Traditional analyses of economic policy whether in macroeconomics or in public finance examine the private sector's responses to once-and-for-all choices of different policy rules that determine the time-path for policy instruments. The normative policy problem amounts to identifying which particular policy rule gives the best social outcome, typically from the viewpoint of a hypothetical enlightened `social planner'. This approach does not square with the way most economic policy decisions are reached in reality, however, and it is potentially misleading when constructing either a positive or a normative theory of policy.

Credibility and Politics
Policy-makers in the real world typically do not set the whole time-path for their policy instruments once and for all. Instead, they choose policy sequentially over time as part of an ongoing process. Policies with desirable long-run properties may therefore suffer from serious credibility problems, because forward-looking private agents realize that policy-makers may want to deviate from these policies in the future. Once private agents take policy-makers' future incentives into account in forming their own expectations and making economic decisions, the resulting outcome may be much worse than that of policies that are desirable but not credible.
Further, policy-makers in the real world are typically not enlightened social planners, but rather politically appointed agents, whose choices of economic policy are coloured by their political ambitions and responsibilities. Successful analyses of policy may therefore have to consider politics: political incentives may affect economic policy choices not only because policy-makers strive to be re-elected, but also because incumbent policy-makers may choose policies in order to `create facts' that will affect the choices of future policy-makers with different political preferences. Moreover, different political institutions may aggregate conflicting interests in society in different ways and thus shape different policy outcomes.
In the last few years researchers have developed a new approach to the analysis of economic policy that takes these factors into account. This new approach treats policy as endogenous in the analysis, specifying policy-makers' objectives and the constraints facing them and often borrowing methods from game theory to address questions of strategic policy choice. By focusing on policy-makers' incentives, researchers have been able to consider which policies are credible or politically feasible. Such questions lie at the heart of understanding actual policy problems, but they were overlooked by the traditional approach to economic policy formulation. We believe that this new approach is summarized and evaluated for the first time in our new book,¹ in which we use simple models of monetary and fiscal policy to illustrate how this new approach can tackle both positive and normative policy issues (Persson and Tabellini (1990a)).
Much of the research in this new field has been undertaken by CEPR Research Fellows and presented at the Centre's conferences and workshops. In this article we summarize some recent contributions to this research programme, with particular emphasis on contributions by CEPR Research Fellows to stabilization and tax policies. Space constraints unfortunately prevent us from describing several related areas of research, including trade and industrial policy, LDC debt and international policy coordination, where much of the work on credibility and strategic policy choice originated (Buiter and Marston (1985)).

Credibility and Macroeconomic Policy
Work in the early 1980s demonstrated how credibility problems may lead to bad policy outcomes, such as an over-expansionary monetary policy and high inflation, or excessive taxation of capital and low savings. Barro and Gordon (1983), Backus and Driffill (1985) and Vickers (1986) analysed whether a policy-maker's incentive to acquire and preserve a `reputation' might help to maintain credibility in monetary policy and thereby keep inflation down. Cohen and Michel (1988) demonstrated how to calculate equilibrium policies in dynamic models where the policy-maker is subject to a credibility problem. Kotlikoff, Persson and Svensson (1988) questioned whether policy-makers' incentives not to break an existing `social contract' might enhance the credibility of capital taxation and thereby maintain a high savings ratio.
At the core of the credibility problem are the political pressures that may bring policy-makers to reverse a previously announced programme. Any institution that makes policy reversals more difficult enables policy-makers to make stronger commitments to future policies and thereby enhances their credibility. In a seminal paper, Rogoff (1985) noted that a simple way to commit is to delegate. Rogoff argued that society could credibly commit to a low inflation rate by delegating monetary policy to an independent central bank and instructing it to keep inflation low. Lohman (1991) recently extended Rogoff's work to examine the details of institutional design, paying particular attention to the trade-off between credibility and flexibility when society delegates policy to an independent central bank.
Recent empirical research has strengthened Rogoff's argument. Alesina (1989) and Grilli, Masciandaro and Tabellini (1991) have shown that countries with more independent central banks do tend to have lower average inflation rates. The latter authors, as well as Alesina and Summers (1990), also show that this lower average inflation is not associated with lower growth, higher unemployment or greater variability of either.
The operations of most existing central banks are characterized by considerable secrecy. Cukierman and Meltzer (1986) demonstrate that it may be in the central bank's interest not to reveal all its information to the private sector, since preserving secrecy allows the bank to intervene in a more powerful way, when the benefits of intervention are particularly high. Basar and Salmon (1989) and Levine and Pearlman (1990) have recently extended this analysis to cover cases of wage and price stickiness.

