Click here for search results


Capital, Gender, and Microenterprise Growth in Ghana

Marcel Fafchamps, David McKenzie, Simon Quinn and Christopher Woodruff, June 2011

Our latest impact note looks at an experiment funded by the Gender Action Plan which re-examines the issue of whether capital alone is enough to get female microenterprises growing.

Despite the emphasis placed by microfinance organizations on lending to female business owners, evidence from three recent randomized controlled trials has cast doubt on the ability of capital alone to grow female-operated microenterprises: in my own previous experiment in Sri Lanka (see impact note 4) women given grants saw no increase in business profits, while the recent randomized trials of microfinance in the Philippines and India also see very little in the way of profit increases when women get loans.

One possible interpretation is that female-owned microenterprises in these countries are already operating at their efficient level of capital, which might be very low – especially in countries where other labor market options for women are limited.

However, an alternative explanation could be that the small scale of many female-owned firms is not efficient, but arises instead from a lack of separation of business and household accounts, and from  inefficiencies in the way people allocate assets between them. These inefficiencies might arise from self-control problems, leading owners to not undertake profitable investments, or from external pressure to share with others. Either of these could lead to a lack of asset integration, so that the form that capital comes into the business will determine the extent to which it helps the business grow.

We conducted a randomized experiment in urban Ghana which enables us to test these ideas.

The Experiment
The experimental design is very similar to the earlier Sri Lanka study: 793 microenterprises (479 female-owned and 314 male-owned) were randomly allocated into treatment and control groups. The treatment group got a one-time grant of 150 cedis (about $120).

 A key feature was that half of these grants were given as cash, with no restrictions on what they could be spent on, whereas the other half of the grants were in-kind, with research assistants purchasing raw materials, inventories, or equipment for the business (with the owner choosing what was purchased).

The sample is twice as large as the Sri Lankan study, offering greater statistical power to look at gender differences, and at the role of cash vs in-kind grants.

Four follow-up surveys were then conducted at 3 months, 6 months, 9 months, and 1 year after the grants were given.

The results of the experiment cast doubt on the ability of capital alone to raise the incomes of subsistence female-owned businesses: Women earning approximately $1 per day at baseline did not see any improvement in profits from either the cash or in-kind grants.

Impct noe 14 women

Women who at baseline were earning in the top 40% (about $5 per day on average) did see large increases in business profits if they got the in-kind grant, but no increase in profits if they got the cash grant: i.e. the form the capital came in mattered a lot.

For males, there was no difference between subsistence and higher initial profit businesses in terms of their response to the grant. Our point estimates suggest that the in-kind grant also had larger impacts than the cash grants for males, but the difference is not statistically significant.

Impact note 14 men

What happened to the cash grants if it didn’t end up in the business? Most of it seemed to have gone to household consumption, with a little bit also going in transfers to others.

Finally we try and understand what causes this difference in response to cash and in-kind grants. We have a number of measures of self-control, as well as a number of measures of pressure to share within the household and outside the household. The difference between impacts of the cash and in-kind grants seems more due to self-control issues than to external pressure.

The results are not consistent with standard models of microenterprise investment and growth: to explain the large difference in outcomes between cash and in-kind grants we need a model with a lack of asset integration, in which the form in which capital comes in affects the extent to which it is invested in the business.

Policy Implications
The results offer mixed news for advocates of directing microfinance at women:

  • Capital alone does not seem to be enough to grow subsistence enterprises. These results are consistent with the efficient scale just being very low for such women – and the best thing to do for them may be policies to draw them into wage work.
  • There is a group in Ghana of higher profitability women for whom access to capital does have very high returns – very few such women existed in the Sri Lankan study. But these returns were only realized when capital came in-kind.
  • Loans and grants to help microenterprises grow may therefore succeed better if disbursed in kind, not in cash.
  • And finally, as in prior work in Mexico and Sri Lanka, the average male-owned microenterprise gains a lot from more capital – so developing financing tools for financing these businesses is an important priority.

For further reading see: Fafchamps, Marcel, David McKenzie, Simon Quinn and Christopher Woodruff (2011) “When is capital enough to get female microenterprises growing? Evidence from a randomized experiment in Ghana”. World Bank Policy Research Working Paper No. 5706.


Permanent URL for this page: