhang and X Jin ldquoRe-examine of the relationship betweenChinarsquos financial deepening and productivity 1987ndash2001rdquo Eco-nomic Research Journal vol 11 pp 35ndash43 2005

[2] K Chen A L Hillman and Q Y Gu ldquoFiscal centralization andchanges of local governmentsrsquo behaviorsmdashfrom lsquohands for helprsquoto lsquohands for grabrsquordquoChina EconomicQuarterly vol 1 pp 111ndash1302002

[3] L Zhou and Z M Wang ldquoEmpirical analysis of regionalfinancial development of China and economic growth 1978ndash2000rdquo Financial Research vol 10 pp 1ndash13 2002

[4] L Zhou ldquolsquoSecond financersquo of Chinarsquos financial industry andfinancial division during the reforming periodrdquo The Journal ofWorld Economy vol 6 pp 72ndash79 2003

[5] L Zhou ldquoGradual transition state capacity and fiscal functionof financerdquo Journal of Finance and Economics vol 2 pp 26ndash372005

[6] E L He and W Liu ldquoFinancial capital local governmentsrsquointervention and industrial structuremdashan empirical analysisbased on the data of central regionsrdquo Economic Issues vol 5pp 37ndash40 2011

[7] X L Dou and W J Wang ldquoThe differences and trends of Chi-nese regional financial development and local governmentsrsquobehavior and analysis of tax sharing reform on the differencesand trends of Chinese regional financial developmentrdquo Shang-hai Finance vol 2 pp 9ndash15 2011

[8] J Zhang and K R Shen ldquoLocal government interventionregional financial development and the transformation ofChinarsquos economic growth modemdashan empirical study under thefiscal decentralization systemrdquo Nankai Economic Studies vol 6pp 122ndash141 2008

[9] G Chen X G Yin and Y Pan ldquoChinarsquos financial developmenttax sharing reform and economic growthrdquo Financial Researchvol 2 pp 99ndash109 2006

[10] P P Ding and Y Fu ldquoBehaviors of local governments therelationship of fiscal and finance and macroeconomic volatilityassociated with Chinamdashan analysis under Chinese-style fiscaldecentralization systemrdquo Comparative Economic amp Social Sys-tems vol 6 pp 87ndash97 2012

[11] X G Yin G Chen and Y Pan ldquoTax sharing reform local gov-ernmentsrsquo intervention and financial development efficiencyrdquoJournal of Finance and Economics vol 10 pp 92ndash101 2006

[12] M G Yu and H B Pan ldquoGovernment intervention the ruleof law financial development and bank loans of state-ownedenterprisesrdquo Financial Research vol 9 pp 1ndash23 2008

[13] D X Wang J Liu and L Li ldquoFiscal decentralization bankcredit and total factor productivityrdquo Journal of Finance andEconomics vol 4 pp 69ndash80 2011

[14] Y X Jiang and R B Jiang ldquoRegional financial innovationefficiency evaluation environmental impact and gap analysisrdquoJournal of Zhejiang University (Humanities and Social Sciences)vol 43 no 4 pp 52ndash65 2013

[15] H W Yang L W Cheng Y Xu and Y W Qi ldquoMeasure andinfluencing factors of financial resources allocation efficiencyunder fiscal decentralization system based on super-efficiencyDEA-TOBIT two-step methodrdquo Journal of Jiangxi NormalUniversity (Natural Sciences Edition) vol 6 pp 587ndash592 2014

[16] N Shen F Z Liu and J Q Zhao ldquoFiscal decentralizationfinancial deepening and regional international poverty prone todeveloprdquo Finance amp Trade Economics vol 1 pp 41ndash46 2006

[17] Q Y Du G P Li and N Shen ldquoFinancial efficiency andoptimization of the regional industrial structure under thecurrent fiscal systemrdquo Xinjiang Social Science vol 4 pp 16ndash212006

[18] T Feng Y W Song and Y Lu ldquoFiscal decentralization localgovernmentsrsquo behavior and regional financial developmentrdquoJournal of Xirsquoan Jiaotong University vol 5 pp 20ndash27 2007

[19] Y A Zhou X J Ma and J Y Zhao ldquoGovernment behaviorsfinancial development and economic growthrdquo Henan SocialSciences vol 1 pp 55ndash60 2007