Exchange Rate Regimes
Another institutional device to gain credibility is to make an international agreement. The international sanctions that may follow if the agreement is broken unilaterally can make reversals of announced policies sufficiently costly to domestic policy-makers to create some perceived commitment. In a sense, joining an international agreement is like delegating some policy decisions to an international agent. Giavazzi and Pagano (1988), in a now classic paper, were the first to point out that the EMS could be viewed in precisely this way.
Subsequent research has attempted to quantify the contribution of the EMS to the credibility of monetary policy in the high-inflation countries, and the evidence it provides is mixed. Giavazzi and Giovannini (1989) used data up to 1986 and found that some of the high-inflation countries had managed to `import' credibility from the Bundesbank, but only several years after the inception of the EMS. Alogoskoufis (1990 and 1991) reports similar findings for a larger group of countries and with more recent data. Weber (1988) attempts to measure the credibility of policy-makers in the various EMS member countries directly; and in a recent extension of this work Weber (1991) suggests patterns that differ significantly across countries and over time.
In its present form the EMS can be described as an institution that permits individual countries to commit to a rule with an `escape clause': in normal times the exchange rate remains fixed, but in abnormal times there is a realignment. Obstfeld (1991) shows that such an institution can lead to far worse equilibria than an irrevocably fixed rate: limited credibility may cause periods without realignments, but with excessively high real wages, real interest rates and unemployment.

Government Capital Structure
In traditional analyses of economic policy, the government's financial structure is irrelevant, but when incentive problems are introduced the financial structure will affect the policy outcome. In a sense, this finding parallels a result from the recent theory of corporate finance: that the Modigliani-Miller theorem, which spells out the irrelevance of the corporate financial structure, breaks down if there are incentive problems between managers and shareholders.
Following Lucas and Stokey (1983), Persson, Persson and Svensson (1987) showed that giving public debt the right maturity structure and composition may raise the credibility of fiscal policy. In particular, balancing the government's nominal position against the private sector may help to keep inflation down by eliminating the incentive for policy-makers to impose a `surprise inflation' tax to dilute the value of the outstanding money stock.
In later work, Calvo and Guidotti (1991) discuss the proper balance between indexed and nominal public debt, given that indexed debt may increase governments' credibility in keeping inflation down while nominal debt may increase its flexibility in financing unforeseen contingencies through an inflation tax. Further, Giavazzi and Pagano (1991) and Alesina, Prati and Tabellini (1991) have shown that shifting a substantial public debt into longer maturities may help to reduce the risk premium incorporated in interest rates by increasing the public's confidence that there is not going to be a debt crisis.
In other recent work, Tabellini (1990 and 1991) has demonstrated that if members of older generations hold claims on the government as prospective pensions rather than as public debt, this may result in different policies. He has also shown, as have Aghion and Bolton (1991), that the solution to problems of default on the public debt may lie in ensuring that the debt is held by a sufficiently wide section of the population to create a `constituency for repayment'.

