28. It is easy to point to defects in existing policies, but we also need to design better ones: and if governments are to be persuaded to pay more attention to policy issues and to abandon previous inappropriate policies and embrace new ones, they need to be convinced that these new policies will bear the fruits promised. Do policy makers have to rely solely on guesses or hunches about what the effects of new policies will be? Or is it possible to provide them with more objective guidance? Is every country a "special case" that can learn only from its own experience in policy issues, or is it possible to draw on experience already obtained in other countries? I believe that we can draw on experience elsewhere but at the same time I suppose that most of us have become wary of lines of reasoning that say that if a policy works in France/UK/USA/USSR it must work in Africa also. We cannot hope to settle in this paper the issue of the transferability of experience but it is a critically important one which we should briefly consider.
29. Let me try to illustrate it, and since I am an economist I win take the example of pricing policies for meat. Let us imagine (assume) a country with a livestock sector composed partly of commercial ranchers, partly of nomadic pastoralists and partly of mixed livestock - grain smallholder enterprises. Let us also imagine that this is a country where in the past livestock and meat prices remained quite low because of the possibility of cheap imports from abroad while at the same time its meat exports were constrained by disease regulations. Domestic livestock output has stagnated and the import bill for meat has climbed alarmingly with almost all imported meat being consumed in the main towns. Projections, based on past trends in consumption and production, predict very large and growing meat imports in the future. A recent donor-sponsored livestock sector review has called for a change in government pricing policy to raise prices to producers in order to elicit more domestic output to replace imports. The government budget is very tight, so that it will be impossible, by means of a consumption subsidy, to raise producer prices without also raising consumer prices unless marketing margins can be reduced. The domestic livestock and meat marketing system is handled by large numbers of private traders. The country has very inadequate statistics on past levels of domestic production, consumption, trade and prices on which to base predictions of the results of policy changes. Data on growth in the human population are, however, available and reliable.
30. In deciding whether to accept the advice to adopt a price-in creasing policy the government ought to consider a number of points. How rapidly will domestic consumption and production increase in the absence of changes in price? What price changes will occur anyway in the absence of a change in policy? Does the government have the ability to raise producer prices if it wants to' How much can marketing margins be reduced in order to allow increases in producer prices without corresponding rises in consumer prices? By how much will an x% rise in consumer prices affect meat consumption (price elasticity of demand)? By how much will a y% change in producer prices affect production for sale (price elasticity of supply)? What will be the income and nutritional effects on different groups of consumers and producers of a change in price policy for meat?
31. These points need consideration by the government in order to compare this (price-increasing) change in policy either with making no changes at all or, more sensibly, with some other alternative policies which might save an equivalent amount of foreign exchange. Whereas ideally one would like to base predictions about the effects of a change in price policy mainly on past experience of price policy changes within the same country, often this -may be impossible either for lack of data on the past or because the country has itself recently altered so far and so fast that predictions based on its own experience may no longer be relevant. It may then have to rely on experience elsewhere.
32. I am going to be brave and hazard some opinions about the extent to which analysis of experience in other countries on these points will provide useful guidance to countries in sub-Saharan Africa which do not have experience of their own on which to rely. These opinions are summarised in Table 1 but only apply subject to the specific assumptions set out three paragraphs previously.
Table 1. The predictability of events in one country on the basis of experience in other countries.
|
Event |
The extent of predictability |
|
Future changes in quantities (assuming no price change) | |
|
(i) Produced |
X |
|
(ii) Consumed |
Ö Ö * |
|
Future changes in prices (assuming no policy changes) |
X |
|
Ability of government to affect prices |
?Ö |
|
Possible reduction in marketing margin |
ÖÖ * |
|
Effect of price changes on quantities | |
|
(i) Produced |
?Ö |
|
(ii) Consumed |
Ö |
|
Effects on different groups of producers and consumers | |
|
(i) Income |
Ö |
|
(ii) Nutrition |
Ö |
Ö = Direction of change predictable but size of change only poorly predictable.Ö Ö = Both direction and size of change are predictable with reasonable accuracy.
