Resumé

SUMMARY: International policy coordination is justified by two different types of arguments. Most of the economics literature stresses the potential welfare gains that arise as countries move from a non-cooperative to a cooperative equilibrium. Empirical suggests, however, that these gaines are likely to be modest and uncertain. The emphasis then shifts to the second justification: to preserve the international public of a stable international regime. At the same time a move from discretionary to rule-based coordination can be observed, particularly in EC economic integration. The latter, particularly the move towards economic and monetary union, has important implications for future efforts at global coordination.

Introduction

Coordination of international economic policies is a strong form of international cooperation. It implies a modification of national policies in recognition of international interdependence, usually supported by an explicit international agreement constrains, or even determines, some instruments of policy. Coordination therefore goes beyond the systematic exchange of information between countries, often with international organizations as intermediares, about policy actions and objectives. Exchange of information is a prerequisite for coordination and may in itself lead to significant in national policy formulation and execution by assuring consistency minimizing conflicts over shared targets.

The following two sections will survey two potential types of benefits from and two main approaches to coordination. Examples from macroeconomic policy coordination, illustrating the major practical difficulties of achieving it, are also given. Nations try to agree on policies in other more detailed policy areas as well, including minimal harmonizationof legislative and regulatory framework within which economic

---

This article draws on joint work over the past five years with Daniel Gros on European monetary integration on the author's contributions to the New Palgrave Dictionary of Money and Finance to be published Macmillan in 1992.

agents operate; establishing an agreed framework for trade and competition policies and for financial regulation and supervision are prime and topical examples. But macroeconomic policy coordination provides sufficient examples of the issues which arise. Particular attention will be given to examples which illustrate different degrees of ambition in international coordination.

Potential benefits from policy coordination

Two approaches may be distinguished in the analysis of the purposes of coordinating
policies, see Kenen (1988): the optimizing approach and the regime-preserving

The former starts from the observation that in the global economy, as well as within regions of nations, the integration of goods and financial markets has proceeded so far that national policies, at least in the main countries, have significant externalities or spill-over effects for other nations. If governments act independently they may underestimate effects and the feed-back to their own economy which they entail. Partial neglect of interdependence will lead to a sub-optimal constellation of policies - a noncooperative Nash equilibrium - which governments could improve upon by moving closer to a joint setting of their policy instruments. Through an international agreement each government could improve its capacity to get closer to its own preferred objective.

Early examples of this type of justification for policy coordination may be found in the Bretton Woods system prior to 1973 or in the efforts at devising coordinated responses the two oil price shocks in 1973 and 1979-80 among the industrialized countries. It was often observed by the international orginazations during the regime of fixed exchange rates and only moderate capital mobility of the 1960s that partial neglect other countries' policies or responses to changes in the external environment had generated collectively undesirable outcomes: the discipline of keeping the current account away from 'fundamental disequilibrium' became too tight, if several major countries were contracting demand at the same time. A rigourous analysis of welfarereducing leading into a low-level equilibrium trap for output and employment was first presented in Hamada (1976). On the other hand, policies would become too lax if several countries were expanding at the same time.

The Bonn Economic Summit of 1978 provides, in the changed international regime of the late 19705, the main practical illustration of the optimizing approach; the group of Seven participants agreed to make mutually consistent adjustments, mainly in their budgetary policies, in order to achieve a more sustainable constellation of current account

In the course of the 1980s the analytical underpinnings of the optimizing approach
were clarified, and elaborate efforts at empirical estimation of the potential benefits

from coordination were undertaken. The net effect has been, however, to weaken rather
than to strenghten the case for policy coordination among the main industrial economies.

A pionering theoretical and empirical contribution was made by Oudiz and Sachs (1984), starting from the constructive and bold assumption that the preferences of the three main industrial countries could be derived from their observed use of monetary and budgetary instruments. Using two prominent models of the economies of the United Japan and Germany and of their interaction the authors calculated the potential gains over the three-year period 1984-86 from submitting their policies to coordination. The gains were found to be modest, but still significant.

These cautions and carefully qualified conclusions raised some critical questions. The most obvious one was that the policy package proposed to bring about a change from non-cooperative to cooperative equilibrium consisted in some further US budgetary - at a time when the US deficit was already very large - and a worldwide lowering of interest rates. This clashed with dominant view of policy-makers, viz. that the US position was rapidly becoming unsustainable while inflation remained an important in Europe and Japan. Second, even the best macroeconomic models could hardly be considered sufficiently reliable for elaborate coordination efforts to rely heavily them.

