ANNEX 1: THE LINKAGE BETWEEN THE SUBSIDY AND ITS EFFECTS

Consider, first, the determination of the effect of a fishery subsidy on the size of the fish stock. The problem is to determine the magnitude of the shock (the subsidy) being exerted on the system (the fishery) and to trace the effects of this shock through the system to the point where it impinges on the fish population. The decision chain would follow the following pattern:

A. Specification of the programme

B. Implementation of the programme

C. Stimulus provided to the firm

D. Firm’s reaction to that stimulus

E. Effect on fish stocks of the firm’s reaction to the stimulus.

Take the case of the United States’ Capital Construction Fund (CCF) which was previously described. Step "A" of the decision chain requires that the requirements for a firm to benefit from the CCF programme be specified and step "B" requires knowledge of the extent of the utilization of the programme. The stimuli, in step "C", are twofold and operate at distinct points in time. At the time the contract between the fisherman and the government is signed, the fund is established. The fisherman knows that any part of his or her fishing income can be protected from income taxes in the CCF. Does the fisherman increase his or her capital stock (i.e. expand his or her fishing operation) at that time? If so, then any resulting profits can be protected from taxes by being placed in the fund. If, in step "D", we assume that the fisherman expands his or her operations as a result of this stimulus, the next question is: does, or when does, the fisherman satisfy the terms of the contract with the government and purchase the new vessel? The firm therefore reacts to a double stimulus: first, the firm expands its fleet in response to the stimulus in anticipation of increased profits that would permit it to accumulate funds in the CCF; and, second, later, the firm uses the CCF fund to purchase a new vessel or modernize the old one. An important question is by how much does the fleet (or, more precisely, the catching capacity of the fleet) expand in reaction to each round of CCF stimulus. An even more critical question is by how much does fishing effort increase, since it is fishing effort and not vessel capacity that directly affects the catch.^[165] Given the fleet expansion, the next question to be answered, in step "E", is: "what is the effect of the expansion of the fleet on the fish population"?

This kind of analysis, "A" to "E" must be applied to each identified subsidy. This generic analysis is applicable to any type of subsidy, regardless of how "subsidy" is defined. If a sufficiently broad definition of subsidy were applied, then, for instance, a general programme of investment tax credits (applicable to all industries) would be a subsidy although it clearly falls outside the realm of the WTO. There is no essential difference in the nature of the analysis of the subsidy’s effects, be the subsidy an investment tax credit, the CCF, a regulation restricting fishing, or any other. Although here we have emphasized the effects of subsidies on sustainability, similar arguments could be made concerning the effects of subsidies on international trade.

Among the approaches to determining the effect of the subsidy on sustainability suggested by FAO’s Expert Consultation at the end of 2000, were simple qualitative models and econometric estimation. The consultants did not elaborate.

Simple qualitative models require the least analysis and the least amount of precise data. Therefore, although the results are of necessity fairly crude, this is the approach that will generally be taken in the absence of well funded, long term projects. As can be attested to by the Milazzo report, the APEC (PricewaterhouseCoopers) study and the OECD analysis of government financial transfers, a major project is necessary just to adequately satisfy the first two of the five steps in the decision chain. And since hard data are at least conceptually available for these two steps, satisfying them is the easier part of the project. This is not to imply that satisfying the first two steps requires anything other than great diligence, much effort, much time, and adequate funding.

Let us continue with the CCF example. We know, or should be able to determine, the specifications of the programme, the amount of money deposited in the funds each year, and the amount withdrawn as per the fishermen’s contract for vessel construction or modification. That takes care of the first two steps.

The third step is easy and verbal: the stimulus is to build vessels. There are counterstimuli in that new vessels are banned in certain fisheries, meaning that to all intents and purposes the firm cannot take money out of the fund. This constraint would have to be considered in any evaluation of the CCF programme.

The remaining steps involve three critical questions: what extra capacity does the CCF programme add to American fishing fleets each year; what is the increased fishing effort; and what is the effect of this increase on fish stocks.

The answer to the first question should be available from government records. Presumably, when a firm wants to withdraw funds from the CCF to buy or renovate a fishing vessel, the government is told what is being done. If the vessel is being modified, the government should know what changes are being made, if only to determine whether the modifications satisfy the terms of the original contract. If a vessel is being replaced, the net increase in fishing capacity is what is wanted and should be estimable.

The second question might be answerable through the study of vessel logs. By how much did effort change? In the absence of suitable econometric work, some very broad assumptions would have to be made to determine, ex ante, the anticipated change in fishing effort.

The final question, satisfying step "E", depends on the state of the fish stock that is to be fished with the new or modified vessel. Since vessels can move about from one fishery to another, assumptions must be made about which fishery is to be the object of the vessel. A population dynamics relation can then tell how much of the fish stock will be caught. Combined with other information, one should be able to judge the effect on the sustainability of the stock.

Arnason^[166] has developed a theoretical, generic, model that can be adapted for applying the above approach. The focus of his analysis is profits: the changes in profits resulting from the government programme; and the response of the firm to the change in profits. Profits are defined in economic terms, so that costs include not only explicit cost outlays but also the opportunity cost (in excess of the explicit cost outlays) to the firm of harvesting fish.^[167] Expected profit stimulates changes in fishing effort.^[168] Changes in fishing effort interact with fish populations. We have, in effect, described the five steps in our decision chain.

The relationships Arnason uses are the following:

A. Catch is determined by fishing effort applied to a fish stock

B. "Outlay" costs are determined by effort

C. Revenues are determined by catch

D. Fishing effort is determined by profits (or expected profits)

E. Fish populations are determined by their natural growth less the catch

F. Profits are determined by revenues less outlay costs less opportunity costs, all of which are affected by subsidies.

The model can be condensed into three functions: profits, fishing effort change, and fish population change. Arnason performs no statistical estimation but assumes "reasonable" values for the parameters (including prices and unit costs) and draws conclusions concerning the effects of subsidies.

Arnason’s approach can be further developed into an econometric model which would involve the development of a series of equations describing each aspect of the fishery. Many an economists’ debate centers on elasticities, the sensitivity of responses to stimuli. Theoretically, it may be clear that a particular response occurs if certain pressure is put on the economic system. Yet the question remains of whether the response is relatively large, substantial, important or whether it is small, insubstantial and inconsequential. "Reasonable" values may be assumed for parameters, but what is reasonable and what is true may be two totally different things. Econometric estimation, done properly with adequate data, enables one to evaluate the nature of the response. Such a model would be larger than that of Arnason, and the parameters would be determined for the most part by statistical estimation. When statistical estimation proves impossible, historical allocators could be used. The data requirements are extensive, but the model would be able to provide good descriptions of the fishery under study and could be used to provide simulations of the effects of various shocks, or subsidy programmes, on the fishery.^[169]

An integrated fisheries econometric model would include a marketing sector (essentially a demand model for the products produced in this fishery), a processing sector (showing relative factor shares and profitability given the revenue and cost structures of the industry), and a harvesting sector (showing the relationships between effort and catch, and catch and fish population).^[170] Some subsidies simply increase or decrease revenues or costs and these can easily be built into the cost and revenue functions. Other subsidies, such as tariffs, import quotas, vessel buybacks and restrictive fishing regulations, to cite just a few, are more difficult to incorporate into the model but, with a little ingenuity, it can be done. Since the net effect of a subsidy on sustainability is the effect of the subsidy less the restraining effect of effective fisheries management, the model should also incorporate both the effects, and the costs, of fisheries management.

Similar modeling exercises can be used to evaluate the effects of fishery subsidies on international trade and economic growth.