Farm price data: are we deluding ourselves?

by B.A. Bishop

This article is intended to raise questions concerning the kind of published farm price information put out by most governments, whether of developed or developing countries. In the former, one purpose of providing such information is to give farmers an idea of the prices they might receive when they market their produce' thus enabling them to improve their marketing plans. I contend that the relationship between such published prices and those actually received by farmers is largely unknown, and that it is probably rather weak. If this is so, it raises questions concerning the manner in which farmers receive their price perceptions, in both developed and developing countries, and whether this affects their marketing practices. It also raises questions concerning the validity of the conclusions that can be drawn from price analyses. Finally, I will include some observations about the way in which the present state of affairs is changing.

It may be noted that the question of how published prices relate to actual prices is not often addressed' since it is commonly assumed that farm price information is clear and unequivocal' and that the greatest difficulty is one of definition: for example, what is the definition of "farmgate" prices and what adjustments have to be made to recorded or published prices to bring them into accordance with this definition? Typical of statements to this effect is this: "The prices at which goods change hands, and the total quantities at which they are traded, are relatively easy to discover." On the contrary, neither the prices at which goods change hands, nor the total quantities traded at any one price are known or easily ascertainable.

One may identity a number of "locations" where seller/buyer transactions take place, where prices are determined, and where ownership is transferred. Transactions may take place, for example, between:

- the producer and the wholesaler, who may be either a national or a regional wholesaler, in a central market

- in a regional market, between the producer and the regional wholesaler - locally, between the producer and the retailer

- or, again locally, between the producer and the consumer directly.

One may suppose that for each of these "locations" there is a characteristic set of prices which vary from location to location partly for such unavoidable reasons as transport costs, partly because of market imperfections. In addition, the manner of the sale may be such that prices are determined at the moment of sale by bargaining or by auction, or they may be determined by a previously agreed contract. It is clear that when a farmer enters a market location he does not know for certain what price he will receive, unless he has a previous contract, but in this context government price fixing is analogous to a contract. Nevertheless it is important for a farmer to know what prices are being established in the market location.

In the short term, he will need such information simply to be able to decide whether or not to sell his produce. If costs of production to the time of harvesting are considered "sunk" costs, he will need to know whether he can at least cover his marketing costs. In the longer term, he needs to know something about the general level of prices in the market, which will enable him to judge whether the prices he himself actually received in the past were acceptable in comparison with what other farmers were receiving and whether or not he can surpass prices previously received by improving his marketing practices. The important point is that the farmer needs to know the market price as a basis of comparison by which he can make management decisions.

The farmer's sources. But as far as the farmer is concerned, the definition of the market price is not really too significant so long as it bears a known relationship to his own prices. The market price may be the price received by a neighbour for a particular quality and quantity of crop, or it may be the price he received from a particular known merchant the last time he went to market. In fact, in market systems where the diffusion of price information is not much developed, these may be the two main sources of price information for an individual farmer. In other words, a farmer is not so much interested in generalized market prices as in an indicator price that he can relate to his own circumstances.

Even where there is government intervention in the market, it is rare that published price information indicates the actual price received by the farmers. There are so many variations on the theme of intervention - floor prices, ceiling prices, guide prices, fixed prices for a certain quality and/or quantity, to which are applied rebates, taxes, or subsidies - that the published price is only an indicator of what the farmer actually receives. In most cases, numerous adjustments are necessary to arrive at this figure. In principle the information for this should be available in government records.

Thus the case is somewhat different for these prices than for those where the basic data are not collected.

In fact, in Western Europe, generalizing on the basis of observations of the situation in five countries, namely the Netherlands, France, Italy, Spain, and (to a lesser extent) the UK, it seems that published data may be more fairly described as indicator prices than as anything else. The reasons for this may be understood by examining more closely the difficulties in the way of publishing price data which reflect better the general price situation in the markets, for instance weighted average prices. In order to be able to calculate these it would be necessary to know both the price and the amount involved in every single market transaction. Merely to state this requirement is to make obvious the difficulties in collecting comprehensive data on market prices. In the first place, if price data are to be collected, the transactions must be observed. In general only transactions taking place in organized wholesale markets can be easily observed. In Europe, on the whole, price data are only collected in such markets. One may suppose that direct producer/consumer sales are not important for most commodities in most European countries and such importance as they once had is tending to diminish. On the other hand, the prevalence of sales under contract, which escape the organized marketing system altogether, is tending to increase, and there is very little good information, for any commodities in any countries, about the proportion of total production which is disposed of in this way. The situation is somewhat paradoxical in the sense that in some cases the contract prices are based on the published prices, which themselves relate to only a small proportion of the total transactions and for this reason may be exceptional. Thus the exception becomes the rule by which prices are set.

In the second place, since price variations take place throughout a season at varying speeds, it follows that price and quantity observations must be made throughout the season, the intervals between observations being determined by the speed and/or irregularity of the price movements. In fact, since prices vary even during a single day, one might suppose that there would need to be several observations during a day.

Third, the existence of regional variations implies that there also need to be observations in various parts of a country, exposed to the same difficulties as just mentioned.

The Netherlands example. In general, it appears the price data in European countries do not match up to the problems posed above. In the first place, the observations are made where transactions are relatively easily observable, namely the national or regional wholesale markets. Second, they are mostly "on-off" observations, for a particular day, with seldom even the range of prices being given. Third, no attempt is made to associate the price observations with the quantities sold at those prices. More commonly, if weighted average prices are calculated, the weights depend on the total quantity of that particular commodity passing through the market in the course of a year. Finally, though it is difficult to be sure about this, the observations of prices do not seem to be made in a very scientific way, being more in the nature of an estimate of a "fair average price" made by an experienced merchant.

The cause of these shortcomings is presumably the cost of doing anything better, though one must admit that it is difficult to know what to suggest by way of improvement. In this connection it is perhaps relevant that the best data on prices of fruits' vegetables, and flowers are those produced in the Netherlands, where records of prices and quantities for all transactions are a by-product of the attempts to improve the institutional arrangements for marketing these products.

The question is whether any improvements which might be achieved in the price data would be commensurate with the extra cost involved. From the point of view of one of the more important groups of users of price data, namely the farmers, as has been noted, their purposes are probably adequately served by indicative prices, and they would not have a requirement for more soundly based general price data. Agricultural policy makers presumably are interested both in the dynamic effect of changes in agricultural prices on production and incomes, and in the relative level of incomes in the agricultural sector and other sectors. For the former, movements in indicative prices would perhaps be a sufficient basis for analysis, but for the latter what is required is to know the real level of agricultural prices. However, the required information about agricultural sector incomes can also be derived from national farm income surveys, and this is in fact done in the UK and other European countries.

In general, this article has suggested that if accurate farm price information were to be published, then many more items of price data would have to be collected and processed than is the case at present. This is now certainly within the bounds of technical and financial possibility for developed countries, through use of computers to record each and every market transaction as regards both price and quantity. Some steps have already been taken in this direction. The irony is that at the same time as the technical possibility is coming closer, the growing prevalence of direct contracts between farmer and processor, which fall outside the scope of organized markets, makes the achievement of this objective less and less reliable.