The Many Reasons Why the Theory of Economic Policy Has an Impact on Institutions

A number of institutional issues arise in organising an eco­nomic and social system, on which the strategic theory of economic planning can throw light, in particular, because the theory of economic policy set up in Chapter 4 states conditions for (1) target controllability, (2) equilibrium existence and (3) multiplicity of equilibria.^[142] These issues involve choosing the optimal institutional mix in terms of market and government, providing for proper signals to be issued by markets and policymakers, searching for a variety of effective policy instruments and their coordination, sol­ving conflicts between groups and ensuring equilibrium for the system as a whole.^[143]

Each of these issues is strictly interconnected with the others.

Economic systems are characterised by instability, which can derive from either short-run fluctuations and/or the existence of multiple long-run equilibria. In the former case, given an equilibrium path, instability takes the form of short-run fluctuations around that path. In order to reduce their size, control of the system by public discretionary action would be required, but this is possible only when a sufficient number of instruments is available.

In the case of multiple long-run equilibria, the economy might be trapped in a suboptimal equilibrium for a long time with high welfare losses or may switch between different equilibria, thus causing instability. Multiple equilibria do indeed depend on a number of situations. Generally speaking, they derive from uncoordinated actions set at a decentralised level and, more precisely, from the absence of mechanisms to coordinate the expectations of the players, even when their target values do no conflict.^[145] This is especially so in the case where the market is more relevant. The case of no conflict is, however, very difficult to find because conflicts over goals predominate. To some extent, conflicts can be at least partially reduced by implementing a decentralised setting with proper limits to freedom of action. The opportunities deriving from decentralised solutions and their limitations will now be discussed.

7.6.1 Decentralised Conflicts

Decentralisation is a dominant political trend of our time as a key to good governance. This needs some justification and clarification. In Chapter 6, we have already dealt with the issue of accountability, finding mixed arguments in favour of or contrary to decentralisation. Possibly, the main determinants for practical cases of decentralisation are the ‘third wave' of democratisation and the explosion of centrifugal forces due in particular to inter-communal ethnic groups (Reilly 2001)^[146] tending to the formation of communities that are homogeneous from the point of view of race and religion. However, scarcity of employment opportunities and defence from immigration have been other important determinants - not exclusive of those already indicated.

As discussed in Chapter 4, in a decentralised setting where there are multiple equilibria that differently influence agents having different goals, announcements can help all players to converge on a certain equilibrium, possibly mediating between their conflicting goals. When, instead of single agents, subsets of agents operate having identical goals - and thus implicit or explicit coalitions exist - conflicts can be overcome, possibly to the benefit of some specific coali­tion. In fact, as an extension of the concept of controllability by a single agent, the theory of economic policy in a strategic context has considered the related possibility of coalitions or implicit coalitions between agents or policymakers. In this case, these continue to act non-cooperatively but appear to operate as if they were in a coalition and thereby benefit by appropriating the coalition gains over the fully non­cooperative solution. For the system as a whole, this pro­duces an intermediate position - partly cooperative and partly non-cooperative or partly centralised and partly decentralised. This is an important case to look at because the gains to the implicit coalition partners will be larger than those they could achieve in a fully cooperative solution, if the coalition is powerful enough to expropriate some of the gains that would otherwise go to the coalition’s non­members.^[148]

In short, the implicit coalition approach has studied com­mon-interest games for some of the agents or policymakers, discovering that there might be no equilibrium if more than one controlling coalition exists having conflicting inter­ests with other coalitions, similarly to what happens in a two-player game. More generally, unique or multiple Nash equilibria or no equilibrium will exist.

Since the coalition outcomes, either implicit or not, will make each participant better off than under purely non­cooperative decision making, and may produce an equili­brium when unfettered non-cooperative decisions do not, we need an institutional framework that allows coalition members to agree on target values. That is largely a matter of trading and agreeing information on the policy targets and expected outcomes for each player, and then modifying those target values as a step necessary to make it possible to realise the gains of the agreed coalition policies for each participant.

In a broader perspective, implicit coalitions may thus be viewed as the credible operating mechanism to implement societal cooperation based on bargaining on the objectives and reinforced by a commitment technology. Policymakers - or people - with different conflicting objectives can agree to pursue sets or subsets of common goals that will be bound by the conditions of implicit coalition. These conditions, together with the technologies of commitment on the target, will ensure the credibility of the policy. They will agree if the cost of such a policy is lower than the cost associated with the creation of a ‘common institution' or self-government to manage their policy and internalise the externalities from the non-cooperative solution. The institutional lesson here is that agreeing on target values, not policy rules, may be the key to achieving better overall outcomes.

7.6.2 CentraIised Solutions

There are a number of scenarios where centralised decision­making is needed together with some kind of institution to support and promote it rather than to leave the field free for decentralised (and possibly self-interested) institutions. This provides a rationale for why such institutions are necessary in a number of rather common cases.