[20] C Hao ldquoDevelopment of financial intermediation and eco-nomic growth the Chinese experiencerdquo China EconomicReview vol 17 no 4 pp 347ndash362 2006

[21] J Zhang Fiscal Decentralization Regional Financial Devel-opment and Chinarsquos Economic Growth Performance NanjingUniversity Press Nanjing China 2011

[22] Y J Yao ldquoGovernment intervention the development of finan-cial intermediation and economic growthrdquo Journal of Financeand Economics vol 8 pp 52ndash58 2010

[23] X Wang and L Li ldquoFiscal decentralization and regional finan-cial development the theoretical and empirical analysis fromthe perspective of central governmentrdquo Public Finance Researchvol 4 pp 31ndash35 2012

[24] W L Zhang and C L Qin ldquoFinancial development fiscaldecentralization and regional economic differencesrdquo EconomicGeography vol 1 pp 583ndash588 2009

[25] Y Ma J Li and B Chen ldquoProvincial differences in Chinesefiscal decentralization financial development industrializationand economic growthrdquo EconomicTheory and Business Manage-ment vol 2 pp 5ndash19 2015

[26] Y X Jiang and R B Jiang ldquoThe spatial external effects ofregional financial innovation financial competition and finan-cial agglomerationrdquo Social Science Front vol 3 pp 79ndash84 2014

[27] K Tone ldquoA slacks-based measure of super-efficiency indata envelopment analysisrdquo European Journal of OperationalResearch vol 143 no 1 pp 32ndash41 2002

[28] Y Q Lu and D G Zhang ldquoChinarsquos regional financial efficiencyand differences of efficiency measurementrdquo Economic Geogra-phy vol 1 pp 96ndash101 2012

[29] A N Berger and D B Humphrey ldquoEfficiency of financialinstitutions international survey and directions for futureresearchrdquo European Journal of Operational Research vol 98 no2 pp 175ndash212 1997

[30] Y J Yao ldquoEmpirical analysis of financial development andurban-rural income gaprdquo Journal of Finance and Economics vol2 pp 49ndash59 2005

[31] Q Li J Song E Wang H Hu J Zhang and Y WangldquoEconomic growth and pollutant emissions in China a spatialeconometric analysisrdquo Stochastic Environmental Research andRisk Assessment vol 28 no 2 pp 429ndash442 2014

[32] J P LeSage and R K Pace Introduction to Spatial EconometricsCRCPress Taylor amp Francis Group Boca Raton Fla USA 2009

[33] Q W Guo and J X Jia ldquoFiscal decentralization organizationalstructure of local governments and the size of governmentexpenditurerdquoEconomic Research Journal vol 11 pp 59ndash87 2010httpwwwcnkicomcnArticleCJFDTotal-JJYJ201011008htm

[34] F Gong and X Lei ldquoThe measurement of Chinese-style fiscaldecentralizationrdquo Statistical Research vol 10 pp 47ndash55 2010

[35] G Fan X L Wang and H P Zhu Chinese Marketing IndexRegional Marketing Process in 2011 Economic Science PressBeijing China 2011

14 Discrete Dynamics in Nature and Society

[36] S S Rosenthal and W C Strange ldquoGeography industrialorganization and agglomerationrdquoTheReview of Economics andStatistics vol 85 no 2 pp 377ndash393 2003

[37] S Park Yoon International Banking and Financial CentersKluwer Academic Boston Mass USA 1989

[38] B Y Qiao J Y Fan and X Y Feng ldquoChinarsquos fiscal decentral-ization and compulsory primary educationrdquo Social Sciences inChina vol 6 pp 37ndash46 2005

[39] W JWang and C L Qin ldquoRegional differences in the behaviorsof local governments and growth effect of fiscal decentralizationbased on the analysis of judgment hypothesis and testingrdquoManagement World vol 1 pp 9ndash21 2008

[40] J K Brueckner ldquoStrategic interaction among governments anoverview of empirical studiesrdquo International Regional ScienceReview vol 26 no 2 pp 175ndash188 2003

[41] F Revelli ldquoOn spatial public finance empiricsrdquo International Taxand Public Finance vol 12 no 4 pp 475ndash492 2005

