versión On-line ISSN 0717-6821
Cuad. econ. v.37 n.112 Santiago dic. 2000
IN SEARCH OF A POSITIVE THEORY OF ECONOMIC POLICY:
LESSONS FROM LATIN AMERICA
Economists have a normative theory of economic policy. They have traditionally seen themselves as having a framework, on the basis of which they can determine whether a certain policy increases social welfare or not. Hence they can say how economic policy should be, to increase social welfare. When economists are asked why policy usually deviates from this optimum, for a long time they have said it is due to "politics". However, without an understanding of how "politics" works the economist may find itself in the equivalent situation of preaching to a monopolist to lower its price. Therefore, economists have felt the need for a positive theory of economic policy. To develop it, economists need to ask questions such as what systematic effects do culture, endowments, external shocks, special interests, and political institutions have on economic policy. This issue is a collection of papers that explore this area, an area that has had a substantial increase in interest during the last two decades. That is, the attempt to develop a positive theory of economic policy.
In this introduction I will briefly describe and summarize some of the literature in what is now called "political economics"; analyze the different strands found in the literature; and locate the four essays of this volume within these strands.
THE LITERATURE ON POLITICAL ECONOMICS AND THE PAPERS IN THIS ISSUE
In the last few years this area has grown greatly (see the references). It already has some basic insights and robust theoretical results, and even more recently there have been empirical studies that tend to confirm these insights. This literature has different strands, and these are reflected in the essays included in this issue.
There is on the one hand an old tradition, that starts with Weber (1958), attributing certain economic policies (and in general success or failure in the economic realm) to "cosmological beliefs" or culture, or to the factor endowment of the country: issues that are out of the control of politicians or economists and that have great endurance. Lal (this volume) is located in this tradition.
In this literature the key question is what explains long run success or failure in development. Are the West's political and economic institutions and valuesdemocracy, the market, protecting human rights, egalitarianismthe only route to prosperity? Some claim, alternatively, that it is unique Asian values which are responsible for the East Asian economic miracles. Who is right, and can we say anything useful about the institutions which promote development? That is the central question Lal and others have sought to answer in the context of the formal institutions as embodied in democracy and the legal infrastructure for a market, as well as the informal institutions of culture as embodied in the family1. Lal, for example, has argued that what matters for intensive growth is that the market should be allowed to function, and that it can be adopted without the need to adopt the two other Western institutionsits family or its polity.
This literature argues that institutions are determined by historical factors such as culture or the initial resource endowment. The appearance of the institutions of capitalism depends, among other things, on the cultural view of the mission of the State in society. There are two views of the State: one in which it merely facilitates individuals to pursue their own ends; another which views the State as the manager of an enterprise seeking to use the law for its own substantive purposes, and in particular for the legislation of morality (see Oakeshott 1993). In the Third World, according to Lal due to religious beliefs, the State is seen as an enterprise, which has led to dirigisme and the control of the market. But dirigisme bred corruption, rent-seeking, tax evasion and illegal activities in underground economies in the same way as it did in the mercantilistic societies (see Hecksher 1955). The key insight here is that institutional development is a form of cultural evolution.
Then there is a more recent strand, that enjoyed a surge in the post oil shock era, when many countries found themselves straitjacketed by institutions that were not functional, but unable to reform them. This literature consists of research that attempts to figure out what conditions are necessary for an LDC to do needed structural reforms (see for example Haggard and Kaufman 1992). The papers by Edwards et al. and Tommassi et al. (in this volume) are part of this literature. Their papers respond to questions such as: Why did reform occur in Colombia? Why did reform occur in Argentina? Why did reform take the form it did in Argentina? This literature is usually very empirical, based either on regressions or on case studies.
