Countries that improved their exchange rate policy and reduced inflation and budget deficits generally increased their GDP per capita growth rates
Fourteen of 28 Sub-Saharan adjusting countries increased the average rate of growth between 1981-86 and 1987-91, while the other 14 had it decrease (figure 1). Ghana, Mozambique, Nigeria, and Uganda had the largest gains--all four exceeded 3 percentage points--while Benin, Cameroon, Congo, and Rwanda had declines of more than 3 percentage points. Eight countries went from negative to positive rates of growth, showing a strong turnaround.
What explains the differences in performance? One hypothesis is the difference in policies. Countries that moved toward macroeconomic stability and improved the relative prices for exports (especially agricultural exports) would be expected to enjoy the biggest payoffs. To the extent that a better policy environment remains in place, the improved growth should be sustainable--possibly heralding the start of a new period of accelerated growth.
An alternative hypothesis is that external factors have driven the better performance--and the worse. Some observers emphasize the recent increase in external transfers and argue that it is a key to the improved performance. Others stress the recent deterioration in the terms of trade and argue that it is a key to the poorer performance.
The conclusion of a recent major study of adjustment in Africa: both policies and external factors were forces in the turnaround, with improved policies more strongly correlated with increases in growth.[1]
Changes in policies affect growth. Countries that improved their exchange rate policy and reduced inflation and budget deficits generally increased their GDP per capita growth rates. The method used in the study divided countries into three groups, depending on whether they had large positive changes in macroeconomic policies, small positive changes, or negative changes, to determine whether the groups differed in their economic performance.
The six countries that improved policies the most had the largest median improvement in GDP per capita growth (1.8 percentage points), and all countries in the group returned to positive (though very low) growth rates. In contrast, among the countries where policies worsened, median growth declined 2.6 percentage points. This group includes the six countries with the sharpest downturns in growth.
The countries that made small improvements in macroeconomic policies had a median increase of 1.5 percentage points. This was progress, but not enough to restore positive GDP per capita growth (this group had a median decline in growth of 0.2 percent a year during 1987-91).
Improvements in macroeconomic policies are correlated not only with larger changes in the rate of GDP per capita growth but also with higher levels of growth. Countries that improved macroeconomic policies the most had positive rates of GDP per capita growth during 1987-91, while those whose policies deteriorated had negative rates. But even for the top performers, the growth rates remained low (at a median of 1.1% a year). Adjustment policies, though instrumental in generating the conditions for higher growth, are thus only part of the solution. Shifting economies to a new growth path requires a good overall policy environment and better use of more traditional development instruments, such as investments in education and infrastructure.
To the extent that macroeconomic policies matter--and they do--getting the exchange rate right is one of the top priorities for short-term growth. Countries that significantly reduced the parallel market premium (by devaluing) and adopted realistic macroeconomic policies enjoyed the biggest payoffs. Dividing countries into three groups based on the real depreciation during adjustment shows that countries that managed large real depreciations (40% or more) enjoyed the largest turnaround, with a median increase in GDP per capita growth of 2.3 percentage points. Of this group, Burundi was the only country whose GDP growth declined. In contrast, countries that experienced real appreciations suffered a median decline of 1.7 percentage points. There is a difference of 4 percentage points in the median changes for the top and bottom groups. The gap between the fixed and flexible exchange rate countries is also sizable (3.4 percentage points), because of the different adjustment strategies followed.
African countries had a clear need for real depreciations in the early 1980s because of deteriorating terms of trade and unsound exchange rate policies. These initial conditions largely explain the effectiveness of devaluations in increasing growth. For countries that already have realistic real exchange rates, there is little scope to further boost growth by depreciating, though gains would be expected from maintaining external competitiveness. But for countries with unrealistic exchange rates, the expected payoffs to real depreciations are large.
The change in GDP growth rates is the most appropriate indicator of whether policy reforms are working in the short and medium term. But over the long term, it is the rate of growth that ultimately matters. And over the medium to long term, the policy stance is more important than the change in policies in explaining the rate of growth. This is because changes in policies are not necessarily correlated with a country's policy stance. Tanzania showed large improvements in macroeconomic policies, and yet it still had a very poor policy stance in 1990-91 because its policies were so poor to begin with. By the same token, a country starting with reasonable policies might maintain an adequate stance even if its policies deteriorate. In Togo, macroeco- nomic policies worsened in the second half of the 1980s (as the real exchange rate appreciated and the budget deficit increased), and yet the country's policy stance remained fair, thanks to a sound macroeconomic framework in the early 1980s.
What is the link between policy stance and the rate of growth? Median GDP per capita growth in countries with adequate or fair macroeconomic policies was 0.4% a year--low but at least positive, and better than the decline of 1.2% a year during 1981-86. Meanwhile, countries with poor or very poor macroeconomic policies had staggering declines in GDP per capita, with the median at -2.1% a year during 1987-91. The median growth rates were also quite different for countries with limited, medium, and heavy government intervention in markets: 1.9%,
-1.1%, and -1.4%, respectively. More than 4 percentage points separated the three countries with the best macroeconomic policies and the least intervention from the two countries with the worst macroeconomic policies and heavy intervention--a powerful incentive for moving toward good policy.
