Sustainable poverty reduction is difficult to achieve without mechanisms that help households manage shocks. Covariate shocks, which impact virtually everyone in a community, occur because of natural disasters, war, price instability and financial crises. Idiosyncratic shocks, which impact only individuals or specific households, commonly arise due to farm-specific crop failure or livestock mortality, illness, injury or death of a household member, or property loss due to fire or theft.
Such shocks often severely affect a household's asset holdings thereby trapping a household in chronic poverty. Shocks may also impact the asset holdings of husbands, wives and commonly held assets differently. Finally, the same covariate shock experienced at the same time in different countries, but with different cultural settings may result in different risk management responses. In order to design effective policies that minimize the impact of shocks it is important to understand these nuanced responses.
Different Countries, Same Shocks
The 2007-2008 spike in world food prices was a shock that taxed households' existing risk management mechanisms worldwide. By analyzing panel data from Uganda and Bangladesh collected immediately before and after the food price crisis we are able to expand our understanding of the gender and culture-differentiated impact of shocks on assets.
We take advantage of detailed information on the type of shock (weather, food price, death, illness etc.), the type of asset (livestock, equipment, jewelry etc.), and if the asset is owned jointly or exclusively by either the husband or the wife. By comparing two countries with very different sociocultural traditions, reflected in differences in family structure, production organization, and gender norms, we can infer how the impacts of shocks might differ between these countries.
Understanding Cultural Norms
In the agricultural sector in Bangladesh women tend to be invisible, owing to the assumption that they are not involved in agriculture, and because of cultural norms that value female seclusion and undervalue female labor. Yet, participation of women in agriculture has increased from less than 20 percent to 33.6 percent of the total. Because of this perception, and because Islamic laws provide that daughters inherit half the share of sons, women have less land and fewer assets than men. They also have not been targeted by agricultural extension services.
Upon marriage, daughters leave their natal villages for their husbands' villages, and dowries are normally given to the groom's family. While dowries are often considered a daughter's share of the family's wealth, the groom's family typically controls the dowries. The one-time costs of dowries and wedding expenses are a large drain on families' resources and are highly correlated with downward mobility into poverty. Men bring more assets to marriage and usually own most of the household's land and assets such as livestock and productive equipment, while jewelry is owned largely by women and considered a woman's asset.
In Uganda, most agricultural production takes place on small family farms. Roughly 60 percent of women actively participate in agriculture and constitute 80 percent of economically active women. These percentages are, in fact, slightly higher than those for men. Gendered norms in Uganda imply that men decide the pattern of work and allocation of resources, with women having some say about women's crops.
There may also be a gendered division of agricultural activities, with men in charge of land clearing and marketing and women in charge of weeding, postharvest processing, and food preparation. Men and women frequently use the same plots for agricultural purposes-particularly because land is in short supply although in some instances men and women do farm separate plots.
In much of Uganda, land is owned collectively by the clan and inherited and distributed in consultation with clan elders. Often, women, who in most cases lack formal ownership rights, are able to access and use land through husbands or male kin. The type of marital regime-customary, church or co-habitation strongly influences the rights Ugandan women have with respect to control over land and other key assets.
Prior to marriage, a bride-price—usually in the form of cattle or other Iivestock—is paid by the husband's family to the wife's family to compensate for the resources needed to raise and educate the woman. Once the bride-price has been paid, the married woman moves to her husband's clan and is no longer regarded as the responsibility of her natal clan. The bride-price can become a basis for control of women's labor by her husband.
The Impact of Exposure to Shocks
Households in Bangladesh reported being affected by a number of shocks as well as life-cycle events. About 5.6 percent of households in the survey communities experienced difficulties due to floods or cyclones, while 7.5 percent were affected by drought. After the 2007-08 global food price increases, 45 percent of households reported having been affected by the rise in prices. Illness was the most prevalent idiosyncratic shock, affecting 10 percent of households. Death, while arguably an extreme shock, affected only a small percentage of households. Dowry and wedding expenses, which we classify as a life-cycle event, affected 4 percent of households. Among the other lifecycle events reported, only 2 percent report receiving remittances and less than 1 percent benefited from inheritance or dowry receipts.
Despite these shocks, households on average were able to accumulate assets over the three-year survey interval. The most important assets other than land owned by households are consumer durables, livestock, and jewelry, with most of these assets held jointly by husband and wife. Husbands, however, owned the bulk of exclusively held assets. There are clear gender differences in the type of assets held within the household. Although women hold the minority share of most assets, jewelry is clearly a woman's asset as they held 40 percent of the household's share in 2007 and 33 percent in 2010. Women also tend to invest in livestock, although their share declined from 16 percent in 2007 to 9 percent in 2010.
In Uganda the prevalence of shocks is quite high from 2007 to 2009. On average, households experienced at least five shocks in the three-year period and at least two shocks in 2009 alone. Drought is the most prevalent shock, affecting almost 90 percent of households. Floods (47 percent) and crop loss (51 percent) due to causes other than climate are quite high as well. About two-thirds of households were adversely affected by large increases in food and fuel prices. Illness shocks affected 38 percent of households. While 65 percent of households report a death in the extended family, a much lower percentage of households—13 percent—experienced death of a household member.
Overall the value of assets held by the household increased. The largest proportion of household assets is held exclusively by the husband, followed by jointly owned assets. While wives have access to a larger share of assets through joint ownership with the head, the fraction of assets exclusively held by the wife is 10 percent. Nevertheless, this share is slightly larger than the share of the wife's exclusively held assets in Bangladesh. The increase in value of total household assets over the period seems to have come entirely from an increase in value of jointly held assets.