There is a also a theoretical tradition, that starts with the time inconsistency literature in the late seventies and early eighties. This literature attempts to respond questions such as what institutions will resolve the credibility problem at the core of the time inconsistency literature (this is the normative strand of the political economics literature). This literature has an empirical side to it, pioneered by Alesina, Persson, Tabellini and others (see the references), that attempts to actually verify the effects of institutions on economic policy. It also has a positive side, which tries to determine how political institutions determine economic policy. The paper by Sapelli belongs to this strand of the literature. It asks why the structural reforms that occurred in Chile in the eighties were not changed by the democratic government that followed. It asks which institutions contributed to the permanence of the structural reforms. It is in some way tied to the previous strand, being the logical sequence: once structural reform occurs, then why would the new policies be credible. How can we make them credible?
SOME INSIGHTS OF THE POLITICAL ECONOMICS LITERATURE
A broad survey of the political economics literature can be found in Persson and Tabellini (1999, 2000). I do not intend to give a full account of what is a rapidly expanding area, but I believe it is useful to summarise some of the key insights that this literature has to offer.
This strand of the literature grew out of the key credibility problem or time inconsistency problem whose importance was pioneered by Kydland and Prescott (1977) and Barro and Gordon (1983). The key problem is that whenever a stock is accumulated in expectation of the permanence of a certain "optimal" policy (be it tax or monetary policy), it becomes optimal for a policy maker to renege on its previous promise and increase the tax on the stock (be that an investment or money) through taxes or inflation. Since rational economic actors know of this time inconsistency, they will invest suboptimaly or will hold less money stocks than optimal, resulting in less growth or more inflation. Unless a policy maker can make a credible commitment to the initial "optimal" policy rule, nobody will believe him. And so the economy will run along a suboptimal path. This is true independent of whether the policy maker actually (cynically) plans to reverse his policy or not. This is a key point. Even benevolent policy makers should ask to be constrained. Both predatory and benevolent policy makers are subject to the costs of the inability of committing to a first best policy rule.
Since this literature is a strand of institutional economics and transaction cost economics, the assumption is always that institutions have to be designed to prevent opportunistic behaviour, and that if contracts permit opportunistic behaviour, then this behaviour will occur. Hence one cannot rely on the benevolence of actors. This dilemma is only solved by an institution that is costly to modify. This institution then benefits everybody.
An institution that makes a commitment credible clearly has the cost of reducing flexibility. That lack of flexibility, however, is only a cost when benevolent policy makers govern. The cost of restraining the "devils" is to also restrain the "angels". What is important is that the institutions that are relevant to solve the credibility problem are those that limit the policy makers freedom to renege on their promises (and also reform the institutions) and re-optimize over time.
Those that believe that policy makers are always benevolent, a strand of the public finance literature that predominated until recently, will see only the costs and not the benefits. Hence all policies that restrain flexibility will appear sub-optimal to them. Not surprisingly, those that usually think like this are those that favor interventionist government policies, and those that favor institutional restrains are those that believe the government is part of the problem rather than part of the solution. Hence in many respects the discussion about the appropriateness of rules, the independence of the Central Bank, the choice of a truly conservative central banker even though the government may not be conservative (Rogoff 1985), etc., replays the discussion between those that favor state activist policies and those that oppose them.
There is, however, a reason to oppose too rigid rules. That is, if a government is factional and tries to favor certain interests, it may try to lock in its policies so that if the opposition wins it may not undo the policies that favor their group. In this case to permit too rigid rules is clearly sub-optimal.
POLITICS AS A PRINCIPAL AGENT PROBLEM
All this has to remind economists of the principal agent problem. The public choice literature (a recent summary is included in Mueller 1997) sees the relationship between society (the principal) and politicians (the agent) as one in which the key is the "social" contract signed, which is made explicit in the Constitution. To them, once the Constitution is written, the evolution of economic policy is fully determined. The only way to make normative economic policy is through changes of the Constitution. Another application of principal agent theory comes from the literature on transaction costs. An extremely interesting application of transaction cost theory to politics is included in Dixit (1998). In his book Dixit models the relationship between politicians and society as a principal agent problem. In the parallel he makes, it is clear that this contract is extremely more complex than that normally considered in a firm. It actually approximates only the most complex of contracts, the one that attempts to resolve the question of how to monitor the monitors; the problem of how to provide adequate incentives to a CEO.