Empirical Evidence on Political Business Cycles
The modern analysis of how political incentives shape macroeconomic policy is fairly recent, but this field of research is now growing fast. Several important papers in this area were presented at the June 1990 conference on `Economic Policy in Political Equilibrium', organized by the present authors on behalf of CEPR and the Institute for International Economic Studies, which is reported in detail in issue no. 40 of the Bulletin, August 1990. A selection of these papers will appear in a forthcoming special issue of the Review of Economic Studies.²
The modern literature has shown that elections may give rise to policy cycles for at least two reasons. Alesina (1987 and 1988) has emphasized the cycles that may arise after elections, as newly appointed policy-makers enact policies that differ from those of their predecessors because of ideological differences. Rogoff and Sibert (1988) and Rogoff (1990a) have emphasized the cycles arising before elections, as incumbent policy-makers try to signal their competence to voters to increase their chances of re-election. Alesina and Roubini (1990) investigate the evidence of such cycles in output, unemployment and inflation in 18 OECD economies during the post-war period. They find strong evidence in support of the first hypothesis and mild evidence in support of the second.
Elections or more generally changes of government may also distort policy-makers' intertemporal incentives. Persson and Svensson (1989) and Tabellini and Alesina (1990) have shown how incumbent policy-makers who are unsure of their reappointment may want to use government debt strategically in order to influence future economic policy. In particular, more unstable governments tend to deviate further from the balanced budget. This prediction is consistent with a large and disparate body of empirical evidence for both industrial countries (see Grilli, Masciandaro and Tabellini (1991)) and developing countries (see Cukierman, Edwards and Tabellini (1990), Ozler and Tabellini (1990) and Roubini (1990)).

Politics of Stabilization Policies
Many observers have noted the recurrent puzzle that developing countries often refrain from enacting seemingly obvious welfare-improving reforms of trade, fiscal and monetary policy. Recent contributions have attempted to explain this by studying the policy-makers' political incentives. Alesina and Drazen (1990) and Drazen and Grilli (1990) suggest that actual stabilization policies can be viewed as the equilibrium outcome of a war of attrition between different groups in society, which tends to delay their implementation. Dornbusch (1990a, 1990b and 1990c) argues that credibility problems may compromise such stabilization programmes, with examples from the stabilizations carried out in Europe in the 1940s as well as in recent years by Latin American countries. Beenstock (1989) analyses political equilibria in which each group's lobbying activities have externalities which induce all groups to lobby more intensively. Fernandez and Rodrik (1990) show that uncertainty about the distribution of the gains from a welfare-improving reform may prevent it from collecting a political majority, so the political system favours the status quo. Persson and Tabellini (1990b) find similarly that large changes in economic conditions may not change policies by much: even if policy-makers want to alter policy, the political system can adapt by appointing new policy-makers, which tends to preserve the status quo.

Efficient Political Institutions
Much of the work in this area shares a conclusion with the related literature on public choice: political incentives and political institutions tend to distort economic efficiency. But this result is not universal: indeed Persson and Tabellini (1990c) have shown how existing political institutions enable society to delegate policy to representatives with policy preferences that will help to overcome credibility problems. This delegation may even arise endogenously through majority voting. Cohen and Michel (1991) characterize a set of credible policy commitments by political candidates under a set of specific assumptions concerning voting behaviour. Rogoff (1990b) discusses the normative question of how to set the electoral rules to promote economic efficiency, emphasizing one possible trade-off to be faced in choosing the statutory period of time between elections: having elections further apart may eliminate costly political cycles, but it may also limit society's ability to get rid of incumbent policy-makers who prove to be incompetent.

Politics and Growth
In the traditional literature on economic development, social institutions and political factors play a major role. Researchers have recently sought to merge the insights from new research on `endogenous growth' with those from the new policy literature, in order to address these issues in a more precise way. The central question is why some countries but not others have enacted growth-promoting policies. Income distribution may play a major role. Romer (1990) studied the distributive effects of growth-promoting trade policies and argued that such policies may be blocked by the owners of the production factors that will lose out. Alesina and Rodrik (1990) and Bertola (1991) focus on the functional distribution of income between labour and capital and find that income inequality may give rise to policies that will hamper growth. Persson and Tabellini (1990d) and Perotti (1990) have shown on theoretical grounds that inequality in the size distribution of income has an adverse effect on economic growth. Persson and Tabellini have also found empirical support for this hypothesis, both in historical and in current data.

Conclusions
This research programme is still in its infancy, and much remains to be done. At an abstract level, we need to know more about how collective decisions are taken and how different institutions shape the policy formation process. There is a wealth of empirical observations that remain unexploited concerning the the wide differences in economic policy and performance among countries governed by different political systems. At a more applied level, this new approach may be fruitfully extended to study new policy issues, in particular the challenges posed by the transformation of socialist systems into free market democracies and the tightening of European integration, where the interplay among economic policy, institutional reform and politics will play a central role.

