all the quantitative analysis of food insecure households is based on data collected prior to 1993.
This reflects the outcome of the previous heavily regulated economic system. Since the early
1990s, the liberalisation process has changed the opportunities available to all households,
including the food insecure, particularly in the rural areas. Little of this process is, as yet,
documented, and it will take time before it results in major changes in the numbers of food
insecure. However, these trends should be kept in mind when considering the implications of the
following assessment.

1.1 Smallholders

 The most recent analysis of poverty amongst rural smallholders in Malawi was undertaken
by the World Bank (1995) based primarily on data from the 1992/93 National Sample Survey of
Agriculture (NSSA). The study focused on those households below the 40th percentile of per
capital annual incomes, the poorer, and those falling below the 20th percentile, the poorest. The
estimates of income include the value of subsistence production and off-farm earnings, including
income in kind.

 In 1992/93 MK, the income cut-off points are MK117 for the 40th percentile and MK54
for the 20th percentile. To illustrate where this falls in terms of absolute poverty, the per capital
annual income per adult equivalent necessary to purchase 200 kg of maize in 1992 prices is MK
98 (the calorie needs line), and to buy a minimum level of basic needs, including clothing and
shelter, is MK 158. 43 percent of the rural smallholder population fall below the basic needs line
and 30 percent below the calorie needs line. Income distribution is highly skewed in rural
Malawi, and 80 percent of the population have per capital incomes below MK241. The estimates
of the number of poor or food insecure are relatively insensitive to the definition used and usually
identify between 40 percent and 50 percent of the population.

 Poverty is relatively more prevalent amongst rural smallholders than urban dwellers. 85
percent of the population lives in the rural areas7 but their share of the poorest is higher at 95
percent. There is a strong geographical bias in the distribution of poorer households. The South
contains 51 percent of the population but 66 percent of the poorer households. Table C1 shows
the distribution of households by ADD.

 Poorer households have distinct demographic profiles. They are more likely to be female-
headed. There is a one in three chance of a male-headed household being amongst the poorest
40 percent but a one in two chance for a female headed-household. The poorer the household,
the larger the household size and the higher the dependency ratio. For households below the 20th



 7 The analysis above excludes the estate sector, which is not covered by the NSSA.
Estate tenants and permanent workers are estimated to make up 14.5% of the rural population.