The second point has since been studied intensively. In particular, Frankel and Rockett demonstrated that uncertainty about which is the proper macroeconomic model and disagreement among policy-makers from different nations on the interpretation their respective structure make it much harder to decide whether there are potential gains from coordination. Using ten prominent macromodels for the United and the rest of the world the authers concluded that such gains were plausible only in about half of all the possible combinations of models. These results may even understate the difficulties facing international negotiators; their views often appear more divergent than the range of model responses could justify, for example because they attach importance to long-run issues in budgetary policies which are not captured in any empirical model framework. Officials would then not even try to persuade their domestic political enviromnent of the virtues of coordination. At a simpler level, the difficulties of evaluating reliably what the current position of an economy is, and how it will be affected by shocks over the policy horizon relevant for the coordination initiatives suggest some humility in this approach.

Finally, the fact that most industrial countries are to-day less ready to set quantitative objectives for their macroeconomic policies than they were in the 1960s and 1970s has reduced the demand for the type of policy coordination that was then popular in the international see Polak (1988), and has favoured a more modest and realistic

The regime-preserving justification for coordination advocated e.g. by Olson (1965), Cooper (1985) and Kindleberger (1986), sees the main purpose as assuring the supply of important international public goods which would be undersupplied in a decentralized a liberal trade and financial regime, stable exchange rates and protection major economic and political shocks. To justify efforts at coordination with this purpose requires no detailed calculations of potential welfare gains, only a shared preception that if international action is not taken, important long-term benefits of an open international system could be lost. Illustrative examples seem more helpful than a general characterization.

Opinion in the US Congress swung in a protectionist direction in 1985 as a result of the widening US trade deficit, itself in part a result of prolonged appreciation of the dollar. To counter new trade legislation the US Administration took action, with its trading partners, to reduce overvaluation. This change in perception, assessed in Destler Henning (1989), led to the Plaza Agreement of September 1985. When the dollar fallen substantially in early 1987, concern shifted to the prospect that a further loss of confidence in the US currency might imply subsequent undervaluation during a 'hard landing' for the dollar. The Louvre Accord of February 1987 aimed, with some success, at stabilization.

A third possible illustration may be provided by the European Community (EC). As the EC moves towards monetary unification, seen by most Member States as a desireable to the creation of a well-functioning internal market in addition to being desirable on cost-benefit grounds, a minimum of constraints on national budgetary is seen as necessary to preserve such a desirable regime. Agreement was accordingly reached to closely monitor national budget deficits and debt levels in relation reference values and to authorize a collective EC body to impose (mild) sanctions offenders.

More generally, the serious budgetary deficits that have confronted national policymakers the industrialized countries over the past decade have undermined the view that budgetary policies can be regarded solely as a result of economic optimization by the governments concerned. Economists and political scientists have increasingly come to see deficits as explainable in terms of inherent difficulties and biases in the management expenditures and taxes in a democracy, see e.g. Alesina and Tabellini (1987) and Roubini and Sachs (1989) and the survey of the literature in Teisen (1991).

With this perspective, it is logical to ask whether commitments to an internationally stable regime can be used to constrain national budget deficits. The answer is yes - if the perceived benefits of preserving the regime are sufficiently clearly perceived in countries asked to accept the self-denying ordinance of limiting their budgetary freedom.This may be fulfilled in an advanced regional regime where much is at stake for a country by not cooperating. It is not currently met in the looser global systemwhere

stemwherethe benefits from preserving it are less obvious - and the loss of sovereigntyis
acceptable.

Approaches to coordination: rules or discretion?

How should coordination be implemented? Here again two main approaches may be
distinguished with general headings familiar from the principles of domestic stabilization
rules and discretion.

The former approach requires countries to agree on a set of rules within which they can then retain some autonomy in their economic management. A prime example is an exchange-rate system such as the European Monetary System (EMS) in which countries to manage the value of their currency inside fluctuation margins vis-avis or more reference currencies and only to change the margins rarely and in agreement with their partners. The Bretton Woods system shared the first characteristic; was supposed to have the second as well, but gradually lost it.

The second approach is ad hoc and procedural. Countries undertake to consult on policy actions shortly before they become necessary, but there is no a priori presumption particular instruments are singled out for use - or, on the contrary, are excluded the adjustment efforts.

The distinction is less clear in practice. A rule-based system may push participants into closer ad hoc management inside the rules and towards more joint exercise of discretion, see Padoa-Schioppa (1984). This is evident in the EMS, where tighter exchange-rate commitments has led to close monetary coordination and, more indirectly, monitoring of national budgetary policies, though without rules in the latter two areas. It is unlikely, on the other hand, that pressures will develop in an international system with discretionary coordination to prompt its own reinforcement through jointly agreed rules.

