local economydownloaded from lec.sagepub.com at b-on: 01100 universidade do porto on september 6,...

http://lec.sagepub.com/Local Economy

http://lec.sagepub.com/content/29/1-2/141The online version of this article can be found at:

DOI: 10.1177/0269094214522115

2014 29: 141Local EconomyAurora Castro Teixeira and Maria João Barros

Local municipalities' involvement in promoting the internationalisation of SMEs

Published by:

http://www.sagepublications.com

On behalf of:

London South Bank University

Local Economy Policy Unit

Partner Organisation: Centre for Local Economic Strategies

can be found at:Local EconomyAdditional services and information for

http://lec.sagepub.com/cgi/alertsEmail Alerts:

http://lec.sagepub.com/subscriptionsSubscriptions:

http://www.sagepub.com/journalsReprints.navReprints:

http://www.sagepub.com/journalsPermissions.navPermissions:

What is This?

- Feb 18, 2014Version of Record >>

at b-on: 01100 Universidade do Porto on September 6, 2014lec.sagepub.comDownloaded from at b-on: 01100 Universidade do Porto on September 6, 2014lec.sagepub.comDownloaded from

http://lec.sagepub.com/

http://lec.sagepub.com/content/29/1-2/141

http://www.sagepublications.com

http://www.lsbu.ac.uk/index.shtml

http://www1.lsbu.ac.uk/lepu

http://www.cles.org.uk/

http://lec.sagepub.com/cgi/alerts

http://lec.sagepub.com/subscriptions

http://www.sagepub.com/journalsReprints.nav

http://www.sagepub.com/journalsPermissions.nav

http://lec.sagepub.com/content/29/1-2/141.full.pdf

http://online.sagepub.com/site/sphelp/vorhelp.xhtml



Feature

Local municipalities’involvement in promoting theinternationalisation of SMEs

Aurora Castro TeixeiraCEF.UP, Faculdade de Economia, Universidade do Porto, Portugal; INESC Porto, Portugal;

OBEGEF, Portugal

Maria Joao BarrosFaculdade de Economia, Universidade do Porto, Portugal

Abstract

Despite extensive research on decentralisation, the role of local governments in promoting the

internationalisation of firms has been rather neglected in the literature. Based on a sample of

144 Portuguese municipalities, and resorting to logistic econometric estimations, we found that:

(1) the majority of municipalities have been involved in activities to promote economic develop-

ment and the internationalisation of firms; (2) municipalities are essentially involved in the brand-

ing of regions (image building) or in organising fairs and trade missions and (3) municipalities more

active in promoting the internationalisation of small and medium-sized enterprises (SMEs) tend to

be more peripheral, with a relatively high area and population density, higher purchasing power,

higher proportion of population with secondary schooling, lower density entrepreneurial context

but with higher amounts of exports. Although there is still a long way to go for a more profound

and comprehensive decentralisation at this level in Portugal, given the knowledge municipalities

possess about the firms that are located in their vicinity, we contend that it would be desirable

that more decentralised efforts be put towards the implementation of information, and educa-

tion/training-related programmes aiming at promoting SMEs internationalisation.

Keywords

decentralisation, exports promotion, firms, local policy, Portugal

Introduction

Governments have an increasingly activerole in supporting exports by small andmedium-sized enterprises (SMEs) throughpublic policies that promote theirinternationalisation (Gil et al., 2008;Lederman et al., 2010; O’Gorman andEvers, 2011). However, in most countries

export promotion agencies and policieshave traditionally operated at the nationallevel, rather than through a decentralisedmodel (Leonidou et al., 2011). The efforts

Corresponding author:

Aurora Castro Teixeira, Faculdade de Economia,

Universidade do Porto, Rua Dr Roberto Frias, Porto

4200-464, Portugal.

Email: [email protected]

Local Economy

2014, Vol. 29(1–2) 141–162

! The Author(s) 2014

Reprints and permissions:

sagepub.co.uk/journalsPermissions.nav

DOI: 10.1177/0269094214522115

lec.sagepub.com

at b-on: 01100 Universidade do Porto on September 6, 2014lec.sagepub.comDownloaded from


to boost SME internationalisation mightoccur nevertheless at the micro (local autho-rities), meso (national trade bodies andEPAs) and macro-levels (overseas embas-sies, EU, etc.). Given the challenges and spe-cificities that countries face at the regionaland local levels, a multiscalar nature of gov-ernance within economies regarding SMEsexport promotion might be advisable.

Bearing in mind the growing process ofdecentralisation (Azfar et al., 2001; Faguet,2014; Litvack and Seddon, 1999; Marinetto,2003; Taylor, 2007) and the spread of stu-dies focusing on local economic develop-ment in the last 30 years (Barberia andBiderman, 2010; Cox, 2004; DeFilippis,1999; Fineberg, 2013), it becomes importantto assess to what extent the local sphere,through local municipalities, can and/orshould seek affirmation as a fundamentalscale of policy action in promoting localenterprises in the global market.

Notwithstanding the important and fairlywidespread scientific literature on the pro-motion of exports, public policies and pro-grammes in this area, covering their results(Shamsuddoha et al., 2009; Wilkinson andBrouthers, 2006) and the degree of satisfac-tion by the beneficiary firms (Calderon et al.,2005; Cassey, 2008; Gillespie and Riddle,2004), there is a relative lack of studies deal-ing with the local dimension or the decentral-isation of the support provided to theinternationalisation of firms (for a review,see Teixeira and Barros, 2014).

The empirical analysis undertaken in thisstudy therefore aims to contribute to reflec-tions on export promotion and internation-alisation policies at a territorial/local scale,based on the factors that, according to theliterature, can contribute to, or restrict, thedevelopment of such measures/policies atthe local level.

The study focuses on a rather unexploredreality, Portugal, a small European countrythat has received a large amount ofEU funding for the development of its

home-market infrastructures (Pinho andMartins, 2010), where internationalisation,in particular of SMEs, is at the fore of thecountry’s political agenda as a means toovercome the economic crisis (Serra et al.,2012), and where regional decentralisationis still an unresolved issue (Nanetti et al.,2004; Sorens, 2009).

The paper is organised as follows: thefollowing section presents a literaturereview on the determinants of the involve-ment of municipalities in promoting theinternationalisation of enterprises. Then,the study’s methodological considerationsare detailed in the next section and theresults obtained are presented in subsequentsection. Finally, Conclusion summarises themain topics covered in the study, as well asits limitations and pointers for futureresearch.

Determinants of theinvolvement of localmunicipalities in promoting theinternationalisation of firms:A literature review

Public policy interventions are many andvaried (Blackburn and Schaper, 2012).Among these, promoting national exportsand the internationalisation of firms is atop priority of many public policy makers,mainly because internationalisation in gen-eral, and national exports in particular, pro-vide the means to increase employmentopportunities for local people, generate for-eign exchange to finance imports, enrichpublic funds with additional tax revenues,create backward and forward linkages inthe economy, and achieve higher economicgrowth and living standards (Leonidouet al., 2011).

Government-designed internationalisa-tion promotion policies include a varietyof programmes intended to support firms’exporting activity, covering a broad spec-trum of activities (Durmusoglu et al.,

142 Local Economy 29(1–2)



2012; Leonidou et al., 2011; VolpeMartincus and Carballo, 2010): informa-tion-related programmes (provision of mar-keting/information/advice; information onmarket opportunities; general informationabout doing business in a specific country;specific information about doing businesswith a particular firm; export publications– newsletters, how-to-export handbooks);education- and training-related programmes(organisation of export seminars/confer-ences for potential exporters; counsellingand training on the export process for inex-perienced exporters; training on exportdocumentation; foreign language support);trade mobility-related programmes (arrangemeetings with potential customers; assist-ance in participating in trade shows/exhib-itions; participation in trade missionsabroad; support by trade offices abroad)and financial aid-related programmes (co-finance the participation in trade missions,shows and fairs, and organise these events;sponsor the creation of export consortia;export loans; export credit guarantees;funds transferring).

