91
in the early 1980s. Annex 3.2 containing data on the area of commercial farms bears this relationship
out as well, particularly for the smallest farm size category (0-79 hectares, figure 3.4). The peaks and
valleys are much more pronounced in the Central and Southern regions despite the fact that, as noted
above, the Southern region only possesses 3.5 percent of the farms in this category and 14.4 percent
of all farms.4.'
Data on number of farms by size category are illustrated in figure 3.5 for the period 1976 to
1988. Associated data on farm area over the same period are contained in figure 3.6. The two smallest
farm sizes—0-79 hectares and 80-199 hectares—are experiencing the most volatile changes. The farms
in the 0-79 hectare range are particularly variable in number and clearly seem to account for most of
the corresponding national jumps and declines. There is a fairly steady decline in their number from
1977 to 1985 and then a phenomenal increase in 1986. One possible explanation is a demand response
by small fanners to price and marketing reforms instituted in the 1980s. A more telling but
analytically elusive possibility is the influence of statistical methodology including possible changes
in the definitions of commercial farms. Small farms are particularly likely to be overlooked in survey
administration. They also may fall in and out of the commercial category to the extent that market and
climate characteristics drive their production below the threshold for commercial farms. Overall there
has been an increase: the average number of farms in the two smallest farm sizes increased from 1,023
in 1976-78 to 1,214 in 1988-90, a 19 percent increase.
Although the other farm sizes tend to follow the national pattern, their lines are extremely
smoothed vis-a-vis the overall trend line and that for the smaller farms. The largest fluctuation occurs
in 1985. Farms in the 200-799 hectare range are relatively stable with 372 in 1976-78 and 397 in
1988-90, a change of 7 percent. Farms of 800-1,999 hectares are also somewhat constant in number
and seem to escape the sharp decline felt in the other farm sizes in 1985. They have, however,
steadily decreased in number over time with 307 on average for 1976-78 and only 256 in 1988-90,
a drop of 17 percent. The largest farms, those with at least 2,000 hectares, also decline in number
over time. The low point is in 1985 with 160 farms and the change in the averages from 1976-78 to
1988-90 is 14 percent, with 225 and 193 farms, respectively.
Data reflecting changes in the area of farms over time tell a similar story. The area of
commercial farms in 1976, according to the Commercial Farm Series Data, was 1,877,490 hectares,
which is roughly 2.5 percent of the 75,261,167 total hectares in Zambia. As with the number of
farms, the area under farming increased dramatically in 1977 to 2,494,960 hectares then fell again in
1978 to 1,434,876 hectares. There was another peak in 1980 of 1,861,280 hectares but only 1,298,030
hectares in 1981—a major drop from which the sector has yet to recover. By 1984, the area had risen
to 1,689,083 hectares where it hovered for the next six years. Surprisingly in 1985, when the number
of farms plummeted dramatically, area did not change significantly with 1,661,825 hectares operated.
By 1990, 1,531,561 hectares were farmed, 345,929 hectares less than in 1976. In fact, the average
from 1976-78 dropped from 1,935,775 to 1,514,233 hectares in 1988-90, a 22 percent decline.
Because data from the early to mid-1970s could not be located, it is not clear if this represents a
significant trend towards shrinkage of farm sizes in the sector overall or if the figures for the late
1970s were simply anomalous.
° Note that the 3.5% figure is taken for 1990 when only 29 farms were reporting. However, in 1989, 415 farms were
reporting, a drop of 386. In fact, the decline in total number of farms from 2,185 to 1,880, a difference of 305, seems to
be largely attributable to the change in this region.