Apart from the differences in institutional dynamics which the two approaches imply, there is another important difference between them. The existence of wellunderstood e.g. an exchange-rate peg which can only be modified with great difficulty, into the expectations of economic agents and conditions their behaviour in a way that could not be approached by even successful and repeated discretionary actions by a number of countries. Paradoxically at first sight, foreign exchange interventions less frequent - and more efficient when they occur - in a system of fixed rates where the participants have explicit commitments to intervene than among currencies float against each other. Rules tend to increase the efficiency of policy instruments in national hands. They may also make the evaluation of the international effects of policy changes easier; some of the model uncertainty referred above arises from the highly uncertain impact of national policy actions on the exchange rate.

The benefits from relying on a rule focusing on a fairly rigid exchange rate have to be assessed against the costs of eliminating, though insistence on the rule, a superior discretionary policy response. It is usually theoretically possible to devise a discretionary response which is superior to an unconditional rule. In practice, however, the difficulties of (1) evaluating the nature of the international disturbance that has to be corrected, (2) assessing a proper international package of discretionary measuures, and (3) securing political agreement in the countries involved, will often prevent an optimal response. This shifts the balance of the arguments towards a preference for rule-based international coordination.

How are the two approaches to implementing coordination related to the justification the latter? Rule-based coordination is by definition at best optimizing in an average sense, but it would appear to be well-suited for regime preservation. Indeed, if what is to be preserved is conceived as the exchange rate regime itself, the relationship between the two dimensions in coordination becomes tautological. That would, however, too narrow; the stable exchange-rate regimes that have been observed in periods, and regionally, were not ends in themselves, but means to preserving the broader end of a liberal and stable international system.

With this in mind the table summarizes the illustrations of coordination that have
been made above:

Most of the actual experience with coordination may be rationalized as being of the regime-preserving variety. Both rule-based and discretionary approaches have been applied; the former has been unfeasible at the global level for the past two decades. The optimizing approach, the subject of most of the academic literature on coordination, only rarely been approximated at the global level.

A rule-based approach to optimization may appear to be a contradiction in terms.
Yet one may regard a very advanced stage of integration in which some externalities
are fully internalized, notably because of monetary unification, while discretion over

other important policy instruments is exercised in an increasingly joint fashion, as such a - so far unobserved - combination. In a full economic union the policy debate will shift from coordination between the participants towards formulation of one policy, regarded as optimal for the union as a whole. Retaining the terminology of the coordinationliterature, e.g Currie, Holtham and Hughes-Hallett (1989), there is then a shift from relative to absolute coordination targeted at the aggregate outcome.

The implications of EMU for global coordination

This final section looks at the probable implications for global policy coordination of the prospect, now a virtual certainty since the amendments to the Rome Treaty agreed Maastricht in December 1991, that most EC Member States will enter an economic monetary union before the end of the decade. How is this development likely influence the purpose and practice of global coordination?

Two considerations suggest that global coordination should be facilitated, see e.g. Dobson (1991). First, European unification simplifies the negotiation process in the Group of Seven through the reduction in the number of European participants from four to one, while making the emerging tripolar system more transparent. Second, if the EC is even moderately successful in its integration, the global system may be encouraged imitate some of the more advanced forms of coordination tried out regionally; relative success of the EMS has had some impact on global attitudes to exchange-rate

These arguments are less than fully convincing. Whether the United States and Japan find it easier or more difficult to reach an understanding with the EC on ad hoc or regime-preserving initiatives of coordination depends crucially on the institutional in economic and monetary union. The inspiring force of the integration in the latter might lead to defensive regional moves in the other main areas of the world economy, as may already be happening in both North America and Japan, rather than to more global coordination. There is further a risk, stressed by Bergsten (1990), that a tripolar system could prove more unstable than the present system as any one of the three main actors fears that the two others will try to 'gang up' to force its hancs in any particular conflict. A more realistic type of risk may be - in an environment which US leverage over the other two has diminished for both economic and political reasons - that each of the three main actors, increasingly confident of being fully in charge of its own economy (or its regional system) disregards signs of tension with the other two sufficiently long for then to become serious, see Thygesen (1991a). External concerns will become more remote and responsibility for then too diffused for effective and early coordination initiatives to be taken.

Monetary and exchange-rate policy supplies then an example of why global coordinationmay
increasingly difficult in the course of the 19905. It may be recalled

that the main efforts - of the regime-preserving variety, as argued above - were in this area in the 1980s, as evidenced by the Plaza Agreement and the Louvre Accord; indeed it is difficult to imagine any major coordination initiatives which would not involve some joint view of the main exchange-rate relationships.

