Lecture 5. Explaining farm policy What we want to learn about this topic Why do countries pursue the agricultural policies they do? To be able to distinguish between the public interest and public choice approaches to agricultural policy analysis Short introduction to the issues Why pursue this line of enquiry? There are a number of reasons for trying to understand why governments pursue the agricultural policies they do: levels of support vary considerably across countries and across commodities - why is this? There appears to be systematic patterns in this variation of government intervention across countries, suggesting that there are common factors at work. For example, a 'stylised fact' is that governments in developing countries tend to tax agriculture and governments in industrialised countries tend to subsidise agriculture. Trying to explain these different patterns of agricultural protection is an issue in positive economics; understanding government behaviour would help to endogenise government behaviour in quantitative models used for projections and policy simulations. For example, if you use an agricultural sector model to project the impact of a ban on the export of beef to Europe (because of BSE fears, say) on farm incomes, you can either assume that government policy will not change in response to the fall in farm incomes (in which case your projections are almost certain to be wrong), or you could try to model government behaviour and predict what the government might do in response to the fall in farm incomes in order to incorporate the likely policy change explicitly into your model (an example of positive economics); some recent research has shown that the way in which decisions are made can have an important influence on the nature of those decisions. Understanding the way policies are made might help to design better rules and thus achieve better policies (an example of normative economics). it is argued that economists might be more effective advocates of reform if they had a better understanding of why governments take the decisions they do. Agricultural economists have used three approaches or 'models ' in explaining farm policy: a public interest model in which government maximises social welfare; interest group models which emphasise collective action by pressure groups; and politician-voter interaction models. The former model lies behind classical welfare economics; the latter two models derive from the political economy literature. We might also mention a fourth tradition, with its roots more in the political science tradition, which sees policy-making as a process and is concerned with describing policy-making as sequential, non-structured behaviour susceptible to random events using an eclectic, non-formal case study approach. Public interest models Also referred to as the public finance model, its distinguishing feature is the assumption of the existence of a social welfare function. In layman's terms, it has some notion of what is the national interest and it pursues policies designed to maximise this function. In these models, governments are referred to as self-willed in that they are pursuing an agenda which represents the social interest. Public interest models, in turn, can be divided into two streams. The first assumes that equal weights are attached to the welfare of each group in the social welfare function and that government intervenes to correct 'market failures' in order to improve social welfare. This is the traditional approach of welfare economics, but it is not well suited to explaining redistributive policies. However, redistributive policies can be incorporated into the second strand of public interest models in which different weights are attached to the welfare of different groups. Redistribution from groups with a low welfare weight to a group with a high welfare weight can now raise welfare in terms of conventional surplus measures. A social welfare function (also called a policy preference function) describes the trade-offs the government makes between the welfare of different social groups. The government is assumed to try to maximise its social welfare function subject to the constraints imposed by the 'economic technology' available to policy-makers, i.e. the different policies and the impacts which they have on the variables of interest to policy-makers. This technology is captured by the idea of the surplus transformation curve which traces out how much welfare one group can gain from a transfer made from another group. The policy choice (i.e. the degree of redistribution) will be determined by the tangential intersection of the policy preference function and the surplus transfer curve (see MacLaren reading). The empirical relevance of this approach is that, on the assumption that observed policies represent an equilibrium on the policy preference function, it is possible to calculate the implied parameters (i.e. the weights the government gives to the welfare of each social group). Armed with a knowledge of these weights, economists could then test the impact of alternative policy proposals on the value of the social welfare function and assess the likelihood of any proposal being adopted. Alternatively, armed with an empirically-estimated preference function, analysts can try to establish the optimal level of a policy instrument (e.g. support prices, deficiency payments) which maximises that function subject to the economic structure of commodity markets. See references to studies of these kinds in the Swinnen and van der Zee survey referenced below. Criticisms of this approach include: doubts over whether a social welfare function actually exists; can the 'public interest' be given any operational meaning? assumes considerable sophistication on the part of governments, that they always choose the most efficient policies to pursue their objectives hard to credit that observed policies are the outcome of an optimising process of this