Piece rates are impossible because of the lack of a clear relationship between output and input. There is lots of noise between an action and an effect in politics. The politician could apply the best policies and even then an exogenous shock could prevent the economy from growing. Hence the need for a contract with straight pay and supervision. But this still leaves incentive to follow policies that yield benefits only in the short run (again similar to the problem of CEO compensation). In the case of CEOs one solution that has been developed is to provide them with many puts at a high strike price. How could one apply this in the case of politicians? Maybe correlating income to the value of public debt? Pay according to relative performance (relative to other relevant countries), using tournament theory? It is easy to find parallels in the nature of the problem, but not in the solutions.
This is because in reality the similarities are not that many. What would happen in a firm if the CEO where elected by workers, hired at a fixed income, and with a pay that is similar to that of the best paid workers? In a firm these low powered incentive contracts require stringent supervision. How do we supervise? Is the incentive to throw them out of office enough? In many respects, this literature is an extension of economics into political science. The question to be answered here is not only what is the correct incentive system, but which are the optimal political institutions.
Grossman and Helpman (1994) develop a theory of "Special Interest Politics". Their AER article on "protection for sale" shows the optimal structure of trade policy when politicians maximize a function that includes social welfare but also contributions from lobbies. They find that this solution depends on factor endowment, product elasticities, and whether certain groups can or cannot solve the free rider problem and constitute an effective lobby. In many respects, this literature is associated to the "capture" theory of regulators and to the idea that policy is determined by deep rooted aspects of the economy, such as the nature of the goods produced in the economy, and the market structure in which they are produced. Here the problem is again one of principal agent, in which the principal is both society at large and all the lobby groups of the economy.
APPLIED POLITICAL ECONOMICS TO FISCAL POLICY AND POLITICAL INTITUTIONS
An interesting example of the results of the Political Economics literature is in the area of fiscal policy. The literature has found incentives to behave sub optimally in the issuance of public debt. Tabellini (1990), and Alesina and Tabellini (1990) show how a government may have incentives to issue too much public debt. An institutional remedy would be to impose a constitutional rule (for example a balanced budget rule). However, to be effective these institutions must be costly to reform. To impose a budgetary rule that is difficult to reform might sound as too risky for many and hence may be hard to do. Persson, Roland and Tabellini (2001) have delved into comparative politics to find alternative solutions. In their paper they present a model of electoral accountability to compare the public finance outcome under alternative political arrangements. They find that presidential and parliamentary systems have substantially different incentives and result in different public finance policies. They say: "A Parliamentary system has redistribution towards the majority, less underprovision of public goods, more waste and a higher burden of taxation, whereas a presidential-congressional system has redistribution towards a minority, more underprovision of public goods, but less waste and a smaller size of government." (page 2).
Aside from political regime, this literature also examines whether the electoral system may also be part of the solution. Empirical studies (such as Persson 2000) have given strong indications that the electoral system is a major determinant of public debt accumulation. Persson studies empirically the institutions governing electoral rules and political regimes and their effect on fiscal policy. He finds that "empirically, electoral rules and political regimes do seem to systematically influence the choice of fiscal instruments, as well as the incidence of corruption." (page 2).
The discussion in Persson is particularly enlightening of the Chilean experience. In Chile one lives the tension between political institutions and politicians. Politicians in Chile criticise constantly the limits that institutions pose on them. However, in this literature this may be a sign of adequate design. This in part is because the present coalition (the opposition to the Pinochet regime) finds this an unnecessary burden inherited from the dictatorship that produces "undemocratic" enclaves in the functioning of political institutions. It is interesting what "undemocratic" means in this context. It means that the majority cannot change all the rules at any point in time. But this is not necessarily optimal. Those that spurn the limits of the current institutions usually also prefer parliamentarian regimes and proportional representation. As Persson proves, those changes in the political regime and the electoral rules would result in more corruption (rents) and a bigger government. It could, alternatively, also be true that these politicians find that the previous government "locked in" policies that do not benefit society at large, but only certain groups that were preferred by the military government.
