Adult Health in the Time of Drought

It is a well-known fact that households in developing countries often undergo weather-related and other shocks that drastically affect incomes. A large and growing literature explores the effectiveness of response to these events. One strand of the literature addresses the strategies that households and governments use to protect against income shocks. A second strand looks at the effectiveness of these strategies in reducing fluctuations in consumption. All the studies find that some, but not all, households are able to smooth consumption. Households facing liquidity constraints, in particular, have limited smoothing ability. For these households income fluctuations lead to a welfare loss.

This study examines the impact of drought on the individual. The handful of studies that have dealt with this subject have all focused on children. By contrast, this study looks at adults and changes in their body mass. It measures body mass for three reasons. First, body mass index (BMI) is a good proxy for underlying adult health, which is a valid indicator of shock effects in its own right. Second, evidence increasingly links adult BMI to agricultural productivity, and therefore to longer-term economic welfare. Third, low BMI correlates with a large number of health-related indicators, including early onset of chronic conditions and increased risk of premature mortality. This study examines the impact of income shocks by assess-ing the effects of the 1994-95 drought on adult health in Zimbabwe.

The Study Sample
The study draws on a unique panel data set of households residing in rural Zimbabwe. It analyzes information for 1994-97 from annual surveys of about 400 households in three agroclimatic regions. These surveys contain adult anthropometric data, together with other relevant individual, household, and community information. In the survey period, two years of reasonably good rainfall were followed by one drought year and one year of especially good rainfall and crop production.

The Effect of Drought on Adult Health
The core result is that, controlling for household and locality characteristics and individual fixed effects, the drought reduced women's BMI by 1.15 percent for every 10 percent negative deviation in rainfall from the long-term average. Male BMIs were unaffected by changes in rainfall. Other results show that changes in agricultural capital stock during the drought year did not affect adult body mass. But livestock held in the year before the drought had a positive impact on the BMI of women who experienced the drought. (This relationship did not hold for men.) Drug shortages in local clinics were associated with reductions in adult body mass (this result was more robust for women than for men). And staff shortages in these clinics did not appear to worsen adult health. The impact of livestock holdings on women's body mass deserves further comment. In the very different environment of semi-arid West Africa, livestock transactions do not appear to play a major role in smoothing household consumption. But in rural Zimbabwe, livestock holdings improve women's health through two possible channels. First, livestock are a relatively liquid store of wealth whose real value has been maintained in Zimbabwe despite persistent inflation. Second, livestock-especially oxen-can substitute for a wide variety of labor tasks. In addition to ox-plowing instead of manual hoeing, oxen can be used to pull carts carrying firewood and water. These activities are regarded as women's responsibilities in the survey area. Not only does animal haulage reduce human energy expenditures, but it also allows greater quantities of wood and water to be collected during a single trip.

"Although women, but not men, were adversely affected by the 1994-95 drought, they did not bear the effects equally. Women in poor households and daughters generally bore the brunt of the shock."

Results further show that the impact of negative rainfall shock does not fall uniformly on all women. Drought did not affect the BMIs of daughters-in-law of the household head. But wives and daughters were adversely affected, with a 10 percent reduction in rainfall corresponding to a 2.2 and 3.9 percent reduction in BMIs, respectively. If a household held livestock, however, wives of the household head retained BMIs. These findings suggest that wives are vulnerable to drought to the extent that households lack assets that they can use as buffers against income shocks.

By contrast, household assets apparently do not help daughters of household heads retain their BMIs. The authors hypothesize that during drought these women have relatively unchanged energy needs. They may continue to have responsibilities for household activities such as collection of water and fuel. If daughters have a harder time finding these resources during drought or are forced to seek employment outside the household, their energy expenditures might actually increase. At the same time, the position of daughters in the household can be somewhat ambiguous. Women who have returned to the household following failed marriages are sometimes stigmatized. Their relatives may expect them to return to their husbands. In such circumstances, household heads may limit the amount of resources made available to daughters. Unmarried daughters who have had children out of wedlock may face similar constraints within their households.

The BMIs of daughters-in-law remain unchanged by drought perhaps because these women have access to resources outside the household. Regulations in the survey area severely limit off-farm work for household heads, but not for their children. Consequently, daughters-in-law may have access to remittance income that offsets the impact of drought.

As can be expected, the drought disproportionately affects poorer households. This is consistent with the argument that livestock holdings built up prior to drought provide a means of smoothing consumption.

These results have focused on ex ante strategies for reducing the effects of drought. What can be said about ex post strategies such as the grain loans offered to drought-stricken households? Findings suggest that the grain loans scheme implemented by the government had no significant impact on adult body mass. However, these results should be regarded with some caution because of measurement difficulties.

Conclusions
Although women, but not men, were adversely affected by the 1994-95 drought, they did not bear the effects equally. Women in poor households and daughters generally bore the brunt of the shock. An ex ante private coping strategy, the accumulation of livestock, protected women who could afford it against the adverse consequences of drought. By contrast, ex post public responses were not effective, though this finding should be treated with caution.

To order a discussion paper, please fill out an online order form, email IFPRI-FCN@cgiar.org or send requests to Food Consumption and Nutrition Division, IFPRI, 2033 K Street, N.W., Washington, D.C. 20006, U.S.A.