Does Microfinance Promote Entrepreneurship and Innovation? A Macro Analysis

Abstract

Innovative entrepreneurs in Africa experience a lack of financial support and underdeveloped financial systems. Microfinance could offer a solution to this complex issue. In this respect, this research paper studies the contribution of microfinance to the innovation process through entrepreneurship. The study first reviews the effect of microfinance on opportunity-driven entrepreneurship and necessity-driven entrepreneurship. Second, it investigates the relationship between entrepreneurship and innovation. According to the literature, opportunity-driven entrepreneurship is more associated with different forms of innovation than necessity-driven entrepreneurship. Using pooled ordinary least squares (OLS) and random effects techniques on an unbalanced panel of 45 countries, our analysis supports the hypothesis that microfinance has a negative effect on necessity entrepreneurship. However, its effect on opportunity entrepreneurship depends on the socio-economic conditions of the countries. Regarding regional differences, there is strong evidence of an entrepreneurial shift from necessity to opportunity motivations in the Middle East and North Africa region and weak evidence in Sub-Saharan Africa.

Keywords:

• Microfinance
• Entrepreneurship
• Innovation
• Global Entrepreneurship Monitor (GEM)

Notes

This theory is inspired from the product-cycle model of Vernon (1966).

http://www.mixmarket.org

http://www.gemconsortium.org

See also Morduch (2000).

According to King and Levine (1993) external finance of innovation is central for two reasons: (i) The labour requirements of innovation are assumed to be much larger than just the entrepreneur's time, so the entrepreneur's wealth is not enough to cover all costs; (ii) the risk of innovation success is entirely diversifiable, so that reliance on any amount of internal finance is inefficient.

The 30 years after the Second World War are characterized by a significant economic prosperity combining high productivity with high average wages and high consumption. In France, this period is named les trentes glorieurses – the glorious thirty. It was characterized by a highly developed system of social benefits (see Angresano, 2011).

After the last period, comes a period of decline. The technology is overtaken and the demand is decreasing. A new product-cycle has to emerge through innovations. 

While interesting, this approach has known a dead-end due to continuous technological progress.

Many African countries are in this middle stage of development. Examples in Sub-Saharan Africa are Angola, Ghana, Botswana and others. The MENA region is also considered to be in the second development stage.

Sub-Saharan Africa: Angola, Ghana, South Africa, Tonga, Uganda, Zambia. Middle East & North Africa: Egypt, Jordan, Lebanon, Morocco, Tunisia, Yemen. Latin America & Caribbean: Argentina, Bolivia, Brazil, Chile, Colombia, Costa Rica, Dominican Republic, Ecuador, Guatemala, Jamaica, Mexico, Panama, Peru, Philippines, Uruguay, Venezuela. Europe & Central Asia: Bosnia and Herzegovina, Croatia, Kazakhstan, Macedonia, Montenegro, Poland, Romania, Russia, Serbia, Turkey. East Asia & Pacific: China, Indonesia, Malaysia, Thailand. South Asia: Bangladesh, India, Pakistan.

In supplementary exercises, we replace the GLP by GLP/borrowers.

The Worldwide Governance Indicators Database reports aggregate and individual governance indicators for 215 economies over the period 1996–2011, for six dimensions of governance: voice and accountability, political stability and absence of violence, government effectiveness, and regulatory quality, rule of law and control of corruption. For our model we test the rule of law variable because it captures perceptions of the extent to which agents have confidence in and abide by the rules of society, and in particular the quality of contract enforcement, property rights, the police, and the courts, as well as the likelihood of crime and violence. A higher rule of law would enhance opportunity entrepreneurship.For more details see: http://info.worldbank.org/governance/wgi/resources.htm

Also, contrary to the FE estimator, the RE estimator offers the possibility to test dummy variables.

Note that estimations with GLP per borrower as the explicative variable give similar results to estimations with GLP except for the random effect estimator of GLP per borrower on NDEA, which is not statistically significant. Estimation results are available upon request.

56% of the variance of the regression on RONEA is due to differences across countries. It is equal to 39% for the regression on NDEA.

Indeed, almost 90% of countries in our sample are middle income countries.

Except for pooled OLS estimator for ODEA regression.

We also tested the logarithm of life expectancy as an indicator of human capital. The results were not significant.