At the macro-level and using a bilateralgravity model, Rose (2007) demonstratedthat the presence of foreign missions, mostnotably associated with embassies and con-sulates, was positively correlated with coun-tries’ exports. At the micro-level, someearlier works (e.g. Cavusgil and Naor,1987; Gencturk and Kotabe, 2001) foundthat government export assistance pro-grammes contributed to firms’ exportsuccess. More recently, it has been demon-strated that experiential activities such astrade shows and trade missions lead tohigher levels of firms’ performance becausethey allowed managers to rapidly acquireinformation about export markets and theprocess of exporting (Wilkinson andBrouthers, 2006). Trade shows constitutean important promotional tool and can pro-vide positive economic benefits to the firm,generating both immediate sales and

product awareness (Gopalakrishna et al.,1995). Through trade shows, representativesof companies which are export ready cangain customers, disseminate information,identify prospects, gather intelligence andreinforce firm morale (Wilkinson andBrouthers, 2006).

In most countries, export promotionpolicies have traditionally operated at thenational level rather than through a decen-tralised model (Lederman et al., 2010;Leonidou et al., 2011). However, givenfirms and regions’ specificities there mightexist benefits from involving local publicauthorities in the promotion of the inter-nationalisation of their firms, that is, froma more decentralised implementation ofexport promotion policies (Teixeira andBarros, 2014).

In a centralised system, politicians makedecisions usually aimed at reflecting thecountry’s interests (Balaguer-Coll et al.,2010; Rodrıguez-Pose and Gill, 2005).Nonetheless, this practice may proverather inefficient when interests differamong regions, since some regions do notbenefit from those national policies. If pref-erences vary across regions, it would bemore efficient to geographically alter theprovision of public services. Thus, the pro-vision of public services by the public sectorcould be more efficient within a decentra-lised government structure (Balaguer-Collet al., 2010). This might be the case ofexport promotion policies, since local muni-cipalities can have branches of economicactivity that vary across regions, andwhich act as competitive advantages.

Barberia and Biderman (2010) maintainthat initiatives should be executed and man-aged at the territorial and local levels andseveral authors recognise that the mainbenefits of decentralisation are: greater agil-ity, competitiveness and flexibility to adaptto changes (Lobao and Kraybill, 2009;Oates, 1999; Taylor, 2007); the creation ofa geographical focus at the local level,

Teixeira and Barros 143



coordinating national, state, district andlocal programmes more effectively (Litvackand Seddon, 1999); and the formulation ofmore creative, innovating and appropriateprogrammes that enable local experimenta-tion (Litvack and Seddon, 1999). It cantherefore be assumed that decentralisingexport promotion to a local scale, specific-ally, to the local municipalities, could resultin a continuing process whose final impactwould ultimately be the higher efficiencyand higher effectiveness of these policiesand, consequently, a higher degree of bene-fit for the targeted local enterprises.

Closer scrutiny shows that the benefitsattributed to decentralisation reside mostlyin the local governments’ greater respon-siveness to local needs (Lobao andKraybill, 2009; Oates, 1999), adapting poli-cies to the preferences of smaller and morehomogeneous groups (Balaguer-Coll et al.,2010; Lobao and Kraybill, 2009; Tiebout,1956; Wallis and Oates, 1988), or in thebetter ability of governments to accommo-date differences in tastes for public goodsand services (Balaguer-Coll et al., 2010;Oates, 1972; Tiebout, 1956), all factorsthat justify decentralisation from the view-point of economic efficiency.

In short, the considerations of efficiencyon which decentralisation discourses arebased (Balaguer-Coll et al., 2010) also com-prise this study’s main line of reasoning,that is, greater efficiency is the main argu-ment in favour of decentralising export pro-motion policies to the local municipalities.

Decentralisation might neverthelessinvolve possible losses. The argumentshere are directly linked to the local scaleitself, such as lack of administrative or tech-nical capacity, or even the transfer ofauthority to individuals who have limitedexperience in management and, in somecases, little interest in taking those respon-sibilities (Andersson et al., 2006; Balaguer-Coll et al., 2010; Chapman et al., 2002;Faguet, 2004); less efficient and effective

services provision (Agrawal and Ribot,1999; Andersson et al., 2006; Azfar et al.,2001; Boone, 2003; Litvack and Seddon,1999); the transfer of responsibilities to thelocal level without adequate financialresources can make the equitable distribu-tion and provision of services more difficult(Agrawal and Ribot, 1999; Andersson et al.,2006; Balaguer-Coll et al., 2010; Boone,2003; Cox, 2004; Faguet, 2004; Litvackand Seddon, 1999) and agents mayback away from new strategies they do notfully understand, perpetuating conservatismwithin the communities and stifling effortsfor improvement (Chapman, 2000;Chapman et al., 1997, 2002).

These arguments are in line with Litvackand Seddon (1999) regarding the import-ance of the organisations’ profile and char-acteristics, implying that there should bebasic knowledge of the strengths and weak-nesses of organisations in the performanceof various types of functions, since the suc-cess of decentralisation depends on thesecharacteristics and also on the appropriatepreparation of the agents of decentralisedadministration (Litvack and Seddon,1999). Teixeira and Barros (2014) furtherhighlight an emergent generation of studiesabout decentralisation that focuses not onlyon the scale of provision and the type ofservice but also on the fundamental natureof the policies and institutions. This emer-gent literature shows that it is a complexmix of institutions that generates receptivelocal agents.

What will ultimately define the finalresult of decentralisation is, besides specificfactors, the interaction between the type ofdecentralisation and the conditions underwhich it takes place. To simplify, the condi-tions that influence the success of a decen-tralisation process, that is, the increasinginvolvement of local entities, can be dividedinto two groups: the attributes of the localmunicipalities (the extent to which localauthorities are motivated to support the




process and the availability of financial andtechnical resources) and structural vari-ables, such as the type and magnitude ofthe resources possessed, the local powerrelationships and the characteristics of thelocal economy (Pacheco, 2004; Teixeira andSilva, 2012).

Indeed, several authors agree that decen-tralisation works differently depending onthe types of powers that are decentralised(Andersson et al., 2006; Litvack et al.,1998; Ribot, 2002; Rondinelli et al., 1989).Others sustain that decentralisation canwork, but only in the context of specificinstitutions, which include mechanisms ofaccountability, supervision and transfer ofresources (Agrawal and Ribot, 1999;Andersson et al., 2006; Blair, 2000;Faguet, 2014; Gibson and Lehoucq, 2003;Joanis, 2014).

It is increasingly acknowledged that fordecentralisation to achieve the potentialbenefits of an efficient and equitable provi-sion of public goods, citizens need to havethe means to send appropriate informationto local actors, so that local politicians canrespond appropriately or be held account-able when this does not occur. It is claimedthat for these conditions to exist, severalinstitutional and social characteristics haveto be assembled. More specifically, theincentives for local politicians to respondto their constituents’ demands are under-stood as being conditioned by institutionalincentives within the framework of nationalpolicy, by constraints from the localpolitical system, and by the formal repre-sentation and articulation of the citizens’preferences in the political structure(Kauneckis and Andersson, 2009).

In the model by Kauneckis andAndersson (2009), formal political institu-tions and the structure of the local societygenerate several incentives and constraintsto the municipalities’ action. Thus, thestructure of local political action is concep-tualised as comprising two levels: the impact

of local political institutions at the nationallevel and the influence of the local munici-palities’ specific institutional and socioeco-nomic characteristics.

In this context, analysing under whichcircumstances decentralisation is moreeffective places the emphasis not on themerits of decentralisation (as opposed tocentralisation) but rather on the mannerand conditions in which it is undertaken.Theoretical premises suggest that decentral-isation depends on institutional regulationsand their interaction with social practices,influencing the achievement of decentralisedgovernance (Kamoto et al., 2013). Thesefactors, according to Azfar et al. (2001),include the distribution of powers amonglevels of government (central governmentsupervision over local government oper-ations), the discipline operating fromwithin and outside government (manage-ment of the involved staff) and the princi-pal-agent information flows (ability for allagents to participate in the decision-makingprocess), which is intimately related to theregion’s human capital. In other words,Agrawal and Ribot (1999) maintain thatthe relationship between decentralisationand its results can be better understood ifit is analysed in terms of actors, powers andaccountability, which makes it relevant toanalyse the relationships between the cen-tral government and local governmentsand between these and local populations(De Vries, 2012; Pacheco, 2004; Ribot,1999, 2001).