Let me describe further the results from Persson (2000). What he studies as "electoral rules" is both district size and electoral formula. The electoral formula determines how many legislators acquire a seat in a voting district. The electoral formula determines how votes are translated into seats. Plurality rule implies that only those with the highest vote shares get seats, while proportional representation implies seats are awarded in proportion to the vote share. The political regime types are: presidential and parliamentary.
I will not extend myself on the theoretical explanation of which rules or regime favors one or another public finance outcome (for that, see the paper). His predictions are that a smaller size of government and less rents occur when there is a combination of presidential regime and majoritarian electoral rules. This is, surprisingly enough, the system currently operating in Chile. As I said before, many politicians would like to change them. Maybe it would be inconvenient for them to be changed, in particular if one thinks that the best for society at large is to commit to a small government. This calls to our attention, as in the public choice literature, the importance of the Constitution and why the characteristics of political institutions should be in the Constitution and hence difficult to reform. In other words, political regimes and electoral rules may be too important to leave them to the politicians.
Interestingly enough, the classification of the countries in and by itself gives some interesting results. Persson presents a map with the different combinations of presidential or parliamentarian, and majoritarian or proportional representation. From the map it is clear that countries are basically classified into:
Europe is mainly parliamentarian and proportional representation,
Commonwealth countries are parliamentarian and majoritarian,
The US is presidential and majoritarian, and
Latin America is mainly presidential and proportional representation.
From the point of view of the theory, the US system is associated to less government, and less rents than all the others. This is also the less frequently seen system. Conversely, the system predominating in Europe would be associated wit the most government and rents. Curiously, the exception in Latin America to the classification described above is Chile, which has a system that is different from the rest of Latin America, in that it is majoritarian, and hence it is similar to that of the US.
Empirical results confirm that presidential countries have smaller governments by 5-9 percent of GDP and majoritarian countries have smaller governments by 2-4 percent of GDP (though this variable is sometimes not significant). Persson also finds that countries with different regimes systematically react differently to external shocks.
Persson says that the fact Chile is less corrupt in Latin America is due to electoral system (no proportional representation). However, the institutional reforms are recent and corruption in Chile possibly had been lower before that. Also, there is a contradiction with Haggard and Kaufman (1992) that argue that the system in Chile is actually prone to corruption since it has many veto gates and should produce weak links between candidates and parties. This points to inconsistencies in the theory that need to be sorted out. The issue is to determine which institutions are key, and which secondary.
Persson concludes that: "presidential regimes are associated with smaller governments than parliamentarian regimes, a smaller response of spending to different events, and a stronger post election cycle but a weaker pre election cycle. Majoritarian elections are associated with smaller broad spending programs than proportional elections and with less corruption, they also have smaller spending responses to events and a stronger pre-election cycle in taxes." (page 18).
All these new theories and research are fascinating. I hope you find the papers that follow fascinating as well. Linking up with my discussion at the beginning, let me finish by discussing briefly the papers in this issue.
If Lal is to be believed, the subject of study of the other strands of the literature concentrate themselves with issues that are of secondary importance as non economic determinants of policy. Political institutions themselves would be determined by deeper forces and so only the study of these forces would let us see what determines the long term trends in growth.
Lal believes in looking at deeply ingrained institutions and to culture (not necessarily political) and its consequences on long term growth. Lal's view is that what predominates are deeply rooted characteristics, such as the cosmological views of the society, the factor endowments of the society, and that these determine the long term trend in economic policy and growth. This long term trend can be punctuated by crisis, which may make it optimal to deviate from it, but these deviations should be expected to be temporary. Hence my reading of Lal is that he is deeply skeptical that the current period of "liberal" economic policies in Latin America will endure.
