in most countries.10 The stagnation in yields and per capita production is especially
noteworthy in Zimbabwe, Kenya, and Zambia, where smallholders’ use of improved maize
hybrids and fertilizer per hectare are the highest in Africa.
However, grain production per capita is increasing in countries such as Ethiopia, Mali, and
Uganda. Each of these countries has implemented important macroeconomic and sectoral
policy reforms over the past decade, but not clearly more so than the other countries listed
in Table 1, where performance has been very mixed. Our conclusion is that the weight of
the evidence indicates that there is a moderately positive but highly variable correlation
between agricultural performance and the extent to which countries have followed donor
specified sectoral and macroeconomic policy adjustments. The most important (but only
partial) determinant of the response of agriculture to reform has been the extent to which
agriculture was supported or depressed prior to reform. In countries where state activities
clearly depressed agricultural production prior to reform (as in Ethiopia and Mali),
agricultural productivity growth in the reform period has been encouraging, despite the
fact that numerous policy and institutional barriers to productivity growth still need to be
resolved. These policy/institutional barriers are examined in Section 3. By contrast, in
countries where smallholder grain production was, on net, supported by state intervention
(as in Zimbabwe and Kenya), agricultural productivity has been stagnant in the post-
reform period, despite the fact that certain aspects of reform have been clearly growth-
promoting (Jayne and Jones 1997; Jayne and Argwings-Kodhek 1997). In addition, the
supply response to sectoral and macroeconomic policy reform seems to have been greatest
in those countries where the reforms were coupled with long-term investments in
agricultural technology and human capital development. For example, Savadogo,
Reardon, and Pietola (1995) show that prior investments in animal traction and access to
infrastructure were important determinants of farmers’ supply responsiveness to output
prices in Burkina Faso.
• Price instability: The transition from state-controlled grain prices to market-oriented
prices has exacerbated intra-annual and inter-annual price instability in some countries
(Barrett and Carter 1997; Kangasniemi et al. 1993). The magnitude of price instability
10 Some of the food output decline in the 1990-94 period can been attributed to the 1992 drought, the
worst in decades. But when the effects of the drought are removed (see note a, Table 1), the general picture
remains intact, and clarifies that the decline is not simply a transitory phenomenon due to drought.
11