One set of situations where institutions are needed to support centralised or coordinated decisions will involve an institutional framework directed by a hegemon or desig­nated leader, aiming at (one hopes) of the common good or a common institution jointly run by the participants as a social planning device. This will be the case when extern­alities or other kinds of interrelationships exist between dif­ferent communities. The need for centralised institutions or coordinated decisions may arise both when there is no defined equilibrium and when there are multiple equilibria: in the former case where we have to pick out a sequence of policy positions that are acceptable (the hegemon/desig- nated leader case) and in the latter where we need to choose one equilibrium that maximises the joint gains to all on a Pareto-optimal or optimal bargaining basis (the social­planner case). A social planner would have no problem coordinating his or her action as he or she would determine all the instruments at the same time on behalf of the common good.

A second set of situations emerges when a coordinating institution is needed to oversee a sequence of policy announcements intended to act as an equilibrium selection device. In this case, the institution coordinates all the infor­mation sets that players need in order to establish the cred­ibility of the announcements. The need for institutions regulating this type of situation derives from cases where there are multiple (and not necessarily equally good) equili­bria or where there is no defined equilibrium and where there are multiple equilibria and where the equilibrium is not optimal.

A centralised solution has many practical problems of implementation, in the case of both common-interest games and conflicts. In the former, for instance, centralisation may have high costs in collecting the information needed or can be subject to costly delays. There are also issues of moral hazard in the transmission of preferences from the members of the community to the social planner. In the latter case, which involves conflicting goals between the members, centralisation also has to solve the problem of which equili­brium the social planner should converge to. Conflicts between the players do not guarantee a clear answer to this question.

In fact, mediation of conflicting targets and target values of the various players, in order to suggest a final acceptable compromise, implies a hard task for a centralised institu­tion. In some cases, some players may facilitate the equili­brium of the system as a whole in a way that satisfies the goals of the other players. In others, some players can exploit to their advantage either centralised or decentra­lised institutions. A further issue relevant to deciding on the degree of centralisation refers to the extent of the con­flict of interests between the government levels. Dreher et al. (2016) show that the extent of misaligned interests and the relative importance of local and central government knowledge affect the optimal choice of policy decision schemes.

In practical life, examples of centralised institutions abound both at a domestic and at an international level, even if they do not perfectly correspond to the ideal types featured earlier. And many of those characteristics may be represented by one real institution.

At a domestic level, practically all the cases where a government performs tasks beyond those of the minimal state to avoid microeconomic and macroeconomic failures will end up coordinating private decisions and, possibly, mediating between the different interests of agents. At an international level, some of the more obvious institutions deal with trade policy, especially where tariff or non-tariff barrier reductions are concerned or international monetary arrangements and the management of exchange-rate regimes (especially with respect to the choice of exchange-rate target values for joint stability). Another example would be institu­tions for protecting the environment.

Examples of coordinating institutions to jointly manage policy (in the sense of a social planner) are the World Trade Organization (WTO) for trade policy and the International Monetary Fund (IMF) for international monetary arrange­ments or the European Commission to oversee Europe's Stability and Growth Pact (SGP), the Fiscal Compact and other common rules. Some of these institutions allow a hegemon or designated leader to act as a coordinated focal point. Among them, the exchange-rate system instituted at Bretton Woods, which allowed the United States to act as hegemon, or the European Monetary Committee (as it was) to oversee the European Monetary System of exchange rates that allowed Germany to act as designated leader for EU monetary policies.

More controversially, perhaps, one could argue that the iterative process of national budget plans being monitored and then adjusted in the light of anticipated changes in European Central Bank (ECB) monetary policies, under the European Commission's convergence and medium-term objectives programme, is an instance of a sequence of policy announcements leading (if everyone keeps to the rules) to a selected equilibrium point. The same argument could be used more generally for cases where fiscal and monetary policies adjust to each other in a series of policy changes, where the fiscal authorities and central bank both have instrument independence (as in the United States and the United Kingdom, for example).

The lesson here is that we often need some form of centra­lised policymaking to devise policies or to get to a policy equilibrium or a stable point where the economic system is over-determined or under-determined since either multiple equilibria or no equilibrium would otherwise prevail as a result of policy conflicts. Policies in these circumstances will break down unless suitable institutions are brought into play. This is not an issue of whether the policies are effective or not but of what to do when there are multiple or no under­lying equilibria. In this case, strong centralised policy insti­tutions are unavoidable. However, they should not be to the benefit of some countries and the specific interest groups therein (or the interest groups of institutions spanning the borders of countries, such as transnational corporations). In other words, we should aim at a system that is different from the current system and ensures ‘global governance without global government', as Stiglitz (2002: 21) says. In fact, Stiglitz suggests that proper government requires representation of all the interests at stake and transparency and accountability of leaders.^[150] Stiglitz (2006: chap. 10) adds at least two other requirements for a proper government: better judicial procedures and better enforcement of the international rule of law.

7.7