kind. the general result that producers receive a greater weighting than consumers and taxpayers adds little to the understanding of how farm policy is actually made, as there is no model or theory which explains how the welfare weights are derived. Hence the interest in political economy models which try to explain the sources of government behaviour. Private interest models Political economy models are a reaction to the assumption of rational government behaviour underlying the public interest model. Government policies are assumed either to be the outcome of bargaining between interest groups (the collective action model) in which the best-organised lobbies have most influence, or to be the outcome of politicians trying to maximise their support among voters. Governments are referred to as clearing-house governments in this perspective, as their role is simply to hold the ring while the protagonists fight it out. A characteristic of politicial economy models is their emphasis on the self-interest-motivated behaviour of politicians, voters, pressure groups and bureaucrats. The models differ in the importance they give to these different agents and the assumptions about their behaviour. Collective action models try to understand government policies in terms of the structural characteristics of the different groups trying to influence policies. They stem from the work of Olsen, who tried to understand why people join groups. The emphasis is on the importance of organisation costs and free-rider effects relative to the benefits or losses expected from a policy. The theory has been used to explain the strength of farmer organisations and the relative weakness of consumer and other groups in formulating farm policy. Empirical work on interest group models has sought to explain differences in the level of support across countries, or across commodities within a country, in terms of structural economic characteristics (e.g. the share of consumers' expenditure on food, whether the commodity is exported or imported, the gap between average farm and nonfarm income) and factors affecting farmer organisation. Such functions are sometimes referred to as influence functions. The politician-voter model sees policy-making as a political market place in which policies are the outcome of a supply and demand process. Voters and pressure groups demand policies with their political resources (votes, lobbying, political contributions) and politicians respond by trying to maximise their voting support. Within this framework, the main determinants of the taxation-subsidisation patterns of agricultural protection are the differences in relative endownment income and the per cpita redistributive efects of farm policy among different groups. The advantage of the politician-voter model is that it can explain the effect on agricultural transfers of an exogenous change in agriculture relative income, while collective action models have difficulty in explaining why unfavourable market conditions increase agricultural protection. However, various authors have pointed out that, empirically, it is difficult to discriminate between these two explanations. A criticism of political economy models is that they are excessively static in nature and they are not good at predicting the timing of important policy changes, e.g. the MacSharry CAP reform. They focus more on the demand for policies to the neglect of the constraints (budgetary and otherwise) under which policy-makers function. Conclusions It may be wrong to see these approaches as competing and exclusive. Most likely, a mixture of motives may be at work. It is also difficult to distinguish between the altruistic actions of politicians and their self-interested actions. For example, are farmers favoured because their welfare is weighted more heavily, or because they are more politically powerful? From a technical viewpoint, it is always possible to assign welfare weights that describe the outcome of a redistributive policy. But it is impossible to prove whether these weights are determined exogenously (i.e. by a self-willed politician) or endogenously (as a response to interest group pressure) (Brooks 1996). Reading suggestions Winters, A. (1987), The political economy of the agricultural policy of industrial countries, European Rev. of Agric. Econ. 14, 3, 285-304 (informal survey which tries to understand agricultural policy as the outcome of political bargaining between interested groups within a particular economic context) Harvey D., 1997, 'Extensions and political analysis of the CAP', Chap. 8 in Ritson C. and Harvey D. eds, The Common Agricultural Policy, 2nd edition, CAB International. (explains and critiques the political economy approach to agricultural policy-making) Swinnen, J. and van der Zee, F., 1993. The political economy of agricultural policies: a survey, European Review of Agricultural Economics 20, 261-290. (a superb, if long, survey article of the two competing approaches to explaining farm policy) De Gorter, H. and Swinnen, J., 1994. The economic polity of farm policy, Journal of Agric. Econ. 45, 3, 312-326. (discusses the advantages and disadvantages of pressure group versus politician-voter interaction models and comes down in favour of the latter). Brooks, J., 1996. Agricultural policies in OECD countries: what can we learn from political economy models?, Journal of Agric. Econ. 47, 3, 366-389. (an accessible article which justifies interest in political economy models to ensure that policy recommendations are mindful of their political context. 'Once the tricks of policy design are discovered, agricultural economists will then have to make doubly sure that their ideas are sound ones'.) Von Cramon-Taubadel, S., 1992. A critical assessment of the political preference function approch in agricultural economics, Agricultural Economics 7, 371-394 (explains and assesses the use of preference function weights in the explanation and assessment of agricultural policies)