Another part of the literature centers on the conditions under which deep reforms to economic policy is possible. This difficulty in doing structural reforms should in some respect appear natural. Institutions are always designed to give inertia to current policy. This is the only way that the expectation of their permanence may have the desired effects. The question, then, is how to break this inertia. Colombia in that sense had been singled out as a county that was able to reform without any clear event triggering reform. Edwards and Steiner question this. They claim crisis do not have to be necessarily economic, they can be political, social or institutional. They argue cogently that Colombia is not an exception in the long literature that finds that a crisis triggers reform.
Tommasi et al. also fall into this category. They claim it is usually thought that the Menem regime was unconstrained and ruled by decree, and that made reform possible. They argue that an important "behind the scenes" logrolling activity was under way, in which votes were being traded, and that the coalitions formed in this trade conditioned the kind of reforms that were possible, their sequencing, and which were not feasible (as the labour market reforms). Hence Tommasi et al. argue the importance of coalition forming and the need to actively seek for the particular political alignment required.
Sapelli studies another question. This question in many respects is the logical follow up to that responded by the literature on when reforms are possible. If the inertia of the institutional framework has been broken, and deep reforms are possible, then how are these policies made credible. If at that juncture the institutions were amenable to reform, what is to prevent another counter reform in a different direction in a short period of time? Alternatively, after reform, what prevents policies to revert back to the old policies? The break with tradition should signal to investors that dramatic changes of rules are possible. The sudden break with the old rules should make the new rules not credible. Since institutions prevent the credibility problem by diminishing flexibility, after the dramatic show of flexibility required for structural adjustment, how do you lock in the reforms to prevent serious credibility problems? In Chile this was a key problem of the transition to democracy. One could expect that during the Pinochet regime the sustainability of policies was guaranteed by the regime itself. But the new democratic government was not bound by this. Sapelli argues that the political institutions worked to favor policy continuance, as they should. Also, that in a rare show of preoccupation for this matters, the design of new institutions took into account this problem. The Chilean case has several examples of institutions that contributed to the sustainability of reforms in the democratic transition. Institutions achieved this by doing what they are supposed to do: neutralise pressures, make certain reforms difficult or impossible, provide incentives to channel special interests to areas where they do not do much harm, etc.
The institutional complexities of reform, as the institutional complexities to solve the credibility problem, are areas that will continue to be of concern to economists for many years to come. Hopefully, in some time, economists will be able to understand why economic policy takes the form it does.
* Profesor, Instituto de Economía, Pontificia Universidad Católica de Chile. email: firstname.lastname@example.org.
Alesina, A. and G., Tabellini (1990), "A Political Theory of Fiscal Deficits and Government Debt" Review of Economic Studies, 57 (July) 403-14. [ Links ]
Barro, R. and D., Gordon (1983), A positive theory of monetary policy in a natural rate model, JPE 91: 589-610. [ Links ]
Barro, R. and D., Gordon (1983) Rules, discretion and reputation in a model of monetary policy, Journal of Monetary Economics, 101-121. [ Links ]
Dixit, A. (1998), The Making of Economic Policy: A Transaction Cost Politics Perspective MIT Press. [ Links ]
Drazen, A (2000), Political Economy in Macroeconomics, Princeton University Press. [ Links ]
