DOES FINANCIAL STRUCTURE MATTER FOR POVERTY ?EVIDENCE FROM DEVELOPING COUNTRIES
Roland Kangni Kpodar (IMF) &
Raju Jan Singh (World Bank)
World Bank Conference on Financial Structure and Economic Development, Washington, June 16,
2011
Objectives of this PaperWe look at the association between the structure of the financial system and poverty;Focusing on developing countries;Including institutional variables.
Results The structure of the financial
system does play a role in reducing poverty;
Bank-based systems are associated with lower levels of poverty;
As institutions grow stronger, market-based systems could benefit the poor.
Theoretical Background
Why would finance matter? Improves access to credit; Allows risk diversification; Creates jobs; Reduces discrimination.
Why would the financial structure matter?
Financial markets operate in an imperfect information setting;
Institutional arrangements play an important role : Collect information Define property rights Enforce contracts
A bank-based system could thus… Provide wider access to credit; Allow for cheaper risk
diversification; Create more jobs; Reduce discrimination further.
Empirical Analysis
Data Panel data from 47 developing
economies Observations averaged over
five-year periods from 1984 to 2008
IFS, WDI, Financial Structure Database, ICRG
Variables
Dependant variable Headcount Index
Control variables GDP per capita Inflation Infrastructure Trade openness
Variables of interest Quality of institutions Financial structure
The Model
where : Pi,t – the indicator of poverty for a country i at a period t; Yi,t – the level of income per capita; Xi,t – a set of control variables FSi,t – a set of variables accounting for financial structure Law – the institutional variable: Rule of Law ui - unobserved country-specific effect; εi,t - the error term
Estimation Method: System GMM
Results (without institutions)
Log of Poverty Headcount (1) (2) (3) (4) GDP per capita (log) -1.250 -1.222 -1.194 -1.207 [0.168]*** [0.126]*** [0.130]*** [0.144]***Inflation (log) -1.449 -1.358 -1.756 -1.677 [0.900] [0.814]* [0.817]** [0.859]*Inflation squared (log) 0.963 0.780 1.036 0.993 [0.556]* [0.502] [0.518]** [0.524]*Road/area -0.008 -0.009 -0.006 -0.008 [0.004]* [0.004]*** [0.004]* [0.004]**Trade openness -0.006 -0.002 -0.003 -0.006 [0.004] [0.004] [0.004] [0.003]**Composite indicator of structure-size 0.381 [0.152]** Composite indicator of structure-activity -0.027 [0.056] Composite indicator of structure-efficiency 0.086 [0.109] Overall measure of financial structure 0.081 [0.102]Constant 13.414 13.178 12.818 13.151 [1.389]*** [1.093]*** [1.066]*** [1.267]*** Observations 121 119 117 111Number of countries 47 45 44 43Sargan/Hansen test 0.43 0.59 0.51 0.47AR2 0.07 0.18 0.21 0.16
Results (with institutions)Log of Poverty Headcount (5) (6) (7) (8) Composite indicator of structure-size 0.544 [0.216]** Composite indicator of structure-activity 0.560 [0.335]* Composite indicator of structure-efficiency 0.653 [0.218]*** Overall measure of financial structure 0.552 [0.296]*Institutions - Law and order (ICRG) -0.276 -0.270 -0.229 -0.236 [0.085]*** [0.093]*** [0.113]** [0.104]**Composite indicator of structure-size*Institutions -0.110 [0.065]* Composite indicator of structure-activity*Institutions -0.151 [0.091]* Composite indicator of structure-efficiency*Institutions -0.159 [0.056]*** Overall measure of financial structure*Institutions -0.136 [0.076]*
Figure 1. Poverty Headcount Ratio as a Function of Financial Structure and Institutional Quality
0.00.5
1.01.5
2.02.5
3.03.5
4.04.5
5.05.5
6.0
-4.0
-3.0
-2.0
-1.0
0.0
1.0
2.0
3.0
4.0
Pove
rty h
eadc
ount
ratio
(log
)
Robustness Tests
Excluded outliers Added additional control
variables Used alternative measures of
absolute and relative poverty
Thank You