Summarising, the key determinants ofthe involvement of local municipalities inactivities that provide support to economicdevelopment and the internationalisation offirms include: (1) the municipalities’ charac-teristics, namely the municipal budget percapita and the number of employees; (2)the municipalities’ development indicators,including telecommunications, distance tothe capital, area, population density, popu-lation with secondary and higher education,




purchasing power and unemployment rateand (3) the municipalities’ characteristicsof entrepreneurial activity, such as thenumber of enterprises, volume of exports,weight of medium and large enterprisesand tertiary educated labour.

Bearing in mind that the formal mechan-isms of government are important only in theway they interact with the local conditions,particular attention should be put on thelocal demand structure regarding the muni-cipalities’ characteristics and not merely onthe analysis of the relations between the cen-tral government and the local governments.In this context, according to Kauneckis andAndersson (2009), the characteristics of amunicipality that most influence the qualityof local service provision are the size of thearea the municipality has to serve, its popu-lation (education and literacy, populationdensity and percentage of population whichrequires that service) and its financial cap-acity. According to the same authors, thelarger the municipal area, the greater is thedifficulty in providing adequate quality ser-vices. Additionally, a more educated popu-lation, clustered in population centres withlow demand for the services under studyshould be easier to satisfy than low-densityclusters of population with low literacy ratesand a high demand for these services(Kauneckis and Andersson, 2009;Rowland, 2001). On the other hand, Lobaoand Kraybill (2009) concluded that localitieswith more population and less educated citi-zens get more involved in activities to pro-mote local economic development. Finally,local municipal governments are restrictedby their financial capacity to respond tolocal demand. In some cases, this relates toa flaw in devolution to local governments; inothers, it may simply reflect the relativevigour of a local economy. The municipalbudget is, therefore, used as a measure ofthe local governments’ ability to respond totheir citizens’ demand (Kauneckis andAndersson, 2009). Equally, to question the

potential efficiency and effectiveness of pro-moting exports at the local level implies theneed to examine the relationships betweenlocal municipalities and enterprises. Bygoing beyond the main arguments on thefactors influencing a process of decentralisa-tion, a healthy export sector is considered ofextreme importance to the nation, states andcommunities (Lewis, 1990). Additionally,the number of employees assigned to a div-ision may serve to indicate the commitmentof a state or municipality to develop exports(Lesch et al., 1990) and the local govern-ment’s capacity (measured in resources andnetwork) is an important determinant of theactivities pursued to promote local economicdevelopment (Lobao and Kraybill, 2009).Finally, business competitiveness withother localities pressures governments toattract businesses (Lobao and Kraybill,2009) and, by entering new markets, regionsmay increase their economic potential,instead of simply competing with eachother in the existing markets (Stagg, 1990).

Methodological considerations

The aim of this study is to assess the deter-minants underlying the involvement of thelocal sphere, most specifically the municipa-lities, in the promotion and support of localfirms’ internationalisation processes.

The municipalities’ sensibility to mattersof economic development and the inter-nationalisation of firms might be analysedconsidering three dimensions of municipali-ties’ involvement: (1) the municipality hasstaff assigned to this area (variable ‘staff’);(2) the municipality establishes contact withenterprises (variable ‘contacts’) and (3) themunicipality carries out the activities understudy (variable ‘activities’).

Based on the literature review conductedin the previous section, the determinants ofthe municipalities involvement are dividedinto three groups (see Table 1): the munici-palities’ institutional characteristics (human




Tab

le1.

Pro

xie

sfo

rth

eke

yva

riab

les

of

the

‘theore

tica

l’m

odel.

Dete

rmin

ants

Pro

xy

Auth

ors

Sourc

e

Var

iable

sch

arac

teri

sing

the

org

anis

atio

n

Finan

cial

reso

urc

es

Munic

ipal

budge

tper

capita

(E)a

Kau

neck

isan

dA

nders

son

(2009),

Lobao

and

Kra

ybill

(2009)

Munic

ipal

itie

s’in

stitutional

websi

tes

Hum

anre

sourc

es

Num

ber

of

munic

ipal

em

plo

yees

[1994]

Lesc

het

al.(1

990)

Sale

sIn

dex

2010

Var

iable

sch

arac

teri

sing

the

regi

on’s

degr

ee

of

dev

elo

pm

ent

Infr

astr

uct

ure

san

d

acce

ssib

ilities

Tele

phone

connect

ions

(thousa

nd

hab

itan

ts)

[2009]

Lobao

and

Kra

ybill

(2009)

INE

Dis

tance

toth

eca

pital

(km

)w

ww

.via

mic

helin

.pt

Geogr

aphic

alco

nte

xt

Are

a(k

m2)

Kau

neck

isan

dA

nders

son

(2009),

Lobao

and

Kra

ybill

,

(2009)

INE

Popula

tion

densi

ty[2

009]

Row

land

(2001),

Kau

neck

isan

d

Anders

son

(2009),

Lobao

and

Kra

ybill

(2009)

Hum

anca

pital

Popula

tion

with

com

ple

ted

seco

ndar

yeduca

tion

b

[2001/2

009]

Popula

tion

with

com

ple

ted

hig

her

educa

tion

c[2

001/2

009]

Eco

nom

icco

nte

xt

Purc

has

ing

pow

er(E

)[2

007]

Kau

neck

isan

dA

nders

son

(2009)

Unem

plo

yment

rate

[2009]

Var

iable

sch

arac

teri

sing

the

entr

epre

neuri

al

activi

ty

Entr

epre

neuri

alac

tivi

tyExport

s(t

housa

nd

euro

s)[2

008]

Lew

is(1

990)

Sale

sIn

dex

2010

Weig

ht

of

mediu

man

dla

rge

firm

sd[2

007]

Num

ber

of

firm

s[2

007]

Sect

or

speci

alis

atio

nW

eig

ht

of

man

ufa

cturi

ng

sect

ore

[2005]

INE

Sour

ce:A

uth

ors

’co

mpila

tion.

Not

es:IN

E:N

atio

nal

Stat

istics

Inst

itute

;N

UT

S:N

om

encl

ature

of

Terr

itori

alU

nits

for

Stat

istics

;(u

nit

of

meas

ure

),[y

ear

].aW

hen

we

could

not

find

the

munic

ipal

budge

tfo

r2011,w

euse

dth

eva

lue

for

2010

or,

inso

me

case

s,fo

r2009.

bIn

perc

enta

ge,ca

lcula

ted

for

the

tota

lpopula

tion,dis

rega

rdin

gsc

hoolag

e.

cIn

perc

enta

ge,ca

lcula

ted

for

the

tota

lpopula

tion,dis

rega

rdin

gsc

hoolag

e.

dIn

perc

enta

ge,co

veri

ng

ente

rpri

ses

with

atle

ast

50

em

plo

yees.

eIn

perc

enta

geof

the

tota

lw

ork

forc

e,

by

NU

TS

II.




and financial resources), extent of the muni-cipalities’ characteristics (infrastructuresand accessibilities, geographical context,human capital and economic context) andthe characteristics of the local enterprises(entrepreneurial activity and humancapital).

In order to operationalise the explana-tory variables that are likely to influencethe involvement of Portuguese municipali-ties in activities to promote economicdevelopment/the internationalisation offirms, we gathered primary and secondarydata, this latter from the Sales Index 2010(Marketest), the National Statistics Institute(INE), the City Halls’ websites and theMichelin website.

The primary data gathering involved theimplementation of a survey to the 308Portuguese municipalities, to see whetherthey engaged in activities to promote localeconomic development and the internation-alisation of local firms. The survey wascomposed of three sections. The firstgroup of questions was intended to addressthe municipalities’ sensibility to the issue ofpromoting economic development/the inter-nationalisation of firms. It enabled us todetermine whether there was a structure(formal or informal) in the municipality

assigned to these actions, its dimension(number of staff members), the level ofqualification of the involved employeesand their capacity for autonomy. Thesecond section addressed the level of themunicipalities’ intervention in this areaand was intended to assess the ways inwhich City Halls relate to local firms (and,when that contact exists, gain insights intoits nature). The last group of questionsfocused on the municipalities’ proactivityand the performance of the activities stu-died, so as to determine the inherent factorsthat influence, positively or negatively, theirinvolvement in such actions.

