Do Good or Do Well? Public Debt Management in a Two-Party Economy

Policy rules can be welfare improving if the government faces a time inconsistency problem when it chooses its optimal policy sequentially. In this paper I show that a conservative government, while favourable in principle to the institution of binding policy rules, may rationally choose not to do so in the presence of electoral competition. The reason is that left-wing governments typically face more serious time inconsistency problems than right-wing ones, because of incentives to raise revenue through inflation or tax capital for distributional reasons. A binding rule may make the left-wing party more "credible" in the eyes of the voters, conversely in the absence of rules, voters will favour the conservative party- for example because of its stronger anti-inflationary preferences. I develop this idea in the context of a simple debt management model, in which the incumbent can decide whether to roll over a given amount of debt in nominal form or indexing debt to the price level. An election follows, and after the election the new government has to repay the debt using taxes and inflation, both assumed to be distortionary. The right-wing party attributes a higher weight to the inflation tax distortion than the left-wing party; the median voter's preferences are in-between the two parties'. Also, both parties face a fixed cost of being out of power. In the absence of a commitment technology, nominal debt will artificially enlarge the inflation tax base, since after the election the government will take nominal interest rates as given and try to reduce the real value of its obligations through inflation. With rational expectations and full information, once the indexation parameter is chosen, voters know who will fully incorporate future inflation. As a result, inflation is too high and taxes too low. If the costs of inflation are convex, voters will be more likely to choose the right-wing party if debt is nominal. The results extend to a setting in which parties differ in their preferences for private vs public consumption. In the presence of uncertainty, a party in power will face a trade-off between the incentive to "tie the hands" of its eventual successor in case of an election loss ("do good") and the incentive to make its opponent look "bad" in the eyes of the voters in order to win the election ("do well").