RDP 8801: Time-Consistent Policy: A Survey of the Issues 4. Competing Political Parties

The importance of the policymaker's time horizon has already been mentioned. This section outlines the implications of competing political parties for the issue of time consistency. In the problem discussed so far, the issue has been whether a government with an infinite horizon would renege on policy promises. It was argued above that a finite horizon changed the nature of the problem and we introduced the possibility of asymmetric information in understanding which party was in power. A new literature[10] has recently emerged, which examines the issue of political parties with difference preferences rotating in power due to a stochastic election process. This literature has been labeled the “partisanship theory” of macroeconomic policymaking. It extends the game to strategic interaction between the private sector and two political parties. The objective function of the government can no longer be assumed to be well defined and stable. The objective function of the government ex-ante is uncertain and ex-poste is only certain for the period of rule. It has very different implications to the “median voter” theory which argues that parties will offer virtually the same policies before an election which appeal to the median voter. The partisanship approach assumes that voters realize that after the election, the winning party will implement policies which favour its traditional constituencies. The parties realise that the electorate understand these motives and therefore convergence of policies is incomplete.

In the discussion above, it was argued that the problem of time consistency was how to bind future government to follow a particular policy. Lucas and Stokey (1983) pointed out that the future behavior of government could be affected by the current government by leaving the economy in a particular state. For example, by leaving a large government debt, a government can limit the ability of future governments to follow expansionary fiscal policy. The partisanship theory makes the argument even more complicated because a government can regain power at a future date and therefore will be careful to choose policies which may restrict the policy moves of the opposing party when it gains office, but doesn't cause severe problems for its own future government.

As an example of the change in policymaker behavior, it is worth highlighting the results in Persson and Svensson (1987). In this paper, the authors show a case where there are two parties; one desires small fiscal deficits and the other desires larger deficits. Suppose the ruling party, which places large weight on small fiscal deficits, is faced with a high probability of the opposition party gaining power. The ruling party may decide to undertake a fiscal expansion and generate a large budget deficit today in order to restrict the policy option of a future expansionary government by leaving it a large debt. This restricts the extent of expansionism that the new administration can follow. Similarly McKibbin, Roubini and Sachs (1987) give an illustration in which a government that dislikes inflation will nonetheless pursue inflationary policies to restrict the inflationary policies of future expansionary governments.

An extension to this analysis would be to examine the implications of a non-political, infinite horizon monetary authority on the nature of the games pursued using fiscal policy between two political parties and a forward looking private sector. The realization in the most recent literature that political and economic institutions are crucially linked is an important development in the theory of economic policy formulation.

A further extension of the literature is linked to the new work on the problem of international policy coordination[11]. Once the problem is extended to that of a multi-level game in which governments in different countries are interacting with each other as well as with private individuals and firms within each country, the game between a domestic government and its private agents may be significantly effected by the game between countries. An example of the possible importance of this case is the agreement by the German government at the 1978 Bonn Summit to reflate the German economy at a time when the domestic political arguments were overwhelmingly against this course. By committing itself to an external agreement, the government was able to dramatically influence the domestic game.


See Alesina (1985), Alesina and Tabellini (1987), Persson and Svensson (1987) and McKibbin, Roubini and Sachs (1987). [10]

See the papers in Buiter and Marston (1985) and Blackburn (1987) for a summary. Also see McKibbin and Sachs (1987) for an empirical application of the game theoretic techniques to a empirical model of the world economy. [11]