The first phase of the study began on23rd February 2011 and ended on 1stApril 2011. From a population of 308Portuguese municipalities, we received atotal of 144 answers, that is, a responserate of 46.8%, where the municipalities ofthe North region and those from theAutonomous Region of Madeira were themost and less receptive to this study, witha response rate of 55.8% and 27.3%,respectively (see Table 2).

Given the non-compulsory nature ofthe survey, the overall response rate byNomenclature of Territorial Units forStatistics (NUTS) II was quite acceptable

Table 2. Respondent municipalities, by geographical region.

Statistic

sub-regions

(NUTS II)

Population:

No. of municipalities

[% total]

Sample: No. of answers

[% total]

Representativeness

in the study:

response rate (%)

North 86 [27.9] 48 [33.3] 55.8

Centre 100 [32.5] 44 [30.6] 44.0

Lisbon 18 [5.8] 7 [4.9] 38.9

Alentejo 58 [18.8] 26 [18.1] 44.8

Algarve 16 [5.2] 6 [4.2] 37.5

Azores 19 [6.2] 10 [6.9] 52.6

Madeira 11 [3.6] 3 [2.1] 27.3

Total 308 [100] 144 [100] 46.8

Source: Authors’ compilation, based on the answers obtained to the survey applied to Municipalities.

Note: NUTS: Nomenclature of Territorial Units for Statistics.




(Chang et al., 2012). Thus, excluding thecase of the Autonomous Region ofMadeira, the sample obtained is representa-tive of the population under study, enablinga rigorous statistical analysis, with the pos-sibility of extrapolating the results to thepopulation.

The binary nature of the variable toexplain variable (municipalities’ involve-ment) restricts the choice of the estimationmodel, with the logistic regression being themost appropriate regression technique tomodel the occurrence (Greene, 2012).1

Empirical results

The majority of municipalities revealed thatthey are involved in activities to promoteeconomic development/the internationalisa-tion of firms (e.g. region’s branding, exportsupport services, trade missions, fairs,market search and publications). Althoughthere is a positive linear association betweenthe three variables that measure the involve-ment of the municipality in export anddevelopment promotion, only the correl-ation coefficient estimates between‘activities’ and ‘staff’, and ‘activities’ and‘contact’, are significant (see Table 3).Thus, municipalities that perform activitiesto promote economic development/inter-nationalisation tend to a larger extent than

those that do not possess personnel assignedto activities to promote such activities andto maintain contacts with local enterprises.Such correlation however is far from per-fect. Indeed, in geographical terms (NUTSIII) we observe that not having staff mem-bers specifically assigned to the promotionof economic development/internationalisa-tion (variable ‘staff’) does not imply thenon-performance of such activities (variable‘activities’) and vice versa (see Figure 1).

As only very small number of municipa-lities has no contact with firms, it was irrele-vant to statistically and econometricallyexplore this dimension of the municipalities’involvement.

At the structural level, we observe thatthree-fourths of the municipalities formallypursue activities to promote economicdevelopment/internationalisation throughan organic unit or division speciallyassigned to that effect defined in the macro-structure. The staff members who work inthis area are qualified and possess a consid-erable level of independence with regard toprocesses of bureaucracy and hierarchy.

Regarding the contacts established withlocal enterprises, and according to the cate-gories of services provided by export pro-motion entities (Lederman et al., 2010),there is a predominance of activities toboost the region’s image building and mar-keting, including fairs and trade missions(at 77.4% and 76.6%, respectively), fol-lowed by market research and publications,and export support services (technicalassistance, seminars), at 38.0% and 29.2%,respectively. Other activities indicated by amuch small number of municipalitiesencompass information and counselling onlegislation and public support mechanismsto enterprises, support to entrepreneurshipand local and regional innovation (some-times associated with start-up centres),financial incentives, local financing mechan-isms, training, visits and tributes to enter-prises, and trade missions to take advantage

Table 3. The municipalities’ involvement in activ-

ities to promote economic development/the inter-

nationalisation of firms – correlation between the

variables ‘staff’, ‘contacts’ and ‘activities’.

Staff Contacts Activities

Staff 1

Contacts 0.103 1

Activities 0.400*** 0.174** 1

Note: **statistically significant at 5% and ***statistically

significant at 1%.




of twinning agreements with other local-ities/regions.

In an attempt to understand why morethan one-third of the local municipalitiesare not involved in activities to promoteeconomic development/internationalisationof firms, the factors identified by localmunicipalities as enhancers or inhibitors tothe performance of such activities wereanalysed.

Within the group of municipalitiesinvolved in these activities, the factors theymost frequently indicated as apparentlyincreasing their involvement included theimportance of local businesses, competi-tiveness in relation to other municipalitiesand the fact that they are requested byfirms to act in this area (with 93.6%,

55.3% and 33.0%, respectively). Thesewere followed by the use of specific localinformation and specific technical or admin-istrative capacity. The ‘sufficient funds’option had no expression (see Figure 2).Among the other factors mentioned wereconcerns about improving the local qual-ity of life, promotion of local economicdevelopment (affirming the local brandand leveraging local resources), EUfunding opportunities and local incentiveschemes.

As for the municipalities which do notperform activities to promote economicdevelopment/internationalisation, the factorthat was most often referred to as hinderingthis action was the lack of sufficient financialresources (54.0% of the cases). This was

Staff

0%–20% 21%–40% 41%–60% 61%–80% 81%–100%

Contacts Activities

Figure 1. Dimension of the municipalities’ involvement in activities to promote economic development/

the internationalisation of firms – ‘staff’, ‘contacts’ and ‘activities’, by NUTS III, in Mainland Portugal, in

percentage of the municipalities where they exist.

Source: Authors’ compilation, based on the answers to the survey applied to Municipalities.




followed by administrative or technical limi-tations and/or shortcomings (32.0%), thefact that local businesses do not justifythis involvement (26.0%), and focuson other benefits (22.0%). The over-centralisation and lack of interest in thematter were highlighted by a fewlocal authorities, whereas the ‘lack of con-tact with local companies’ had no expres-sion (see Figure 3). Other factors pointedout included the existence of an activetrade association in the area, not havingreceived such requests, and the fact thatthe organic division in question isrecent (which was the case of three of therespondents).

The results corroborate the argumentsagainst decentralisation described in theDeterminants of the involvement of localmunicipalities in promoting the internation-alisation of firms: A literature review sec-tion, namely with regard to the constraintsderiving from administrative or technicalshortcomings at the local level (Andersson

et al., 2006; Balaguer-Coll et al., 2010;Chapman et al., 2002; Faguet, 2004) andthe lack of adequate financial resources(Agrawal and Ribot, 1999; Anderssonet al., 2006; Balaguer-Coll et al., 2010;Boone, 2003; Cox, 2004; Faguet, 2004;Litvack and Seddon, 1999). Broadlyspeaking, these factors sustain the notionthat the organisations’ characteristicsdetermine the quality of the service theyprovide (Litvack and Seddon, 1999;Rodden, 2003).

The descriptive results based on the cor-relation between the relevant variables (seeTable 4) show that the existence of staffassigned to the promotion of economicdevelopment/the internationalisation offirms is positively and significantly relatedto the volume of exports and the numberof local enterprises. This means that,based on bivariate correlations, municipali-ties with a larger volume of exports andfirms tend, on average, to involve morehuman resources in activities related to the

55

47

33

22

94

00

10

20

30

40

50

60

70

80

90

100

Compe��vidade face a outros municípios

Solicitação por parte das empresas

Aproveitamento de informação local específica

Capacidade administra�va ou técnica específica

Importância do tecido empresarial local

Suficientes recursos financeiros

Competitiveness in relation to other

municipalities

Requested by enterprises

Use of specific local information

Specific administrative or technical capacity

Importance of local businesses

Sufficient funds

Figure 2. Importance of the factors which favour the respondent municipalities’ involvement in activities

to promote economic development/internationalisation of firms, in (in % total).





promotion of economic development/theinternationalisation of firms.

