4. THE ROLE OF THE STATE IN THE FUNDING AND DELIVERY OF RESEARCH

A key issue is: what are the appropriate roles of the public and private sectors in fishery research? There is no universal answer to this question.

Generally speaking there are strong arguments for the public sector to fund some research but the arguments tend to be weaker in favour of public sector delivery.

The arguments in favour of public funding for research rest on market failures of various kinds. Thirtle and Echeverria (1994) identify three major failures:

• inappropriability - research is likely to have public good characteristics (see Table 1) so that social gains exceed private gains leading to under-investment in research by the private sector
• uncertainty - since it is not possible to insure against uncertain events, risk-averse firms will tend to under-invest in risky research
• indivisibilities - increasing returns to scale may lead to monopoly power, with the classic result that output (of research) is less than socially optimal.

Table 1 presents the standard classification of goods used by economists. Knowledge is non-subtractable: whether it is in the public good category or the toll good category depends on excludability. This in turn depends on institutional arrangements which differ from country to country. The private sector can be expected to attempt to develop brand loyalty as a way of retaining research benefits but the most important factor is likely to be intellectual property right (IPR) legislation. It is important to note that goods and services can change categories as institutional arrangements change-for instance, fisheries in developing countries that were regulated under traditional management systems were effectively private goods (albeit communally managed). With the breakdown of traditional management they have often moved to the common pool category. On the other hand, moves to implement individual quota systems tend to move fisheries towards the private goods category.

Table 1. The economic classification of goods

The more research is basic, defined by Thirtle and Echeverria (1994, p. 32) as "the disinterested pursuit of scientific knowledge without a specific technological objective in view", then the more private research is expected to be inadequate, and the more public support will be required. This theoretical case for public funding is supported by empirical evidence showing very high rates of return to such research investments.

Similar conclusions are reached by Alston et al. (1995, p. 12) who suggest that where the objective is to correct a market failure, the public sector should focus its support more heavily on research that has a high social payoff but which the private sector has relatively little incentive to support, for example, socio-economic research or base-line research where the benefits may be particularly difficult to estimate and where it may be very difficult to prevent free-riding. However, they also note that government intervention in the provision of R&D should not crowd out private investment, giving three alternative conditions which must apply before government intervention should occur:

(i.) that incentives are such that markets fail to produce socially optimal amount of research or
(ii.) that economies of size and scope in research threaten the competitive structure of markets or
(iii.) opportunities exist for exploiting the 'complementarities' between research, education and extension.

They suggest that these conditions are more likely to apply in developing countries than others and hence the argument for government intervention in the provision of research is likely to be stronger. Along similar lines, Troadec and Christy (1990) argue that the private sector can undertake technological advance but relies on the public sector for research relating to institutional change. This division is not hard and fast, sometimes for example because of risk, public sector funding of technological research may be justified, and sometimes the private sector may invest in institutional research, if it promises profitable changes for them. There may also be infant industry arguments for public research funding in developing countries.

Even if there are some strong arguments in favour of publicly-funded research, the issue of research funding is increasingly under the spotlight. In discussing agriculture, particularly in sub-Saharan Africa, Benyon et al. (1998, p. 1) write "research and extension (R&E) systems ...have typically been dominated by donor-dependent centralised public sector institutions widely perceived to be underfunded and underperforming". Much the same is true of fisheries research.

They identify three reasons why there is now growing interest in finding alternative ways of financing and delivering R&E:

1. "The global reappraisal of the role of the state and (the) associated ideological shift towards economic efficiency and< market-based solutions for resource allocation and service provision" (ibid.).
2. Current policies of structural adjustment and fiscal restraint make it unlikely that resources can be increased.
3. A general perception that R&E has failed to deliver successful innovations in African agricultural technology systems.

The first two reasons certainly apply also to fisheries. The third has to be interpreted rather differently in fisheries, at least in the case of capture fisheries, where the failure has been to deliver exploitation and management systems capable of generating fisheries wealth on a sustainable basis. If anything, successful technological innovations have been rather too frequent in fisheries in situations where adequate management institutions were either absent or deficient.

Benyon et al. (1998, p. 2) also highlight the difference between financing and delivery of research. This crucial distinction is frequently overlooked or ignored because States seek to achieve multiple objectives through research. In particular, there is a tendency to want to achieve high-level research and training simultaneously but these objectives often conflict in practice with the result that neither is achieved as effectively as it might be.

Because they focus on financing, Benyon et al. identify a 2 x 2 matrix to describe the financing and delivery issue. This matrix is presented in Table 2 below.

Table 2. A model of research financing and delivery (developed from Benyon et al., 1998)

Traditionally, cell 1 has dominated, especially where the public sector is funding its own research institutes (e.g. a Fisheries Ministry funding a research institute under its control). In terms of developing away from cell 1, there seems to be much interest in cell 2 where public sector research institutes can (arid are increasingly expected to) fund themselves directly through contracts with the private sector. One potential drawback of this approach in terms of the concentration on the short-term leas already been mentioned.

When considering delivery of research, it may be useful to develop the above matrix somewhat (Table 3). For instance there is frequently an important difference between domestic and foreign research institutes.

Table 3. A modified model of research financing and delivery

Traditionally, research financing and delivery has been dominated by cells 1 and 3. However, Table 3 makes it clear that there are many options that could be considered. It seems that this domination arises at least in part from the mixing of objectives: research and training (or development). However, research is a tradable service just like any other and the law of comparative advantage applies just as strongly. It may well make more sense for a country to consider cells 9, 11, 13 and 15 where the quality of the research output is the only consideration. Issues of training and development may be better dealt with as such.

Table 3 could easily be developed further for example to distinguish between public sector research institutes and universities. Research planners may well find it worth while to develop a table such as this looking first at the relative amounts of funding going to different kinds of institutions and second comparing this with the research payoff. Budget allocations could be altered according to the marginal efficiencies of research investments, although account would also have to be taken of the abilities of different institutions to undertake different kinds of work.