Determinants of household activity choice, rural income strategies and diversification - Institutional Constraints to Small Farmer Development in Southern Africa - page 63

Table 3.2 provides a breakdown of the various income sources and their relative importance.
More than 30% of the respondents from Ndakana had at least a member engaged in wage
employment. In the case of Mlungisi, participation in wage employment was approximately
3% higher than Ndakana among the survey households. A total of 39 respondents,
representing more than 50% of all respondents, engaged in agricultural activities as a means
of livelihood in both communities. In Ndakana, the more rural community where field
sizes are relatively larger, a larger proportion of the respondents, about 58%, engaged in
agricultural activities, and earned an average income of R1,942, while in the peri-urban area
of Mlungisi, about 44% of the respondents from that sub-sample engaged in agricultural
activities, earning slightly lower average agricultural income of R1,396. Interestingly,
for both communities, despite a large proportion of households engaging in agriculture,
agricultural incomes turned out to be extremely low relative to other income sources.
The results show that households derived income from other sources such as operation of
own businesses, pensions and grants, and remittances from family members living outside
the community. The dominant own businesses observed in the communities were owning of
tuck-shops, vending of cellular phone re-charge vouchers (or 'airtime'), selling of vegetables,
running of taverns and shebeens (an unlicensed drinking place for alcoholic beverages)
and food processing. The data from the two communities indicate that about 18% of the
respondents in Ndakana operate their own business enterprises and 17% in Mlungisi, and
in each case, this activity contributed about 17% of household income (see Table 3.2). As
Table 3.2 revealed, pensions and grants emerged as important sources of income in both
communities, accounting for the bulk of incomes received by 64% in Ndakana and 76% in
Mlungisi. Remittances were clearly more important in the rural community than the peri-
urban community (Table 3.2).
3.6.3 Dynamics of poverty reduction
The estimates of poverty incidence in this paper employ Foster, Greer and Thorbecke
(FGT) method (1984) mathematically expressed as:
P α = 1/n Σy i <z [z-yi i / z] α
where y i is the real per capita income of a household i, is the total number of household
members, z is the poverty line, and α = 0, P 0 is the headcount ratio or the proportion of
household population whose annual per capita income falls below the poverty line. If α = 1,
P 1 = is the poverty gap ratio representing an index of the depth of poverty, which is the sum
of all individual poverty gaps expressed as a fraction of the poverty line divided by the total
number of households. If α is 2, P 2 is the squared poverty gap ratio representing an index
of the severity of poverty, which is the total number of households. P 2 gives greater weight
to the income shortfall of the poorest of the poor. Poverty line employed in this analysis is