[42] N Shen and T Y He ldquoFinancial efficiency and regional eco-nomic growth from the perspective of fiscal decentralizationrdquoReform vol 1 pp 61ndash65 2006

[43] J M Liu X M Hu and J G Wu ldquoThe threshold effect ofsub-national fiscal decentralization revenue and expendituredecentralization on provincial industrial transformation andupgrading based on the data of 14 cities in Hunan ProvincerdquoPublic Finance Research vol 8 pp 49ndash52 2014

Submit your manuscripts athttpwwwhindawicom

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

MathematicsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Mathematical Problems in Engineering

Hindawi Publishing Corporationhttpwwwhindawicom

Differential EquationsInternational Journal of

Volume 2014

Applied MathematicsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Probability and StatisticsHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Mathematical PhysicsAdvances in

Complex AnalysisJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

OptimizationJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

CombinatoricsHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Operations ResearchAdvances in

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Function Spaces

Abstract and Applied AnalysisHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of Mathematics and Mathematical Sciences

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Algebra

Discrete Dynamics in Nature and Society

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Decision SciencesAdvances in

Discrete MathematicsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom

Volume 2014 Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Stochastic AnalysisInternational Journal of

14 Discrete Dynamics in Nature and Society

[36] S S Rosenthal and W C Strange ldquoGeography industrialorganization and agglomerationrdquoTheReview of Economics andStatistics vol 85 no 2 pp 377ndash393 2003

[37] S Park Yoon International Banking and Financial CentersKluwer Academic Boston Mass USA 1989

[38] B Y Qiao J Y Fan and X Y Feng ldquoChinarsquos fiscal decentral-ization and compulsory primary educationrdquo Social Sciences inChina vol 6 pp 37ndash46 2005

[39] W JWang and C L Qin ldquoRegional differences in the behaviorsof local governments and growth effect of fiscal decentralizationbased on the analysis of judgment hypothesis and testingrdquoManagement World vol 1 pp 9ndash21 2008

[40] J K Brueckner ldquoStrategic interaction among governments anoverview of empirical studiesrdquo International Regional ScienceReview vol 26 no 2 pp 175ndash188 2003

[41] F Revelli ldquoOn spatial public finance empiricsrdquo International Taxand Public Finance vol 12 no 4 pp 475ndash492 2005

[42] N Shen and T Y He ldquoFinancial efficiency and regional eco-nomic growth from the perspective of fiscal decentralizationrdquoReform vol 1 pp 61ndash65 2006

[43] J M Liu X M Hu and J G Wu ldquoThe threshold effect ofsub-national fiscal decentralization revenue and expendituredecentralization on provincial industrial transformation andupgrading based on the data of 14 cities in Hunan ProvincerdquoPublic Finance Research vol 8 pp 49ndash52 2014

Submit your manuscripts athttpwwwhindawicom

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

MathematicsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Mathematical Problems in Engineering

Hindawi Publishing Corporationhttpwwwhindawicom

Differential EquationsInternational Journal of

Volume 2014

Applied MathematicsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Probability and StatisticsHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Mathematical PhysicsAdvances in

Complex AnalysisJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

OptimizationJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

CombinatoricsHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Operations ResearchAdvances in

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Function Spaces

Abstract and Applied AnalysisHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of Mathematics and Mathematical Sciences

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Algebra

Discrete Dynamics in Nature and Society

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Decision SciencesAdvances in

Discrete MathematicsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom

Volume 2014 Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Stochastic AnalysisInternational Journal of

Submit your manuscripts athttpwwwhindawicom

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

MathematicsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Mathematical Problems in Engineering

Hindawi Publishing Corporationhttpwwwhindawicom

Differential EquationsInternational Journal of

Volume 2014

Applied MathematicsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Probability and StatisticsHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Mathematical PhysicsAdvances in

Complex AnalysisJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

OptimizationJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

CombinatoricsHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Operations ResearchAdvances in

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Function Spaces

Abstract and Applied AnalysisHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of Mathematics and Mathematical Sciences

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Algebra

Discrete Dynamics in Nature and Society

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Decision SciencesAdvances in

Discrete MathematicsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom

Volume 2014 Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Stochastic AnalysisInternational Journal of