¹Torsten Persson and Guido Tabellini, Macroeconomic Policy, Credibility and Politics,
(ISBN: 3-7186-5029-0) from the current Fundamentals in Pure and Applied Economics Series (ISSN: 0191-1708), $42.00, available from Harwood Academic Publishers, STBS, P.O. Box 90, Reading RG1 8JL, UK, or STBS, P.O. Box 786, Cooper Station, New York, NY 10276, USA.
² Review of Economic Studies, Special Issue, Spring 1992,
available from Basil Blackwell Ltd., 108 Cowley Road, Oxford OX4 1JF, UK or Basil Blackwell, Inc., 3 Cambridge Center, Suite 302, Cambridge MA 02142, USA.

Aghion, P and P Bolton (1991), `Government Domestic Debt and the Risk of Default: a Political-Economic Model of the Strategic Role of Debt', in Dornbusch and Draghi (1991).
Alesina, A (1987), `Macroeconomic Policy in a Two-Party System as a Repeated Game', Quarterly Journal of Economics.
Alesina, A (1988), `Credibility and Policy Convergence in a Two-Party System with Rational Voters', American Economic Review.
Alesina, A (1989), `Inflation, Unemployment and Politics in Industrial Democracies', Economic Policy, 8.
Alesina, A and A Drazen (1990), `Why are Stabilizations Delayed?', mimeo.
Alesina, A, A Prati and G Tabellini (1991), `Public Confidence and Debt Management: A Model and Case Study of Italy', in Dornbusch and Draghi (1991).
Alesina, A and D Rodrik (1990), `Redistributive Policies and Economic Growth', mimeo.
Alesina, A and N Roubini (1990), `Political Cycles in the OECD Economies', CEPR Discussion Paper No. 470.
Alesina, A and L Summers (1990), `Central Bank Independence and Macroeconomic Performance', mimeo.
Alogoskoufis, G (1990), `Real Effects of Nominal Exchange Rate Regimes', mimeo.
Alogoskoufis, G (1991), `Monetary Accommodation, Exchange Rate Regimes and Inflation Persistence', CEPR Discussion Paper No. 503.
Backus, D and J Driffill (1985), `Rational Expectations and Policy Credibility Following a Change in Regime', Review of Economics and Statistics.
Barro, R and D Gordon (1983), `Rules, Discretion and Reputation in a Model of Monetary Policy', Journal of Monetary Economics.
Basar, T and M Salmon (1989), `Credibility and the Value of Information Transmission in a Model of Monetary Policy and Inflation', CEPR Discussion Paper No. 338.
Beenstock, M (1989), `A Democratic Model of the Rent Sought Benefit Cycle', CEPR Discussion Paper No. 289.
Bertola, G (1991), `Factor Shares, Savings Propensities and Endogenous Growth', mimeo, Princeton.
Buiter, W and R Marston (eds.) (1985), International Economic Policy Coordination, Cambridge University Press for CEPR and NBER, Cambridge.
Calvo, G and P Guidotti (1991), `Indexation and Maturity of Government Bonds: An Exploratory Model', in Dornbusch and Draghi (1991).
Cohen, D and P Michel (1988), `How Should Control Theory be Used to Calculate a Time-Consistent Policy?', Review of Economic Studies.
Cohen, D and P Michel (1991), `Which Rules Rather than Discretion in a Democracy? An Axiomatic Approach', CEPR Discussion Paper No. 539.
Cukierman, A, S Edwards and G Tabellini (1990), `Seigniorage and Political Instability', CEPR Discussion Paper No. 381.
Cukierman, A and A Meltzer (1986), `A Theory of Ambiguity, Credibility and Inflation Under Discretion and Asymmetric Information', Econometrica.
Dornbusch, R (1990a), `Credibility and Stabilization', CEPR Discussion Paper No. 454.
Dornbusch, R (1990b), `Experiences with Extreme Monetary Instability', CEPR Discussion Paper No. 455.
Dornbusch, R (1990c), `Policies to Move from Stabilization to Growth', CEPR Discussion Paper No. 456.
