Access to Credit and Women ntrepreneurship: vidence from Bangladesh Dhaka, Bangladesh 1
Outline Introduction Research Question Methodology Results Conclusion 2
Introduction Access to capital has been recognized as one of the factors that contribute to the higher level of welfare of households. The credit constraint has a gender characteristic Women are more likely to be constrained than men ntrepreneurship is considered as one of the factors that help poor people to escape poverty. 3
Introduction The limitations of the formal financial sector and the informal financial sector in providing financial services, especially credit, to poor and women encouraged various microcredit programs to evolve in Bangladesh. 4
Introduction : Microcredit Microcredit is instalment based small collateral free loans for the purpose of poverty alleviation. Micro-credit was first initiated in 1976 in Chittagong as part of an action research program. Target of the Micro-credit Summit 1997 reaching 100 Million of World s Poorest Families by 2007. 5
Introduction: Microcredit Target of the Micro-credit Summit 2006 nsuring credit for 175 million of the world's poorest families, especially the women of those families, are receiving credit for selfemployment and other financial and business services by the end of 2015. Target of the Micro-credit Summit 2006 Bringing 100 million families rise above the US$1.25 a day threshold adjusted for purchasing power parity (PPP), between 1990 and 2015. 6
Introduction : Microfinance Sector in Bangladesh Items 2011 No of branches 17,851 No. of employees 231,098 Number of active members (in million) 33.06 Women members 93% Number of active borrowers (in million) 27.17 Amount of Loan Outstanding (USD Million ) 3,588 Net savings of members (USD Million) 2,386 7
Research Question Does an access to microcredit contribute to the development of women entrepreneurship? 8
Previous vidence Chowdhury (2008) indicate that the participation in the microcredit programs does not promote women entrepreneurship. Limitations: The sample size is small (N=920) the survey of the study was conducted in only one of the main seventeen districts in the country. It uses pipeline control to take care of the selection bias problem. Probit technique is used. 9
stimation Strategy C C X X C D D j j C Z v v C.. (1) (2) Two error terms, and C become correlated if the factors that influence access to credit (C in equation 1) also determine the outcome variable entrepreneurship i.e., in equation (2). 10
stimation Strategy: Identification C Since and are not correlated, I use the instrumental variable (IV) technique. The land ownership of the household is used as an instrument as it is a criteria that is followed by MFIs in Bangladesh. Table 1: Microcrdit Program Participation and Land Ownership Criterion Microcredit Program Land Ownership Less than 50 Decimals Total Participation Yes No Yes 2,641 (86.11%) 426 (13.89%) 3,067 No 6,628 (72.25%) 2,546 (27.75%) 9,174 Total 9,269 2,972 12,241 11
stimation Strategy: Identification Other instruments are: per capita formal sector loan availability in the district, per capita availability of loans from MFIs in the district, and an interaction variable of land criterion and agricultural land ownership of the household. 12
stimation Strategy: quations Three types of specifications of the access to credit (C in equation 1) have been formulated to assess the impact of these on women entrepreneurship. X X L X MFASA IJ D j D D j L j C C MF 1 C 2 MFO IJ MF NMF 1 2 1 C L C C NMF IJ L NMF v L 1 MFGB IJ, v L MF NMF MF NMF 1 2 IJ IJ 1, L v, MFBRAC IJ (1a) (1b) (1c) 13
Data Household Income and xpenditure Survey data set 2010 (HIS 2010) from BBS. The framework of Integrated Multipurpose Sample (IMPS) was designed based on Population and Housing Census 2001 with 1000 Primary Sampling Unit (PSUs) of which 640 PSUs belonged to rural area and 340 PSUs to urban area. ach PSU consists around 200 households. Two stage random sampling technique was used to generate the sample. In the first stage of the random sampling, 612 PSUs were selected out of 1000 PSUs. 14
Data In the second stage, 20 households were randomly selected from each of the selected 612 PSUs. In total, HIS 2010 collected data from 12,240 households (N=12,240). Table 3: ntrepreneurs by sex Gender ntrepreneur Total Yes No Male 3,352 12,621 15,973 (20.99%) (89.12%) Female 267 17,010 17,277 (1.55%) (98.45%) Total 3,619 29,631 33,250 15
Results: Specification Testing The Wu-Hausman tests (Wu 1973; Hausman 1978) indicate that the null hypothesis of 'no endogeneity present' has been rejected for all three second stage regression models (Model 1: p<0.01; Model 2: p<0.05; and Model 3: p<0.05). The set of instruments for access to credit from MFIs used in all three models for women (Model 1: F=34.22; Model 2: F=35.12; and Model 3: F=24.12 ) meets the standard of Staiger and Stock (1997). For all three models for women entrepreneurship, values of Sargan statistic are 3.08 (Model 1: p=0.21), 0.91 (Model 1: p=0.6342), 1.01 (Model 1: p=0.6044). The test for over identifying restrictions failed to reject the null hypothesis in all three models. 16
Results Table 7: ffect of Access to Credit on Women ntrepreneurship Ivreg 1 2 3 Variables Access to Credit - Microfinance 0.0830** 0.1105*** 0.1407** Access to Credit - Non-Microfinance 0.0944 0.1907* 0.1895* Total microcredit loan -0.0077 Square of Total microcredit loan 0.0001 Total Gameen Bank loan -0.0351** Square of Total Gameen Bank loan 0.0010** Total BRAC loan -0.0041 Square of Total BRAC loan 0.0000 Total ASA loan -0.0704** Square of Total ASA loan 0.0078** Other MFIs Loan -0.0147** Square Other MFIs Loan 0.0001** Total Non-microcredit loan -0.0059* -0.0059* Square of Total Non-microcredit loan 0.00002* 0.00002* N 17277 17277 17277 17
Results Table 8: ffect of Access to Credit on Men ntrepreneurship IVreg Variables 1 2 3 Access to Credit - Microfinance 0.6369*** 0.7399*** 0.9494*** Access to Credit - Non-Microfinance 0.0889 0.2571 0.2323 Total loan - Microfinance -0.056*** Square of Total loan - Microfinance 0.0004*** Total Gameen Bank loan -0.265*** Square of Total Gameen Bank loan 0.0071*** Total BRAC loan -0.043*** Square of Total BRAC loan 0.0002*** Total ASA loan -0.454*** Square of Total ASA loan 0.0467*** Other MFIs Loan -0.100*** Square Other MFIs Loan 0.0007*** Total loan - Non-microfinance -0.0010-0.0003 Square of Total loan - Non-microfinance 0.000003 0.000002 N 15973 15973 15973 18
Conclusions Non-credit effects compared to credit effects of microfinance program participation are more important for women entrepreneurship. Most of the loans from different microfinance sources are not significant determinants of women entrepreneurship. The loans from Grameen Bank and nonmicrofinance sources significantly determine women entrepreneurship. However, these two types of loans are not positively effective for women entrepreneurship when sizes are small. 19
Conclusions The reality is that the sizes of loans which women take from microfinance and non-microfinance sources are small. Non-credit aspects of microfinance program participation of women also help significantly positively male members of households in becoming entrepreneurs. Loans from different microfinance sources are significantly more effective for men than women in terms of becoming entrepreneurs. Loans from microfinance and non-microfinance sources significantly increase the total market value of businesses of men and those of households. 20
Thanks 21