With regard to the relation between theindependent variables, the results show thatthere is a high correlation between thenumber of municipal employees and mostof the other variables, which also occurswith telephone connections. There are alsostrong bivariate correlations among otherindependent variables, such as the purchas-ing power and the human capital proxies.Potential problems of multicollinearitythat may arise, which recommend the esti-mation of three separate models in the caseof municipalities’ involvement proxied bythe variable ‘Activities’.

The group of municipalities which areinvolved in activities to promote economicdevelopment/internationalisation of firms,by either assigning human resources (vari-able ‘staff’) or performing concrete activities(variable ‘activities’), present higher means,than municipalities which do not, in termsof number of municipal employees, area,

population with higher education andexports. Additionally, the differences inmeans of Kruskal–Wallis non-parametrictest shows that municipalities which assignhuman resources to activities in economicdevelopment/internationalisation are char-acterised by being more ‘accessible’ in geo-graphical terms (i.e. are closer to thecapital), have more population density, ahigher percentage of people with secondaryeducation, higher purchasing power, aremore industrialised and have more enter-prises, mostly micro and small enterprises(see Table 5).

According to the goodness of fit meas-ures (Hosmer–Lemeshow test and percent-age of correct classifications) (see Table 6),the estimated models prove to be adequate.2

All the group of determinants – organ-isation, region’s development, entrepreneur-ial activity – are relevant for explainingthe involvement of the municipalities inthe promotion of SMEs internationalisa-tion. Individually considered, the region’s

32

54

0

14

22

26

2

0

10

20

30

40

50

60

Limitações e/ou incapacidade administra�va ou técnica

Insuficientes recursos financeiros

Ausência de contacto com empresas locais

Excessiva centralização por parte do Estado

Aposta em outras vantagens do Município

Tecido empresarial do Município não o jus�fica

Falta de interesse pela área

Administrative or technical limitations

and/or shortcomings

Lack of suf f icient f inancial resources

Lack of contact with local companies

Over-centralization

Focus on other benef its

Local businesses do not justify it

Lack of interest in the area

Figure 3. Importance of the factors which restrict the respondent municipalities’ involvement in activ-

ities to promote economic development/internationalisation of firms (in % total).





Tab

le4.

Corr

ela

tion

mat

rix.

AB

12

3a

3b

4a

4b

5a

5b

6a

6b

7a

7b

7c

8

A.D

ependent

vari

able

‘sta

ff’

1

B.D

ependent

vari

able

‘act

ivitie

s’0.4

62

***

1

1.M

unic

ipal

budge

tper

capita

�0.0

04

0.1

03

1

2.N

um

ber

of

munic

ipal

em

plo

yees

0.2

27*

0.0

46�

0.2

89

**1

3a.

Tele

phone

connect

ions

�0.1

59�

0.0

89

0.3

21

***�

0.0

73

1

3b.D

ista

nce

toth

eca

pital

�0.1

50�

0.0

37

0.1

61�

0.3

83

***�

0.0

33

1

4a.

Are

a0.0

40

0.0

06

0.1

24

0.2

06*

0.3

19

***�

0.0

17

1

4b.Popula

tion

densi

ty0.1

71

0.0

36�

0.4

90

***

0.4

79

***�

0.5

15

***�

0.1

29�

0.6

64

***

1

5a.

Popula

tion

with

seco

ndar

yeduca

tion

0.0

91

0.1

69�

0.0

87

0.5

97

***

0.2

31*�

0.5

97

***�

0.0

04

0.2

43

**1

5b.Popula

tion

with

hig

her

educa

tion

0.0

23

0.0

73�

0.1

34

0.6

17

***

0.2

74

**�

0.3

03

**0.0

78

0.2

57

**0.7

22

***

1

6a.

Purc

has

ing

pow

er

0.2

22*

0.2

02*�

0.1

33

0.6

99

***

0.2

12*�

0.4

84

***

0.0

26

0.3

31

***

0.8

41

***

0.8

06

***

1

6b.U

nem

plo

yment

rate

0.0

19

0.0

23�

0.0

99

0.2

07*�

0.1

32

0.1

76�

0.2

32*

0.3

39

***

0.0

09�

0.0

57

0.0

18

1

7a.

Export

s0.3

20

***

0.1

99*�

0.3

65

***

0.5

89

***�

0.3

28

***�

0.2

21*�

0.1

45

0.6

28

***

0.2

90

**0.3

87

***

0.5

16

***

0.0

15

1

7b.W

eig

ht

of

mediu

man

dla

rge

ente

rpri

ses

0.0

93

0.0

65�

0.0

033

0.0

07�

0.2

44

**0.1

99*�

0.2

50

**0.2

59

**�

0.1

71�

0.2

11*

0.0

35

0.2

19*

0.3

09

***

1

7c.

Num

ber

of

ente

rpri

ses

0.2

69

**0.0

46�

0.5

09

***

0.8

73

***�

0.2

63

**�

0.2

16*

0.1

03

0.6

62

***

0.4

08

***

0.5

15

***

0.5

71

***

0.1

74

0.7

00

***

0.0

67

1

8.W

eig

ht

of

man

ufa

cturi

ng

0.1

95�

0.1

44�

0.4

01

***

0.0

81�

0.4

82

***

0.2

11*�

0.3

15

***

0.4

94

***�

0.3

10

***�

0.2

85

**�

0.1

83

0.3

11

***

0.3

37

***

0.4

94

***

0.2

96

**1

Not

e:T

he

grey

cells

repre

sent

stat

istica

llysi

gnifi

cant

est

imat

es

(i.e

.p-v

alue<

0.1

0).

*Sta

tist

ical

lysi

gnifi

cant

at10%

,**

stat

istica

llysi

gnifi

cant

at5%

and

***st

atis

tica

llysi

gnifi

cant

at1%

.




Tab

le5.

Diff

ere

nce

sin

mean

sin

the

vari

able

sin

anal

ysis

(Kru

skal

–W

allis

non-p

aram

etr

icte

st).

Prox

y

Munic

ipal

itie

s’in

volv

em

ent

‘Sta

ff’

‘Act

ivitie

s’

Yes

No.

Kru

skal

–W

allis

test

(p-v

alue)

Yes

No

Kru

skal

–W

allis

test

(p-v

alue)

Munic

ipal

budge

tper

capita

1553

1959

0.1

42

1751

1707

0.2

21

Num

ber

of

munic

ipal

em

plo

yees

410

182

0.0

00

348

206

0.0

33

Tele

phone

connect

ions

26.0

27.6

0.1

53

26.3

27.5

0.4

56

Dis

tance

toth

eca

pital

249.8

486.5

0.0

00

307.8

468.6

0.1

75

Are

a348.1

272.0

0.0

44

340.2

257.9

0.0

17

Popula

tion

densi

ty423.1

170.4

0.0

93

293.2

316.0

0.6

94

Popula

tion

with

com

ple

ted

seco

ndar

yeduca

tion

5.3

4.4

0.0

01

5.1

4.5

0.1

82

Popula

tion

with

com

ple

ted

hig

her

educa

tion

3.9

3.3

0.0

42

3.8

3.2

0.0

73

Purc

has

ing

pow

er

81.0

66.7

0.0

00

76.6

69.5

0.3

63

Unem

plo

yment

rate

7.0

6.8

0.5

35

6.9

7.1

0.5

89

Export

s83971.6

19057.8

0.0

02

75585.6

22047.2

0.0

62

Weig

ht

of

mediu

man

dla

rge

firm

s0.0

054

0.0

094

0.0

31

0.0

050

0.0

118

0.5

15

Num

ber

of

firm

s5405.8

2045.8

0.0

00

4371.6

2692.6

0.1

56

Weig

ht

of

man

ufa

cturi

ng

31.1

26.3

0.0

26

28.0

30.4

0.2

92

Sour

ce:A

uth

ors

’co

mpila

tion,bas

ed

on

the

answ

ers

toth

esu

rvey

applie

dto

Munic

ipal

itie

s.

Not

e:T

he

grey

cells

repre

sent

stat

istica

llysi

gnifi

cant

est

imat

es

(i.e

.p-v

alue<

0.1

0).




Tab

le6.