* = In these cases the necessary data collection and analysis are, for the most part, already complete. In the other cases marked Ö it is affirmed that such studies could be successfully carried out.
X = Neither the direction nor size of change can be successfully predicted on the basis of experience in other countries.
33. On most of the points I am fairly confident about the extent to which experience in other countries can usefully guide policy and predict events in a particular country. I have put a question mark, however, against two points: the ability of government to affect prices; and the effect of price changes on the amount produced. In principle a cross-country study should be able to reveal the extent and the conditions under which a government can enforce a policy, e.g. on meat and livestock prices. However, it is a capability which is frequently overestimated and the science of assessing administrative capacity is a young one; so perhaps some caution is needed.
34. We can illustrate the kind of discrimination which needs to be exercised in drawing on experience in other countries from the case of the effect of price changes on the quantity of livestock and meat produced for sale (supplied). In brief, some controversy exists as to whether price rises are likely, in Africa, to lead to more or less livestock being supplied for slaughter. Some people argue that African livestock producers have limited needs for cash and a rise in price will merely enable them to sell less animals to obtain the same amount of cash, and that there are other factors which cause livestock owners to want to increase their herd size. Others argue that livestock owners in Africa respond in the same way as farmers do everywhere, and if prices rise they increase output. Others argue that although in the long term livestock holders will plan to increase output, in the short term, for sound economic reasons, they may withhold animals from market; and yet others argue that African pastoralists may not increase output in response to a price rise because it is not profitable for them to do so. (For further discussion see Low, 1980; Jarvis, 1980; Sandford, 1983; Ariza-Nino and Shapiro, 1984; Ndzinge, et al, 1984).
35. If we are to proceed beyond this controversy, we need to distinguish between different kinds of livestock owners who may behave in different ways and study each kind separately. For example, many traditional African pastoralists in remote areas lack access either to consumer goods on which to spend extra cash or to inputs purchaseable for cash (e.g. feed concentrates, veterinary medicines) with which to increase output. The response of such pastoralists to higher prices for their livestock is quite likely to be to sell fewer for they have other uses to which livestock can be put. The response of commercial ranchers, however., (depending on how permanent they think an increase in prices will be) may be to sell off more animals quickly, to hold back animals temporarily from the market in order to put more weight on them, or to convert from a mainly breeding to a mainly fattening operation or vice versa. The response of mixed grain - livestock smallholders will in turn be somewhat different, and in drawing on experience from other countries we need to be very aware of the systematic similarities and differences between production systems in different countries so that we can draw only on experience which is relevant to the situation of our own countries.
36. This example of policy in respect to prices will, I hope, be useful in illustrating the extent to which experience in other countries can help policy makers. The crucial issue is how closely the natural and social aspects of the production, consumption or marketing systems in the individual country match those of other countries from whose experience it is hoped to learn. Obviously there will be no exact match and to that extent the causal sequences will not be identical either. For a variety of reasons the match is generally likely to be closer in the production system between an African country and other African countries than between an African country and countries in other parts of the world: but as the pricing example shows even then one must have close regard to the details of the production system. Experience in other continents may be more applicable to Africa in respect of consumption and marketing systems than it is for production; even so, considerable caution must be exercised.
37. By implication the example of pricing policy also shows the extent to which one particular kind of research, research which compares experience in different countries ("cross-country" research), can be useful to the same policy makers. At the moment, because of lack of policy-oriented research, policy makers often have to guess at the probable results of the policies they propose. We cannot ever expect to eliminate entirely the role of guesswork in policy-making. But good research, both cross-country research and research which analyses in depth the experience of particular countries, can greatly reduce the amount of guesswork which has to be done and so improve the reliability with which the results of different policies can be predicted. Here at ILCA we are conducting a number of both cross-country and in-depth, single-country studies on policy issues.
