A 1989 study (Chilowa and Shively 1989) surveyed 200 poor urban households, measured their income and expenditure and nutrition levels in children. Household expenditure was estimated at MK124 monthly, or MK24 per capital. This is equivalent to MK288 per annum. 68 percent of the budget was spent on food. A 1991 study of low income urban households shows an average monthly household income of MK136. It is difficult to compare these figures to the ones given by the World Bank in its poverty profiles, not only because the years are different, but because poverty lines for urban households are generally estimated to be higher than for rural areas. Certainly these households are not wealthy, and those at the bottom of the distribution would be on a level with poorer rural households, but it is difficult to quantify what percentage overall of urban households would be in that position. The nutritional data indicated a lower level of stunting than the national average, but a higher level of wasting, or acute malnutrition. This is consistent with the results from the 1995 MIS survey, which show 8.7 percent wasting, as opposed to a national average of 7 percent. This rises to 13.2 percent when the figures for urban Lilongwe are taken separately. A 1991 survey (Roe 1992) highlights the importance of food as a source of income in the urban informal sector, both in terms of selling and processing and through food production. 70 percent of low income households either had a garden in the immediate area or at home in the rural sector. Links were maintained with the extended family. Those households that consumed food from their garden usually had enough to last for eight months. A quarter of them had sold food in the previous year. This means that low income urban households may be more vulnerable to drought than has previously been acknowledged. However, as with rural households, the main problem is simple lack of resources. 2 Transitory Food Insecurity Transitory food insecurity, i.e. food insecurity resulting from an exogenous shock, which is anticipated to be temporary, can result from climatic shocks, drought or flooding, from political shocks, civil war or trade blockades, and from economic shocks such as unanticipated devaluation. In Malawi the major cause of transitory food insecurity, and the one most commonly discussed, is drought, which has been more frequent in the last decade. As with chronic food insecurity, transitory food insecurity basically reduces to an issue of poverty. If there were no poverty, there would be minimal transitory food insecurity. For example, in general, smallholders farming more than 2 hectares do not suffer from transitory food insecurity. Their income may be reduced when there is a production shortfall, but they have sufficient income reserves to ensure access to food. As discussed above, landholding size is a rather crude and overly simplistic indicator of rural food insecurity. Nonetheless, as an initial attempt to quantify the extent of the problem, it could be argued that smallholders farming less than 0.5 ha suffer from chronic food insecurity and make up 41 percent of rural households. In a major drought, however, the number of food insecure could be increased by the 31 percent of households who farm between 0.5 and 1 hectare.