The

munic

ipal

itie

s’in

volv

em

ent

inac

tivi

ties

topro

mote

eco

nom

icdev

elo

pm

ent/

inte

rnat

ional

isat

ion

of

firm

s:lo

gist

icre

gress

ion.

Dete

rmin

ants

Pro

xy

‘Sta

ff’

‘Act

ivitie

s’

Var

iable

sch

arac

teri

sing

the

org

anis

atio

n

Finan

cial

reso

urc

es

Munic

ipal

budge

tper

capita

(dum

my,

1if

above

mean

)

1.5

61

*0.5

72

0.6

07

0.5

76

Var

iable

sch

arac

teri

sing

the

regi

on’s

degr

eeofdev

elo

pm

ent

Infr

astr

uct

ure

san

dac

cess

ibili

ties

Dis

tance

toca

pital

(ln)

�0.3

11

1.1

47

**0.8

60

*0.7

71

Are

a(ln)

1.6

15

5.8

33

*2.2

13

5.0

78

*

Popula

tion

densi

ty(ln)

1.4

43

5.5

16

*1.8

90

4.7

55

*

Hum

anca

pital

Pro

port

ion

of

popula

tion

with

com

ple

ted

seco

ndar

y

educa

tion

�0.3

46

0.3

03

0.6

37

*

Pro

port

ion

of

popula

tion

with

com

ple

ted

hig

her

educa

tion

�0.6

54

**�

0.5

90

*�

0.2

63

Regi

on’s

eco

nom

icdev

elo

pm

ent

Purc

has

ing

pow

er

(ln)

8.0

27

**8.6

38

**5.9

26

**

Unem

plo

yment

rate

�0.0

58

0.0

08

0.1

05

0.0

61

Var

iable

sch

arac

teri

sing

the

entr

epre

neuri

alac

tivi

ty

Entr

epre

neuri

alac

tivi

ty

Export

s(ln)

0.2

12

0.5

81

**0.6

48

**0.4

92

*

Weig

ht

of

mediu

man

dla

rge

firm

s

�144.2

38

*�

42.7

70�

20.8

26�

10.0

71

Num

ber

of

ente

rpri

ses

(ln)

�1.0

76

�6.4

21

**�

2.4

60

�5.6

18

**

Sect

or

speci

alis

atio

nW

eig

ht

of

man

ufa

cturi

ng

0.0

51

�0.0

31

�0.0

40

�0.0

24

Const

ant

�38.0

02�

53.3

51�

12.9

24�

38.8

80

Obse

rvat

ions

Num

ber

of

obse

rvat

ions

(N)

71

71

71

71

Munic

ipal

itie

sw

ith

Staf

f/A

ctiv

itie

s45

51

51

51

Oth

er

munic

ipal

itie

s26

20

20

20

Godness

of

fitG

oodness

of

fit

Hosm

er–

Lem

esh

ow

test

(p-v

alue)

0.8

90

0.6

55

0.1

49

0.2

27

Perc

enta

geof

corr

ect

clas

sific

atio

ns

74.6

77.5

76.1

76.1

Not

e:T

he

grey

cells

repre

sent

stat

istica

llysi

gnifi

cant

est

imat

es

(i.e

.p-v

alue<

0.1

0).

*Sta

tist

ical

lysi

gnifi

cant

at10%

;**

stat

istica

llysi

gnifi

cant

at5%

.




level of development (measured by its percapita purchasing power) and international-isation characteristics (measured by the sizeof its exports) emerged as the most criticaldeterminants for the quest for decentralisa-tion in what SMEs’ promotion policies andmeasures are concerned.

As would be expected, and in line withKauneckis and Andersson (2009), munici-palities with a higher municipal budget percapita tend, on average, ceteris paribus, tobe associated with a higher involvement inthe promotion of economic developmentand SMEs internationalisation.

Municipalities characterised by lowweight of medium and large firms, low dens-ity of firms and low share of highly edu-cated (i.e. citizens with post-secondary(‘higher’) education) are associated with ahigher level of involvement in the promo-tion of economic development and SMEsinternationalisation. This seems to indicatethat the municipalities’ involvement may tosome extent constitute an attempt to ‘com-pensate’/overcome the difficulties and obs-tacles raised by a lack of preparation of theregion’s agents in terms of human and otherresources (e.g. organisational, financial) toface internationalisation processes. Suchevidence reinforces the argument in favourof increasing decentralisation of public poli-cies and measures in the area of SMEsinternationalisation.

Accessibilities and geographical contextemerge as (much more) relevant in explain-ing the municipalities’ involvement in the‘activities’ for promotion of SMEs inter-nationalisation. In fact, all else remainingconstant, on average, larger municipalities,with relatively higher population densitiesand more distant from the capital tend todecentralise activities to promote economicdevelopment/internationalisation of firms,replacing (or complementing) the action ofcentral government institutions. The argu-ments of Lobao and Kraybill (2009) arealso corroborated, in that municipalities

with a larger area and population densityand with more educated citizens (in thiscase, at the secondary education level)tend to perform more activities to promoteeconomic development/the internationalisa-tion of firms. Thus, some population/area/human capital threshold might be requiredto justify the decentralisation of suchactivities.

At the level of entrepreneurial activity,and as has been mentioned in the literature(see Determinants of the involvement oflocal municipalities in promoting the inter-nationalisation of firms: A literature reviewsection), the higher the volume of exports,the higher, on average, is the municipalities’propensity to perform activities to promoteeconomic development/SMEs internation-alisation. This fact is in line with Lewis(1990) regarding the extreme importanceof exports in this context. On the contrary,and regardless of their industrial specialisa-tion, municipalities with fewer enterprisesand lower proportion of medium and highfirms tend, on average, to be more proactivein this area, assigning more staff and per-forming more of such activities. This evi-dence might be rationalised, following thearguments of Lewis (1990) and Lesch et al.(1990), in light of the idea that, in the pres-ence of underdeveloped local businesses,municipalities will attempt to encourageentrepreneurial and exporting activities,acting as a supporter for local agents.

Conclusion

This study analysed whether local munici-palities could be considered as valid decen-tralised agents to promote economicdevelopment and the internationalisationof firms. With this in mind, we approachedthe matter from the perspective of the muni-cipalities themselves.

To assess the municipalities’ degree ofinvolvement in this domain, regardinghuman resources, contacts with local




enterprises and performance of such activ-ities, we created and implemented a surveyto the 308 Portuguese municipalities, gather-ing 144 responses. Broadly speaking, in all thedimensions mentioned above, the majority ofthe municipalities claim to be involved inactivities to promote economic developmentand the internationalisation of firms.

In spite of the myriad of government-designed internationalisation promotionpolicies that purport to support firms’exporting and internationalisation activity(Durmusoglu et al., 2012; Leonidou et al.,2011; Volpe Martincus and Carballo, 2010),Portuguese municipalities are essentiallyinvolved in region branding (image build-ing) or in organising fairs and trade mis-sions. There is thus a long way to go for amore profound and comprehensive decen-tralisation at this level. Given the know-ledge municipalities possess about thefirms that are located in their vicinity, itwould be desirable and reasonable toexpect that more efforts be put in imple-menting and arranging information-relatedprogrammes (including the provision ofmarketing/information/advice, informationon market opportunities and general infor-mation about doing business in a specificcountry), and education- and training-related programmes (namely, counsellingand training on the export process for inex-perienced exporters, training on exportdocumentation and foreign language sup-port). At present, the austerity programmethat the Portuguese government faces puts alot of financial pressures and restrictions onmunicipalities, which justify that trademobility-related programmes (most not-ably, assistance in participating in tradeshows/exhibitions or participation in trademissions abroad) or financial aid-relatedprogrammes (e.g. co-finance the participa-tion in trade missions, sponsor the creationof export consortia, export loans, exportcredit guarantees or funds transferring) arestill considerably centralised.

Our multivariate econometric estimatesshow that municipalities with a higherbudget per capita and located in regions withhigher purchasing power tend, on average,ceteris paribus, to assign more humanresources to SMEs internationalisation pro-motion activities, which concurs with the find-ings byKauneckis and Andersson (2009), andshows that the municipal budget determinesthe municipal government’s capacity torespond to local demand, and that municipa-lities’ financial and economic context is a rele-vant factor to explain their involvement inlocal firms’ internationalisation promotion.

