In general, this report deals with analyses that can be carried out on the basis of published or already tabulated sources of data. However, recently there has been more emphasis on small-scale, inexpensive surveys for the purpose of generating timely data that may be used to monitor the effects of policy changes, so it may be appropriate to review some considerations that can guide the design of such surveys.
The restrictions on a timely information survey are that it be small enough to be processed within a week, by hand if necessary, but preferably with micro computers or laptop computers; that it be repeated at frequent intervals; and that the information it captures should be of interest to policy makers. These requisites also dictate that the format of the survey be simple.
A policy monitoring survey for specific questions also is quick to process but is somewhat different in structure: its questions are more tightly focused and the sample may be purposive--emphasizing only certain groups of farmers, for example. The nature of the questions in a policy monitoring survey depends in part on its periodicity. An annual or seasonal survey can pose questions concerning the last growing season, such as: Were inputs delivered to farmers in a timely fashion? What prices were paid for the inputs? Which input suppliers were used? Was agricultural credit made available quickly enough and in sufficient quantities? What prices were received for outputs? If the output was sold to a public agency, was the payment for it made in a timely fashion? (Ccama, Heckerman et al. provide an example of such a survey for Peru, based on about 200 farmers stratified by region.) Cross tabulations of the survey's results can yield interesting information on differences in performance by input supplier, by purchasing entity, by region, by the farm's distance from metropolitan centers, by farm size, and so forth.
A seasonal survey also can attempt, with a larger number of respondents, to find out if the area planted has increased in the current growing season, with respect to the previous one. Barring unforseen crop failures due to natural factors, and adjusting for any major changes in technologies of production, this information is a good indicator of the trend in production in many countries. It also tells policy makers whether farmers are responding to changes in policies on pricing and subsidies.
A monthly survey is appropriate for monitoring prices, especially at the retail and wholesale levels. With products like beef cattle, it is difficult to stay abreast of producers' decisions if price data are collected only once a year. For basic grains, prices at the wholesale and retail levels respond to inventory levels as well as to production levels, and monitoring them on a monthly basis is vital to projecting the need for imports. In some cases, price data are collected on a weekly basis, but generally in agriculture the monthly availability of price data, processed and reported quickly, would mark a considerable advance over currently available information.
Timely information surveys should be regarded as opportunities to bring the policy makers into closer and more frequent contact with what is happening in the farmers' world. Viewed in this way, it is surprising that they are not used with greater frequency in developing countries. In a thoughtful review paper on data issues, Murphrey, Hyslop and Ferguson (1984) present a strong case for more attention to timely surveys.
The main potential hindrance to their usefulness stems from delays in processing the surveys. It is crucial that the survey results be processed and reported in less than a month, in the case of seasonal, quarterly, or monthly surveys. Normally this kind of speed can be achieved only by an independent survey unit, free of the pressures and bureaucratic restrictions that are faced by line agencies of the government. Again, micro computers (with standard software for data management) can play a crucial enabling role for this kind of survey.
Although it is not practicable, or necessary, to compute effective protection rates or DRCs on a monthly basis, if domestic price surveys are available it is feasible to update nominal protection rates at least quarterly, using published trends in international prices to update the past year's c.i.f. or f.o.b. prices, and using the same farm-to-border marketing and processing costs as used previously (the same in real terms). This would be a very useful procedure, as so much of the sector's behavior hinges on protection rates; at the very least, nominal protection rates should be calculated annually, though this is rarely done.
Recent information from a market price survey also permits the updating of calculations of the intersectoral terms of trade. Often policy makers become aware of a change in the terms of trade a year or more after the farmers themselves perceive it.
By the same token, normally there is considerable interest, on the part of producers as well as government bodies, in the latest international price quotations, especially for major exports such as coffee, sugar, rice, beef, and the like. Producers' associations often provide their members with this information on a regular basis. Government bulletins of agricultural information should also include this information.
A bulletin of timely statistical information can be quite useful to track the sector's performance and document the concerns of producers and consumers as they arise. Normally it will generate interest beyond the narrow circle of policy analysts and policy makers. At intervals that are feasible in light of the information collection procedures, a bulletin also can contain data on input prices, credit volumes, trade volumes, and production levels themselves. It can be an outlet for disseminating on a regular basis the kinds of indexes discussed in sections 3 and 4 above, plus other information on an occasional basis.
Policy analysis tends to express itself in the form of in-depth analytic reports, completed at irregular intervals, and it does not give enough emphasis to the provision on a systematic basis of timely information and the simple indexes and coefficients that are discussed in this report. Timely surveys of farms and markets can provide the backbone of new information for regular publications, and once established those publications can, of course, be forums for occasional reviews of a more probing nature.
In any technical field, communication is as important as research, and yet agricultural policy analysis appears to have paid very little attention to questions of communication and dissemination of results beyond the audience of researchers. The relatively simple calculations and interpretations described throughout this report have the aim of fostering a wider (and more timely) understanding of what is happening in the agricultural sector. Without giving some consideration to means of communication, these exercises can become rather vacuous.
