Drazen: Political economy in macroeconomics
From WikiSummary, the Free Social Science Summary Database
Drazen. 2000. Political economy in macroeconomics. Princeton: Princeton University Press.
CHAPTER 4: THE PROBLEM
THE PUZZLE OF TIME INCONSISTENCY
- Time inconsistency happens when an actor as an incentive to change his strategy in a later stage. In itself, this isn't all that interesting. When time inconsistency gets interesting is when one actor must fool a second for the second actor's own good. For example, a professor might tell his class at the beginning of the semester that there will be a final exam. The students therefore will study hard. At the time of the final, the professor says there won't really be an exam. Everybody wins--the professor wins because the students studied hard (and he doesn't have to grade), the students win because there isn't a final. But the only reason that everybody wins is that the professor deceived the students--his time inconsistency was for their own benefit.
Time inconsistency problems (Y) arise when there is sequential decisionmaking (X1) and conflicting interests (X2).
COMMITMENT VS FLEXIBILITY
Since we can't specify all possible future states of the world ex ante, it isn't necessarily good to bind your central bank too tightly. There are two issues here: how much commitment is desireable, and how to get it. Drazen considers only the second. Possible ways of binding the central bank while maintaining some flexibility:
- Escape clauses: specify [macroeconomic] conditions under which the central bank may ignore regular procedures and use only discretion
- Flexibility only in seignoriage policy ["seignoriage" refers to overprinting currency to raise state revenue]: unexpected inflation is okay only in response to unexpected revenue needs. The problem here: hard to commit credibly to this limitation.
CHAPTER 5: SOME POLICY SOLUTIONS TO THE PROBLEM
How can we make the state goals at time t credible for time t+1? What would be more credible than simply publicizing an intention/goal not to raise inflation?
- Laws: Although policy makers can always change or ignore a law [about inflation rates], there are transaction costs to changing it and penalties for breaking it. Thus, laws can enhance credibility.
- Constitutions: A law is constitutional if it (1) restricts gov't authority; (2) sets out policy making processes; (3) usually it will concern more fundamental issues, like basic rights; (4) usually it will be harder to amend. Thus, constitutions enhance credibility in much the same way that laws do, only more so [due especially to the very high transaction costs of changing it]. This might be contingent on having a government that is stable and generally has credibility when it comes to respecting its constitution/laws.
- Social norms: an unwritten law, or "a pattern of behavior that is customary, expected, and self-enforcing. Everyone conforms, everyone expects others to conform, and everyone wants to conform given that everyone else conforms." e.g. the US Federal Reserve is not formally guaranteed independence, but its independence is a strong norm.
- Delegation of authority: Giving up some control can make your agreements more credible, though this introduces p-a problems.
- Central bank independence: An important form of delegation. Macroeconomic policy requires a much longer time horizon than citizens--and therefore politicians--usually have, so politicians [wisely] delegate to technocrats with long terms in office and little exposure to politics. Not only do these technocrats make decisions with great autonomy, but it is hard to overturn their decisions. Of course, you still have the issue of how to control the central bank. Some authors encourage appointing a conservative [anti-inflation] central banker; some speak of setting publicly-announced "inflation targets" (a range within which the central bank must keep inflation rates]; others look at incentives for the central bankers; others note the beneficial effects of having a committee running the bank (as in the US).
CHAPTER 6: SOME REPUTATIONAL SOLUTIONS TO THE PROBLEM