Marcel Fafchamps, David McKenzie, Simon Quinn, and Christopher Woodruff (2011). BREAD Working Paper No. 306, June.

In contrast to the results of three recent randomized controlled trials conducted in Asia, this randomized controlled trial finds that grants benefit male enterprise owners more than female enterprise owners, and wealthier female enterprise owners tend to benefit more than poor female enterprise owners.

Three recent randomized controlled trials conducted in Sri Lanka (focusing on grants) and India and the Philippines (focusing on more widely available microloans) all suggest that increasing the availability of capital has no impact on the profitability of microenterprises operated by women.  In Sri Lanka and the Philippines, this was contrasted with a positive return for males (in India, the study considered only females).  This has been an unhappy finding for microfinance institutions and development organizations that have historically focused on lending to women. 

However, external validity has been a concern because no comparable studies have been conducted in Africa or Latin America.  In this paper, Fafchamps, McKenzie, Quinn, and Woodruff argue that “there are reasons to believe the situation might be different outside south Asia.  In much of Africa, for example, women are more integral to household income generation than in other regions.  This is reflected in labor force participation rates, which are much higher among females in many African countries than in Sri Lanka, India and the Philippines.  We might expect that the stronger attachment to the labor force might be associated with positive marginal returns to capital” (2). 

This study represents the first randomized controlled subject conducted in Africa addressing the provision of capital to female versus male enterprise owners.  The trial was modelled on the Sri Lanka experiment.  Conducted in Ghana, grants of 150 cedi (approximately $120) were provided to randomly selected male and female owned enterprises with zero employees at the time of the baseline survey.  Half the grants were provided in cash and half in kind.  The authors report that “The choice of Ghana was motivated by the desire to provide evidence in an African context, in a country known for a history of involvement of women in business, which provides a setting that is potentially conducive to female business success” (13).  In particular, women and men in Ghana have similar labor force participation rates, and women are more likely to be self employed.

The authors conducted a Ghana Microenterprise Survey in Accra and the proximate industrial city of Tema.  3,907 eligible individuals were identified in 7,567 households screened.  The authors then randomly sampled 907 microenterprises for study distributed across survey enumeration areas.  538 females and 369 males composed this microenterprise owner sample.  The authors conducted a pre-treatment survey of these owners in October-November 2008, and re-surveyed them in February 2009 after the grant treatment.  They dropped owners that could not be found from the sample, generating a final sample of 793 firms consisting of 479 females and 314 males used to produce the results of this study.

Some interesting findings emerged from this trial.  In kind grants were far more effective than cash grants.  However, there were differences across genders and across income groups.  Cash grants of 150 cedi increased profits by 10-14 cedi at most.  In kind grants of 150 cedi increased profits by 37-39 cedi.  However, the impact on females was concentrated among the wealthiest.  The authors report that “among females, the in-kind grants only lead to profit increases for about the top 40 percent of businesses in terms of initial size.  Women running smaller subsistence businesses, those earning $1 per day on average, saw no gains from access to this additional capital” (3).  The conclusion is that “As in Sri Lanka, capital alone does not appear to be enough to grow subsistence businesses run by women” (3).

The authors then ask, why are cash grants less effective, particularly for women?  They find that enterprise owners tend to spend unconditional cash grants or transfer them out of the household, especially for female owners of small businesses.  They propose two channels that might account for such transfers: self-control issues, as caused by time inconsistent preferences, high discount rates, or an inability to save; or alternatively pressure from others to share funds.  The author find support for the first channel.  Individuals with more measured self control in the sample benefit significantly more from the cash grants, while there is little treatment heterogeneity varying with external pressure.  This supports some previous work finding that bias towards the present is an important constraint upon the ability of microenterprises to expand.

The authors conclude with some policy recommendations worth reproducing at length.  They write: “The results offer partially good news for advocates of directing microfinance at women. We do find in Ghana a relatively large group of women whose profits increase a lot when given in-kind transfers. Microcredit has recently been argued as providing a way to allow individuals to overcome present-bias by providing self-discipline and encouragement through regular payments and group meetings…  In such a context it is more likely to be used like the in-kind grants than the cash grants here, thereby leading to improvements in business outcomes. However, our findings suggest this effect to be more powerful for women who are already earning more to begin with, suggesting possible limits on the ability of capital alone to generate business growth among poor subsistence-level female enterprises.  Moreover, as in prior work in Sri Lanka and Mexico, the results show that the average male-owned microenterprise gains a lot from being granted additional access to capital, suggesting that microfinance is focus primarily or exclusively on women is not providing access to a large group of people with a need for more capital” (35).