Excluding land, about three quarters of assets is comprised of consumer durables. Asset shares of jewelry and productive equipment went down by half from 2 percent to 1 percent and from 9 .2 percent to 4.1 percent, respectively. The share of livestock increased from 16.4 percent in 2007 to 18.3 percent in 2009. The largest share of assets held by wives is cash savings at 3 1 percent. The next highest share of household assets held by women is livestock, of which they own 25 percent. The ownership shares remain more or less constant over the period except for share of cash savings held by women, which increased by 10 percent.
How Do Men and Women Respond?
Although many shocks are similar in both countries, commonly experienced shocks do not necessarily have the same effects on men's, women's, and jointly owned assets. The differences in the relative impact of the shocks, and their impacts on different types of assets depending on whether these were owned by men or women, show that responses to shocks are context-specific, and that gendered responses to shocks are even more so.
The impact of a weather shock on men's and women's assets appears to depend on the level of involvement in agricultural production and thus the increased exposure to weather risk. For example, women are less involved in agricultural production in Bangladesh, whereas they manage land and grow their own crops in Uganda. Flood and drought affect a large proportion of households in both Bangladesh and Uganda. Yet, flood shocks negatively affect husband's land accumulation in Bangladesh, and have a positive effect on husbands' land in Uganda. In contrast, drought shocks have a negative impact on husband's assets (mostly agricultural and consumer durables) in Bangladesh and wives' assets in Uganda.
The lack of impact of these weather-related shocks on wives' assets in Bangladesh may reflect lack of direct exposure to agricultural risk because women rarely cultivate land independently or own and control agricultural assets. The one exception is that droughts are associated with lower growth of livestock owned by wives. This may be due to the fact that livestock are more vulnerable to drought because of scarcity of water or fodder. Or it may be that priority is given to preserving the household's working animals, which are generally jointly or exclusively husband-owned, rather than the small livestock owned by women.
Rise in Food Prices
Forty-five percent of households in Bangladesh and 67 percent of households in Uganda were negatively affected by the rise in food prices. Land and other assets, in aggregate, were relatively well insured against food price increases in Bangladesh, but jointly held and wives' assets in Uganda were negatively affected. However, some portfolio reallocation appears to have occurred, because Bangladeshi husbands decreased their holdings of agricultural durables and livestock, and households disposed of jointly held jewelry. In Uganda, food price increases reduced holdings of jointly owned and wife-owned durable goods.
Life Cycle and Idiosyncratic Events
Marriage, death, and illnesses are events that have significant and long-lasting effects on asset accumulation and de-accumulation, even if they do not affect all households at the same time. In Bangladesh, it is a family's responsibility to provide a dowry to guarantee a daughter's successful marriage, and both families incur significant wedding expenses. Dowry and wedding expenses reduce the wife's landholdings. Dowry receipts, which occur when a son marries, reduce both joint and husband's land, signaling the devolution of parental land to the next generation upon a son's marriage.
The relatively small impact of covariate and idiosyncratic shocks on joint landholdings may indicate that the household collectively tries to protect its main source of livelihood, disposing of exclusively held assets when shocks occur. In Bangladesh, illness shocks-both illness of the main income earner and the expenses incurred-are the most prevalent idiosyncratic shock. These shocks end up eroding women's assets more than men's. Death of a household member is a serious matter in Uganda. It often results in a reduction of the husband's landholdings, possibly because the head of the household may be responsible for funeral expenses. Death also reduces a wife's asset holdings, but not jointly held or a husband's assets.
Knowing the type of shock that most affects men's and women's assets may help in designing appropriate social protection schemes. For example, in Bangladesh, weather-related shocks, such as floods or droughts, have a larger impact on men's assets, while illness shocks take their toll on women's assets. In Uganda, drought shocks affect wives' assets, but not husbands' assets. One could surmise that weather-based insurance could be marketed to husbands in Bangladesh and wives in Uganda, but that health insurance might be more readily taken up by wives in Bangladesh. Designing social protection schemes should also take into account the prevalence of shocks, the severity of their impact, and whose assets are used to cope with them. While the food price shock emerged as the most important shock from 2006-2010, in Bangladesh, illness was the most prevalent shock between 1996 and 2006, most severely affecting women's assets.
Differences in the institution of marriage and cultural concepts of joint and individual ownership may affect the extent to which joint or individually owned assets are used to cope with shocks. In Bangladesh, the results showing generally insignificant impacts on aggregate land and asset holdings-while individual assets are sacrificed at the margins-indicates that husbands and wives try to preserve the economic base of the household unit. In contrast, in Uganda, husbands' assets appear better insured than wives' or even joint assets. Policy interventions to enable households to manage risk need to take into account both individual and joint asset ownership, and ensure that social protection schemes do not-intentionally or unintentionally-widen the gender asset gap.
Behrman, J. 2011. HarvestPlus Uganda Orange Flesh Sweet Potato (OFSP) Impact Evaluation: Qualitative Report of Findings. Washington, DC: International Food Policy Research Institute.
Birner, R., A. Quisumbing, and N. Ahmed. 2010. Governance and Gender: CrossCutting Issues. Paper commissioned by the Government of the People's Republic of Bangladesh for the Bangladesh Food Security Investment Forum, Dhaka, May 26-27.
Doss, C. 2010. "Women, Marriage and Asset Inheritance in Uganda." Working Paper presented at the Chronic Poverty Research Centre/Overseas Development Institute Roundtable "Inheritance and the Intergenerational Transmission of Poverty," London, October 11.
Publication made possible by support in part from the US Agency for International Development Cooperative Agreement No. EDH-A-00-06-0003-00 through the Assets and Market Access CRSP. All views, interpretations, recommendations, and conclusions expressed in this paper are those of the authors and not necessarily those of the supporting or cooperating organizations.