Improving women’s economic empowerment continues to be the focus of a wide range of development projects, ranging from vocational trainings to microcredit to cash and asset transfers. In rural farming areas, such programs face the question of which approach is most effective—improving women’s access to agricultural inputs and value chains, or strengthening their opportunities to generate income outside of agriculture.

Some argue that farming is still the backbone of the rural economy, and that the largest opportunities for women’s economic empowerment reside in this sector. But empowering women in this area is not always an easy endeavor. Farming is physically hard work, it is often stereotyped as a male activity, and agricultural production and incomes face a multitude of risks, which can make farming relatively unattractive. Rural households also increasingly diversify their livelihoods by investing in non-farm businesses. Would women prefer to be empowered via those types of activities?

We asked women (and men) this question directly by eliciting their preferences for income diversification. Through IFPRI’s Ghana Strategy Support Program (GSSP), we conducted an experiment among 1,527 men and women from 799 farming households in two regions in Ghana. This recent working paper describes our findings thus far.

Framed as a rural development program aiming to improve income-generating capacity in targeted areas of Ghana, the experiment asked participants to allocate an in-kind investment grant between investments in crop farming and non-farm income diversification. To make sure that participants told us their true preferences, we selected one man and one woman in each community through a lottery; they received inputs and assets valued at the size of the grant according to their chosen allocation.

In the experiment, we varied the total value of the grant (with participants receiving 12 tokens each worth either 20 or 50 Ghanaian cedis or GHS); the relative value of investments in crop farming vs. non-farm income diversification (with one activity yielding an investment of 50 GHS per token, and the other activity yielding an investment of only 20 GHS per token) , and the availability of training (available at a cost of 3 tokens). We also asked a range of questions on perceptions and norms. In addition, all participants were asked to specify how they would invest the portion of the grant allocated to crop farming, and the portion of the grant allocated towards non-farm income diversification, if selected through the lottery to receive the grant.

Our data sheds light on the question how women and men prefer to invest these in-kind grants, focusing on preferences for crop farming in the Northern Region, and tree crop farming in the Western Region (where farmers were typically growing cocoa), compared to other businesses. The experiment also aims to understand, if this preference indeed differs for men and women, whether this is related to differences in access to physical capital, the availability of training and extension, vs. differences in the profitability or riskiness of crop farming versus non-farm activities, as well as constraints posed by gendered perceptions and social norms.

What did we find?

1. There are strong social norms and gender stereotypes that would suggest men should allocate their investment grant towards on-farm income generating activities, and women towards investments in other, off-farm business activities. The vast majority of both men and women indicated that men are better at crop farming, while women are generally better at running other non-farm businesses. 
2. Given these strong social norms, gender differences in actual investment choices are actually smaller than we expected, with both men and women revealing a preference for diversifying their investments in both on-farm and other business activities. We do, however, observe strong gender differences in the type of investments in other business activities; women are most likely to invest the grant in activities such as trading food (44%), trading other products (18%), and food processing (18%), while for men, a dominant activity is livestock rearing (47%).
3. The observed gender differences are strongly associated with gender stereotypic perceptions of men’s and women’s skills in these sectors. Consistent with the perception that crop farming is male-dominated, men associated more positive attributes to crop farming and ranked themselves as better skilled than others in farming. Similarly, women perceived other business opportunities more positively and ranked themselves more skilled in running a business. Controlling for these differences reduces the estimated gender gap in investment preferences.
4. Access to financial capital or training, which we varied in the experiment, did not influence investments as much as we expected a priori. Expanding women’s access to financial capital did not strongly influence the share of the grant that they allocated to crop farming, and participants even left significant resources on the table in order to maintain a diversified investment choice. In addition, instead of shifting investments towards crop farming when offered the opportunity to take a training in crop farming, women preferred business trainings.

These research results are relevant to policy makers and practitioners, as they shed light on three main issues: Whether beneficiaries in the Ghanaian context would prefer to invest in farm or non-farm income generating activities; whether the provision of additional financial capital or skills training could help bridge gender gaps in such preferences; and whether other factors, such as social norms and gender stereotyping, are associated with women’s preferred pathways to economic empowerment.

To our own surprise, the results show that, at least in this context, men and women both have a strong preference for diversified investments—and thus, neither projects focusing on farm activities nor projects focusing on non-farm activities would be misplaced. Strong gender norms and a perception of men (women) being better at crop farming (running non-farm businesses) did not prevent women from investing significant amounts in crop farming. However, increasing financial capital and access to extension or training did not further close the gender gap in preferences. In fact, providing skills training led to more specialization, with men and women investing more in the sector that stereotypes prescribe.

This is not to say that policy makers should expect better development outcomes when aligning investments with prevailing norms and stereotypes instead of encouraging women to participate more in activities stereotypically seen as “male.” We did not analyze what type of activity maximizes women’s incomes or empowerment, and it could be that larger gains can be expected from investments in crop farming. Our findings do suggest, however, the importance of diversification in this sector, and the limited responsiveness of women’s investment portfolios to increasing financial capital, or access to additional skills trainings. Such nuances are important to keep in mind when designing programs focusing on women’s economic empowerment and analyzing their impacts.

Berber Kramer is a Research Fellow with IFPRI’s Markets, Trade, and Institutions Division; Isabel Lambrecht is a Research Fellow with IFPRI’s Development Strategy and Governance Division. This post is based on research that is not yet peer-reviewed. This work is supported by the U.S. Agency for International Development (USAID).