2. Agricultural Performance

Table 5 gives the estimated growth rates of agricultural value added per capita. Agricultural value added is contrasted with GDP for each country in figure 1. For all 3 periods and both groups, the rate of expansion and contraction of agricultural value added per capita was less than for GDP per capita. That is, the agricultural sector expanded more slowly than the economy during the early and late growth periods and contracted less during the recession period. Agricultural performance in the rapid growth countries is notable both for the rapid rate of decline during the recession and its rapid rate of growth in the recent period.

Agricultural value added per capita, while a useful indicator of sectoral performance, should be considered in the context of other agricultural performance indicators. If countries experienced significant rural-urban migration and/or restructuring of the economy (away from agriculture), value added per capita may fall despite a healthy economy and a healthy agricultural sector. To account for economic restructuring, we consider labor productivity in the agricultural sector as a second measure of agricultural performance.

Table 6 shows labor productivity measured by the growth rate of agricultural value added per capita of rural population. The growth in labor productivity was in all cases greater than the per capita agricultural growth rates - growing faster on average at a rate of 2% per year. So while total output for the sector could barely keep up with the growth in population, labor productivity in the sector was increasing, implying economic restructuring, rural out-migration, and higher average rural incomes. Agricultural labor productivity growth rates were comparable to the macroeconomic growth rates (table 1) in the growth periods, but were higher than the macroeconomic growth rates during the recession period. In other words, the average output of labor in agriculture grew during the recession while the average output per worker in the whole economy fell.

Relating growth in agricultural value added to GDP growth in Figure 2 shows that the two are consistently related with an elasticity of the order of 0.6 through all three periods. Hence, a 10% growth rate in GDP is associated to a 6% growth rate in agricultural value added. Thus, agriculture has a slower growth than the overall economy when the economy expands and a slower rate of decline when there is a recession. This is the well known phenomenon of the hysteresis of agriculture in the economic cycle. Only with slow growth in the recovery period did the growth of agriculture exceed that of GDP, with an elasticity of 1.9. What does this say about the performance of agriculture in the context of economic reforms? Two observations. One is that the general context of economic growth is indeed fundamental to agricultural growth. Success of the agricultural reforms has thus been tied to success in restoring overall economic dynamics. However, rapid growth has been associated with rapidly appreciating real exchange rates (an annual rate of -6.8 under rapid growth compared to 0.6 under slow growth. See Table 2), and this contributes to dampen the growth transmission effect on agriculture compared to slower growth. There are some clear growth failures for agriculture in slow growth countries, specifically Mexico, Venezuela, and Bolivia. Otherwise, however, successful macroeconomic reforms and the associated reforms in agriculture have been able to restore a modest level of agricultural growth, certainly higher than the growth which agriculture was achieving in these same countries under import substitution industrialization and debt accumulation. The second observation is thus that the reforms have not been without effect, even though performance is still modest. We thus do not find support for the pessimistic assessment of the impact of policy reforms on agriculture espoused by Gardner.