Dornbusch, R and M Draghi (eds.) (1991), Public Debt Management: Theory and History, Cambridge University Press for CEPR, Cambridge.
Drazen, A and V Grilli (1990), `The Benefit of Crises for Economic Reforms', mimeo.
Fernandez, R and D Rodrik (1990), `Why is Trade Reform so Unpopular?', CEPR Discussion Paper No. 391.
Giavazzi, F and A Giovannini (1989), Limiting Exchange Rate Flexibility The European Monetary System, MIT Press, Cambridge MA.
Giavazzi, F and M Pagano (1988), `The Advantage of Tying One's Hand: EMS Discipline and Central Bank Credibility', European Economic Review.
Giavazzi, F and M Pagano (1991), `Confidence Crises and Public Debt Management', in Dornbusch and Draghi (1991).
Grilli, V, D Masciandaro and G Tabellini (1991), `Political and Monetary Institutions and Public Financial Policies in the Industrial Policies', Economic Policy, 13.
Kotlikoff, L, T Persson and L Svensson (1988), `Social Contracts as Assets: A Possible Solution to the Time-Consistency Problem', American Economic Review.
Levine, P and J Pearlman (1990), `Credibility, Ambiguity and Asymmetric Information with Wage/Price Stickiness', CEPR Discussion Paper No. 409.
Lohman, S (1991 forthcoming), `Optimal Commitment in Monetary Policy: Credibility versus Flexibility', American Economic Review.
Lucas, R and N Stokey (1983), `Optimal Fiscal and Monetary Policy in an Economy Without Capital', Journal of Monetary Economics.
Obstfeld, M (1991), `Destabilizing Effects of Exchange Rate Escape Clauses', CEPR Discussion Paper No. 518.
Ozler, S and G Tabellini (1990), `External Debt and Political Instability', mimeo, Harvard.
Perotti, R (1990), `Political Equilibrium, Income Distribution and Growth', mimeo.
Persson, M, T Persson and L Svensson (1987), `Time Consistency of Fiscal and Monetary Policy', Econometrica.
Persson, T and L Svensson (1989), `Why a Stubborn Conservative Would Run a Deficit: Policy with Time-Inconsistent Preferences', Quarterly Journal of Economics.
Persson, T and G Tabellini (1990a), Macroeconomic Policy, Credibility and Politics, Harwood Academic Publishers, London.
Persson, T and G Tabellini (1990b), `The Politics of 1992: Fiscal Policy and European Integration', CEPR Discussion Paper No. 501.
Persson, T and G Tabellini (1990c), `Representative Democracy and Capital Taxation', mimeo.
Persson, T and G Tabellini (1990d), `Is Inequality Harmful for Growth? Theory and Evidence', mimeo.
Rogoff, K (1985), `The Optimal Degree of Commitment to an Intermediate Monetary Target', Quarterly Journal of Economics.
Rogoff, K (1990a), `Equilibrium Political Budget Cycles', American Economic Review.
Rogoff, K (1990b), `How Often Should Elections Be Held?', mimeo.
Rogoff, K and A Sibert (1988), `Elections and Macroeconomic Policy Cycles', Review of Economic Studies.
Romer, P (1990), `Trade, Politics and Growth in a Small Open Economy', mimeo.
Roubini, N (1990), `Political Determinants of Inflation and Fiscal Deficits in Developing Countries', mimeo.
Tabellini, G (1990), `A Positive Theory of Social Security', CEPR Discussion Paper No. 394.
Tabellini, G (1991), `Intergenerational Redistribution, Altruism and Politics', Journal of Political Economy.
Tabellini, G and A Alesina (1990), `Voting on the Budget Deficit', American Economic Review. Vickers, J (1986), `Signalling in a Model of Monetary Policy with Incomplete Information', Oxford Economic Papers.
Weber, A (1988), `The Credibility of Monetary Policies, Policymakers' Reputation and the EMS-hypothesis: Empirical Evidence from 13 Countries', CentER Discussion Paper No. 8803.
Weber, A (1991), `Reputation and Credibility in the European Monetary System', Economic Policy, 12.