Entrepreneurial activity, human capitaland the characteristics of the municipalities(population density, area and geographicallocation) also arise as important determin-ants of the decentralisation of the activitiesstudied here. Specifically, larger and periph-eral municipalities, with low density of firmsand low shares of medium and large firmsbut a high volume of exports and a reason-ably high (formally) educated population,tend to be more involved in economic devel-opment/internationalisation activities.

Notes

1. Details of the derivation of the econometric

specification are provided in the appendix.2. The null hypothesis of the Hosmer–

Lemeshow test that the models have a good

fit to the data is accepted, because p-value isgreater than 0.10; additionally, the modelsforesee correctly about three-fourths of the

dependent variable observations.

Funding

This research received no specific grant from any

funding agency in the public, commercial or not-

for-profit sectors.

References

Agrawal A and Ribot JC (1999) Accountabilityin decentralization: A framework with SouthAsian and West African Cases. The Journal ofDeveloping Areas 33: 473–502.




Andersson KP, Gibson CC and Lehoucq F

(2006) Municipal politics and forest govern-ance: Comparative analysis of decentraliza-tion in Bolivia and Guatemala. World

Development 34(3): 576–595.Azfar O, Kahkonen S and Meagher P (2001)

Conditions for Effective DecentralizedGovernance: A Synthesis of Research

Findings. Washington, DC: IRIS Center,University of Maryland, The World Bank.

Balaguer-Coll MT, Prior D and Tortosa-Ausina

E (2010) Decentralization and efficiency oflocal government. Annals of RegionalScience 45(3): 571–601.

Barberia LG and Biderman C (2010) Local eco-nomic development: Theory, evidence, andimplications for policy in Brazil. Geoforum41(6): 951–962.

Blackburn RA and Schaper MT (2012)Introduction. In: Blackburn RA andSchaper MT (eds) Government, SMEs and

Entrepreneurship Development: Policy,Practice and Challenges. Farnham: Gower,pp. 1–13.

Blair H (2000) Participation and accountabilityat the periphery: Democratic local govern-ance in six countries. World Development

28(1): 21–39.Boone C (2003) Decentralization as political

strategy in West Africa. ComparativePolitical Studies 36(4): 355–380.

Calderon H, Fayos T and Cervera A (2005) Amodel for valuation of governmentexport promotion policies: An empirical

analysis in the Spanish context from amarket oriented perspective. InternationalReview on Public and Non Profit Marketing

2(2): 34–49.Cassey AT (2008) State trade missions. Paper

seminar, School of Economic Sciences,

Washington State University, Spring 2008,pp. 1–29. Available at: www.ses.wsu.edu/seminar/papers_Spring08/Cassey_State_trade_missions_4pdf (accessed 7 October

2010).Cavusgil ST and Naor J (1987) Firm and man-

agement characteristics as discriminators of

export marketing activity. Journal ofBusiness Research 15(3): 221–235.

Chang Y-C, Chen M-H, Lin Y-P, et al. (2012)

Measuring regional innovation and

entrepreneurship capabilities: The case of

Taiwan Science Parks. Journal of theKnowledge Economy 3(2): 90–108.

Chapman D, Barcikowski E, Sowah M, et al.

(2002) Do communities know best? Testinga premise of educational decentralization:Community members’ perceptions of theirlocal schools in Ghana. International Journal

of Educational Development 22(2): 181–189.ChapmanDW(2000) Trends in educational admin-

istration in developing Asia. Educational

Administration Quarterly 36(2): 283–308.Chapman DW, Mahlck LO and Smulders A

(1997) From Planning to Action: Government

Initiatives for Improving School LevelPractice. Paris: International Institute forEducational Planning.

Cox KR (2004) Globalization and the politics of

local and regional development: The questionof convergence. Transactions of the Instituteof British Geographers 29(2): 179–194.

DeFilippis J (1999) Alternatives to the ‘‘newurban politics’’: Finding locality and auton-omy in local economic development. Political

Geography 18(8): 973–990.De Vries MS (2012) Decentralisation: What does it

contribute to? The added value of decentralisa-

tion for living conditions in core cities of the EU.Public Policy and Administration 11(4): 545–562.

Durmusoglu SS, Apfelthaler G, Nayir DZ, et al.(2012) The effect of government-designed

export promotion service use on small andmedium-sized enterprise goal achievement:A multidimensional view of export perfor-

mance. Industrial Marketing Management41(4): 680–691.

Faguet J-P (2004) Does decentralization increase

government responsiveness to local needs?Evidence from Bolivia. Journal of PublicEconomics 88(3–4): 867–893.

Faguet J-P (2014) Decentralization and govern-ance. World Development 53: 2–13.

Fineberg A (2013) Promoting sustainable localeconomic development for all areas:

Looking forward or looking back? LocalEconomy 28(7–8): 914–920.

Gencturk E and Kotabe M (2001) The effect

of export assistance program usage onexport performance: A contingency explana-tion. Journal of International Marketing 9(2):

51–72.




Gibson C and Lehoucq F (2003) The local pol-

itics of decentralized environmental policy inGuatemala. Journal of Environment andDevelopment 12(1): 28–49.

Gillespie K and Riddle L (2004) Export promo-tion organization emergence and develop-ment: A call to research. InternationalMarketing Review 21(4/5): 462–473.

Gil S, Llorca R and Martınez Serrano J (2008)Measuring the impact of regional export pro-motion: The Spanish case. Papers in Regional

Science 87(1): 139–147.Gopalakrishna S, Lilien GL, Williams JD, et al.

(1995) Do trade shows pay off? Journal of

Marketing 59(3): 75–83.Greene W (2012) Econometric Analysis. 7th ed.

Upper Saddle River, NJ: Prentice Hall.Joanis M (2014) Shared accountability and par-

tial decentralization in local public good pro-vision. Journal of Development Economics107: 28–37.

Kamoto J, Clarkson G, Dorward P, et al. (2013)Doing more harm than good? Communitybased natural resource management and the

neglect of local institutions in policy develop-ment. Land Use Policy 35: 293–301.

Kauneckis D and Andersson K (2009) Making

decentralization work: A cross-nationalexamination of local governments and nat-ural resource governance in Latin America.Studies in Comparative International

Development 44(1): 23–46.Lederman D, Olarreaga M and Payton L (2010)

Export promotion agencies: Do they work?

Journal of Development Economics 91(2):257–265.

Leonidou LC, Palihawadana S and Theodosiou

M (2011) National export-promotion pro-grams as drivers of organizational resourcesand capabilities: Effects on strategy, competi-

tive advantage, and performance. Journal ofInternational Marketing 19(2): 1–29.

Lesch WC, Eshghi A and Eshghi GS (1990) Areview of export promotion programs in the

ten largest industrial states. In: Cavusgil STand Czinkota MR (eds) InternationalPerspectives on Trade Promotion and

Assistance. New York: Quorum Books,pp. 25–37.

Lewis RJ (1990) Foreword. In: Cavusgil ST and

Czinkota MR (eds) International Perspectives

on Trade Promotion and Assistance. New

York: Quorum Books, pp. viii–xiv.Litvack J, Ahmad J and Bird R (1998)

Rethinking decentralization in developing

countries. The World Bank Sector StudySeries, Paper no. 21491. Washington, DC:The World Bank.

Litvack J and Seddon J (eds) (1999)

Decentralization Briefing Notes. Washington,DC: The World Bank.

Lobao L and Kraybill D (2009) Poverty and

local governments: Economic developmentand community service provision in an eraof decentralization. Growth and Change

40(3): 418–451.Marinetto M (2003) Governing beyond the

centre: A critique of the Anglo-governanceschool. Political Studies 51(3): 592–608.

Nanetti RY, Rato H and Rodrigues M (2004)Institutional capacity and reluctant decentra-lization in Portugal: The Lisbon and Tagus

Valley Region. Regional and Federal Studies14(3): 405–429.

Oates WE (1972) Fiscal Federalism. New York:

Harcourt Brace Jovanovich, Inc.Oates WE (1999) An essay on fiscal federalism.