Feeny, D., 1988, "The development of property rights in land: a comparative study", in R. H. Bates (ed.): "Toward a Political Economy of Development", University of California Press, Berkeley. [ Links ]
Grossman, G. M. and E. Helpman (1994), "Protection for Sale", American Economic Review, 84 (September), pp. 833-50. [ Links ]
Haggard, S. and R. Kaufman (eds.) (1992), The Politics of Economic Adjustment Princeton University Press. [ Links ]
Hecksher, E. (1955), Mercantilism, 2 vols., Allen and Unwin, London. [ Links ]
Kydland, F. and Prescott E.C. (1977), "Rules rather than discretion: the inconsistency of optima plans", Journal of Political Economy, 85, 473 - 492. [ Links ]
Lal, D. (1998), "Unintended Consequences: The Impact of Factor Endowments, Culture and Politics on Economic Performance in the Long run", The Ohlin Lectures, MIT Press, Cambridge Mass. [ Links ]
Lal, D. (1998a), "The Political Economy of Reform in Latin America", in R. Skidelsky (ed.): The Politics of Economic Reform, Profile Books, London. [ Links ]
Lal, D. and H. Myint (1996), "The Political Economy of Poverty, Equity and Growth", Clarendon Press, Oxford. [ Links ]
Lin, J. Y. and J. B., Nugent (1995), "Institutions and Economic Development", in J. Behrman and T. N. Srinivasan (eds.): Handbooks of Development Economics, Vol. 3A, North Holland, Amsterdam. [ Links ]
Morse, R. M. (1964), "The heritage of Latin America" in L. Hartz (ed.): The Founding of New Societies, Harcourt, Brace and World, New York. [ Links ]
Mueller, D. editor (1997), Perspectives on Public Choice: A Handbook Cambridge University Press. [ Links ]
North, D. C. (1981), Structure and Change in Economic History, New York. [ Links ]
North, D. C. (1990), "Institutions, Institutional Change and Economic Performance", Cambridge University Press, Cambridge. [ Links ]
North, D. C., and R. P., Thomas (1973), "The Rise of the Western World", Cambridge University Press, Cambridge. [ Links ]
Oakeshott, M. (1993), "Morality and Politics in Modern Europe", Yale University Press, New Haven. [ Links ]
Persson, T. and G., Tabellini (1995), "Double edged incentives: Institutions and policy coordination", in Grossman and Rogoff (eds.) Handbook of International Economics, Volume III, Amsterdam, North Holland. [ Links ]
Persson, T. and G., Tabellini (2000), Political Economics: Explaining Economic Policy, MIT Press. [ Links ]
Persson, T. and L., Svensson (1989), Why a stubborn conservative would run a deficit: policy with time inconsistency preferences. QJE 104: 325-345. [ Links ]
Persson, T.; G., Roland and G., Tabellini (1997), Separation of powers and political accountability, QJE 112: 1163-1202. [ Links ]
Persson T., G., Roland and G., Tabellini (1998), Towards micropolitical foundations of public finance, European Economic Review 42: 685-694. [ Links ]
Persson, T.; G., Roland and G., Tabellini (2001), Comparative Politics and Public Finance. Journal of Political Economy, forthcoming. [ Links ]
Persson, T. (1998), "Economic policy and special interest politics," Economic Journal, 108, 310-327. [ Links ]
Persson, T. and G., Tabellini (2000), "Political economics and macroeconomic policy", in Taylor, J. and M. Woodford (eds.), Handbook of Macroeconomics, North-Holland: Amsterdam. [ Links ]
Persson, T. and G., Tabellini (1999), "The size and scope of government: Comparative politics with rational politicians", 1998 Marshall Lecture, in European Economic Review 43, 699-735. [ Links ]
Persson, T. (2000), Do political institutions shape economic policy? Stockholm University. [ Links ]
Persson, T., G. Roland and G. Tabellini (1997), "The theory of fiscal federalism: What does it mean for Europe?," in Siebert, H. (ed.), Quo Vadis Europe?, J.C.B. Mohr: Tübingen [ Links ]
Persson, T.; G., Roland and G., Tabellini (2000), Comparative Politics and Public Finance, JPE, forthcoming. [ Links ]
Rogoff, K. (1985), "The optimal degree of commitment to an intermediate monetary target", Quarterly Journal of Economics, 100, 1169 - 1689. [ Links ]
Tabellini, G. (1990), "A Positive Theory of Social Security" NBER WP 3272. [ Links ]
Weber, M. (1958), The Protestant Ethic and the Spirit of Capitalism. New York: Charles Scribner's Sons. [ Links ]
Williamson, O. E. (1985), The Economic Institutions of Capitalism, Free Press, New York. [ Links ]