In the economic and monetary union envisaged a few years from now a European Central Bank (ECB), committed to price stability as a primary objective and remarkably of political authorities, is likely to have a dominant position in macroeconomic at the EC level, including exchange-rate policy, see Thygesen (1991b). The Council of Finance Ministers may, if it decides unanimously to conclude a formal agreement with non-EC countries on an exchange rate for the ecu, override the ECB, though it will also in such a case be obliged to consult the ECB 'in an endevaour reach a consensus consistent with the objective of price stability' (Art. 109 of the proposed Treaty). These are stringent conditions; they appear to exclude the less formal understandings embodied in the 1985 and 1987 negotiations in the Group of Seven to influence the main exchange rates. If the Council - or the foreign parties in similar - is not prepared to go as far as to conclude a formal agreement, the Council can only, by qualified majority in that case, formulate 'general guidelines' for the union's exchange-rate policy. With these institutional arrangements it is likely to be difficult to reach even the limited results of international policy coordination obtained in the 1980s.

In view of the EC's preoccupation with the achievement of internal convergence in the years leading to economic and monetary union and thereafter and to give maximum attention to the objective of price stability, the time is hardly ripe for major efforts to reorganize the international monetary system along lines that would be appealing also to the United States. Examples of global macroeconomic policy coordination are likely to become even more rare that in the past. That would be unfortunate; as argued by Krugman (1991) the international adjustment process has worked in the second half of the 1980s and it is clear that it was helped at important points by international policy coordination, aiming first at bringing the dollar down and then at moderating its fluctuations the deep fall of 1985-87. Similar efforts could be required in the future.

References

Alesina, Alberto and Guido Tabellini. 1987. A positive theory of fiscal deficits and government in a democracy. NBER Working Paper No. 2308. Cambridge, Mass.

Bergsten, C. Fred. 1990. The World Economy.
Foreign Affairs 69: 96-112.

Cooper, Richard N. 1985. Economic interdependence
coordination of economic

policies. In Ronald W. Jones and Peter B. Kenen (eds.), Handbook of International Economics Vol. 2. Amsterdam, p. 289-331.

Currie, David, Gerald Holtham and Andrew Hughes-Hallett. 1989. The theory and practice of international policy coordination: coordination pay? In Ralph C. Bryant et al. (eds.) Macroeconomic poli-

cies in an interdependent world. Washington
C: 14-58.

Dobson, Wendy. 1991. Economic Policy Coordination: Requiem or Prologue? Policy in International Economics 30. Washington D. C, Institute for International

Frankel, Jeffrey and Katherine Rockett. 1988. International macroeconomic policy coordination policy-makers do not agree on the true model. American Economic Review 78: 318-40.

Hamada, Koichi. 1976. A strategic analysis of
monetary interdependence. Journal of Political
83: 677-700.

Bang Jespersen, Lene Louise. 1991. Fiscal implications of EMU - A case for coordination? Institute of Economics, of Copenhagen.

Kenen, Peter B. 1988. Managing exchange
rates. London.

Kindleberger, Charles. 1986. International public goods without international government. Economic Review 76; 1-13.

Krugman, Paul. 1991. Has the adjustment process worked? Policy Analyses in International 34. Washington D. C. Institute for International Economics.

Olson, Mancur J. 1971. The logic of collective
action. Cambridge, Mass.

Oudiz, Gilles and Jeffrey D. Sachs. 1984. Macroeconomic policy coordination among the industrial economics. Brookings on Economic Activity, Vol. 1:1-75.

Padoa-Schioppa, Tommaso. 1984. Money, economic policy and Europe. Luxembourg, for Official Publications of the European Communities.

Polak, Jacques. 1988. Economic policy objectives policy making in the major industrial In Wilfred Guth, ed. Economic coordination, Washington D. C:l-

Roubini, Nouriel and Jeffrey D. Sachs. 1989. Political and economic determinants of budget deficits in industrial democracies. European Economic Review 33: 903-38.

Teisen, Hans Christian. 1991. Økonomi og
politik i europæisk integration.Atø//øna/økonomisk
129; 148-64.

Thygesen, Niels. 1991a. Economic and monetary in Europe and global economic cooperation. International Spectator XXIV: 70-86.

Thygesen, Niels. 1991b. Monetary constitutions a framework for economic and political stability. Unpublished paper for Conference '175 Jahre Oesterreichische Nationalbank' Wien.