Journal of Economic Literature 37(3):

1120–1149.O’Gorman C and Evers N (2011) Network inter-

mediaries in the internationalisation of newfirms in peripheral regions. International

Marketing Review 28(4): 340–364.Pacheco P (2004) What lies behind decentraliza-

tion? Forest, powers and actors in lowland

Bolivia. European Journal of DevelopmentResearch 16(1): 90–119.

Pinho JC and Martins L (2010) Exporting bar-

riers: Insights from Portuguese small- andmedium-sized exporters and non-exporters.Journal of International Entrepreneurship

8(3): 254–272.Ribot JC (1999) Accountable representation and

power in participatory and decentralizedenvironmental management. Unasylva, FAO

50(4): 18–22.Ribot JC (2001) Local actors, powers and

accountability in African decentralizations:

A review of issues. Paper prepared forIDRC, assessment of social policy reformsinitiative, World Resources Institute,

Washington, DC.




Ribot JC (2002) Democratic Decentralization of

Natural Resources: Institutionalizing PopularParticipation. Washington, DC: WorldResources Institute.

Rodden J (2003) Comparative federalism anddecentralization: On meaning and measure-ment. Comparative Politics 36(4): 481–500.

Rodrıguez-Pose A and Gill N (2005) On the ‘eco-

nomic dividend’ of devolution. RegionalStudies 39(4): 405–420.

Rondinelli D, McCullough J and Johnson R

(1989) Analyzing decentralization policies indeveloping countries: A political-economyframework. Development and Change 20(1):

57–87.Rose AK (2007) The foreign service and foreign

trade: Embassies as export promotion. WorldEconomy 30(1): 22–38.

Rowland AM (2001) Population as a determi-nant of local outcomes under decentraliza-tion: Illustrations from small municipalities

in Bolivia and Mexico. World Development29(8): 1373–1389.

Serra F, Pointon J and Abdou H (2012) Factors

influencing the propensity to export: A studyof UK and Portuguese textile firms.International Business Review 21(2): 210–224.

Shamsuddoha AK, Ali MY and Ndubisi NO(2009) Impact of government export assis-tance on internationalization of SMEs fromdeveloping nations. Journal of Enterprise

Information Management 22(4): 408–422.Sorens J (2009) The partisan logic of decentrali-

zation in Europe. Regional and Federal

Studies 19(2): 255–272.

Stagg L (1990) Is Minnesota’s export portfolio a

good mix? In: Cavusgil ST and Czinkota MR(eds) International Perspectives on TradePromotion and Assistance. New York:

Quorum Books, pp. 3–24.Taylor MZ (2007) Political decentralization and

technological innovation: Testing the innova-tive advantages of decentralized states.

Review of Policy Research 24(3): 231–257.Teixeira AAC and Barros MJ (2014)

Decentralization of public policies for the

promotion of SMEs’ internationalization. Atheoretical account. Revista Portuguesa deEstudos Regionais 35(1): 16–27.

Teixeira AAC and Silva C (2012) A new perspec-tive on local political entrepreneurship:Evidence from Portugal. Local Economy27(4): 332–354.

Tiebout CM (1956) A pure theory of local expen-ditures. Journal of Political Economy 64(5):416–424.

Volpe Martincus C and Carballo J (2010) Exportpromotion: Bundled services work better.World Economy 33(12): 1718–1756.

Wallis JJ and Oates WE (1988) Decentralizationin the public sector: An empirical study of thestate and local government. In: Rosen HS

(ed.) Fiscal Federalism: Quantitative Studies,National Bureau of Economic Research.Chicago: University of Chicago Press,pp. 5–32.

Wilkinson TJ and Brouthers LE (2006) Tradepromotion and SME export performance.International Business Review 15(3): 233–252.

Appendix: Derivation of the econometric specification

The binary nature of the dependent variable (‘municipalities’ involvement’: 1 – Yes; 0 – No)restricts the choice of the estimation model, and in this case, logistic regression seems themost appropriate regression technique to model the occurrence (Greene, 2012). Thus, thisstudy’s analysis is conducted in the context of the general framework of probabilistic models.

Prob event occurs jð Þ ¼ Prob Y ¼ jð Þ ¼ F relevant effects : parameters½ �

whereY¼ 1 if the municipality is involved in activities related to promoting economic devel-opment/internationalisation of firms.Y¼ 0 in the opposite case.




To explain the municipality’s involvement in these activities, there is a set of factors thatpotentially determine the results, so

Prob Y ¼ 1ð Þ ¼ F X,�ð Þ

Prob Y ¼ 0ð Þ ¼ 1� F X,�ð Þ

The X vector includes a set of factors, such as the financial and geographical context,among other variables (see Determinants of the involvement of local municipalities in pro-moting the internationalisation of firms: A literature review section). The set of � parametersreflects the impact of changes in X in the probability of the municipality being involved inactivities related to the promotion of economic development/internationalisation of firms.

In the logistic regression model, the parameters are estimated using the method of themaximum likelihood. Specifically, to test if factors such as, for example, human capital andentrepreneurial activity are significant determinants of the involvement of municipalities inactivities related to the promotion of economic development/internationalisation of firms,we use the estimation of general logistic regression with the following specifications:

ProbðMunicipalities InvolvementÞ ¼1

1þ e�Z

Z ¼ �0 þ �1ln Budget per capitaþ �2ln Number employees|fflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflffl{zfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflffl}institutionalcharacteristics

þ�3ln Telefþ �4ln Dist Lxþ �5ln Pop densþ �6Secondþ �7Sup

þ�8ln Purchasing powerþ �9Un rate Area|fflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflffl{zfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflffl}developmentcharacteristics

þ �10ln Exportþ �11Manufacturingþ �12M B entþ �13ln Number enterprises|fflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflffl{zfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflffl}activitycharacteristics

The definitions of the proxies for the variables are detailed in Table 1.

We chose to proceed to an adjustment of the logistic equation for the model rewritten interms of the odds that the event will occur, which helps to interpret the coefficients of thelogistic function more clearly and directly. In this case, the logistic model is obtained asfollows:

logProbðMunicipalities Involvement

ProbðNo Municipalities Involvement

� �

¼ �0 þ �1ln Budget per capitaþ �2ln Number employees|fflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflffl{zfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflffl}institutionalcharacteristics

þ�3ln Telefþ �4ln Dist Lxþ �5ln Pop dens

þ�6Secondþ �7Supþ �8ln Purchasing powerþ �9Un rate Area|fflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflffl{zfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflffl}developmentcharacteristics

þ �10ln Exportþ �11Manufacturingþ �12M B entþ �13ln Nmber enterprises|fflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflffl{zfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflffl}activitycharacteristics

þ"i




A way of interpreting the logistic coefficient would be based on the changing ratio of oddsassociated with a unitary change in the independent variable

�0þ�1ln Budget per capitaþ �2ln Number employees|fflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflffl{zfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflffl}institutionalcharacteristics

þ

�3ln Telefþ �4ln Dist Lxþ �5ln Pop densþ �6Secondþ�7Supþ �8ln Purchasing powerþ �9Un rate Area|fflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflffl{zfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflffl}

developmentcharacteristics

þ

�10ln Exportþ �11Manufacturingþ �12M B entþ �13ln Nmber enterprises|fflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflffl{zfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflffl}activitycharacteristics

þ"i

ProbðMunicipalities InvolvementÞ

ProbðNo Munciipalities InvolvementÞ¼ e

In this case, e elevated to �i (i¼ 1,. . ., 13) is the factor by which the odds change when theith independent variable increases in one unit. When �i is positive, this factor will be greaterthan 1, which means that the odds have increased and that the factor positively influences themunicipality’s involvement in activities to promote economic development/internationalisa-tion; if �i is negative, this factor will be less than 1, which means that the odds havedecreased, that is, the factor negatively influences the municipality’s involvement in thepromotion of economic development/internationalisation; when �i is equal to 0, the factorwill be equal to 1, which means that the odds remained unaltered, thereby the factor showsno impact on the municipality’s involvement in the said activities. The " represents therandom term.




local economydownloaded from lec.sagepub.com at b-on: 01100 universidade do porto on september 6,...

Documents