“Mario Draghi”:http://www.ecb.europa.eu/press/key/date/2011/html/sp111118.en.html
bq. Let me use this occasion to dwell a bit further on monetary policy in the current environment. Three principles are of the essence: continuity, consistency and credibility. Continuity first and foremost refers to our primary objective of maintaining price stability over the medium term. Consistency means to act in line with our primary objective and with our strategy both in time and over time. Credibility implies that our monetary policy is successful in anchoring inflation expectations over the medium and longer term. This is the major contribution we can make in support of sustainable growth, employment creation and financial stability. And we are making this contribution in full independence. Gaining credibility is a long and laborious process. Maintaining it is a permanent challenge. But losing credibility can happen quickly – and history shows that regaining it has huge economic and social costs.
“Daniel Davies”:https://crookedtimber.org/2006/11/29/reputations-are-made-of/ (on the Iraq War, but trust me – the logic travels).
bq. At this late stage in the occupation of Iraq, many of Henry Kissinger’s old arguments about Indo-China are being dusted down. One of the hoariest and worst is that we need to “stay the course” (or some similar euphemism) in order to maintain “credibility” – to demonstrate our resolve to our enemies, who will otherwise continue to attack us. It reminds me of my one and only contribution to the corpus of game theory.
bq. The Folk Theorem in game theory states that any outcome of a repeated game can be sustained as an equilibrium if the minimax condition for both players is satisfied. In plain language, it can be summarised as stating that “if we take strategic considerations into account, there is a game-theoretic rationale for practically anything”. This formulation leads on to my contribution, the Davies-Folk Theorem, which states that “if we take strategic considerations into account,there is a game-theoretic rationale for practically fucking anything” (it’s a fairly simple corollary; proof available from author on request).
bq. The point being that since game theory in general provides the analyst with so many opportunities to twist himself repeatedly up his own arse like a berserk Klein bottle, if a given real-world course of action appears to have nothing going for it other than a game-theoretic or strategic justification, it’s almost certainly a bad idea. Thus it is with that bastard child of deterrence, “credibility”.
bq. … The idea is that the war is costing huge amounts of money and lives with no real prospect of success and a distinct danger that it is making things much worse. However, to do the logical thing would send the signal to our enemies that we will give up if fought to a pointless bloody standstill. Therefore, for strategic reasons, we must redouble our efforts, in order to send the signal to our enemies that we will fight implacably and mindlessly in any battle we happen to get into, forever, in order to dissuade them from attacking us in the first place. It’s got the kind of combination of “counter-intuitive” thinking and political convenience that always appeals to the armchair Machiavelli, as well as to the kind of person who thinks it’s witty to describe things as “Economics 101” (Airmiles has been all over this one for ages, naturally). What’s it like as a piece of game-theoretic reasoning?
bq. Lousy. It is certainly true that one of the benefits of doing something stupid is that it saves you from having to spend money on maintaining your reputation as an idiot. However, is the reputation of an idiot really worth having?
bq. It turns out that it can be proved by theorem that the answer is no. If the game of being a belligerent idiot with no sensible regard for one’s own welfare _was_ worth the candle, in the sense of conferring benefits which outweighed the cost of gaining it, then everyone would want to get that reputation, whether they were genuinely an idiot or not. But if everyone wanted that reputation, then everyone would know that simply acting like an idiot didn’t mean that you were one, in which case it would be impossible to establish a reputation as an idiot in the first place. The point here is that it’s one of the more important things in game theory that a signal has to be a _costly_ signal to be credible; like membership of the Modern Languages Association, a reputation in deterrence theory is something that is worth having, but not worth _getting._ People who use the word “signal” in this context (usually on the basis of a poorly understood or second-hand reading of Schelling) don’t always seem to realise that they are explicitly admitting that the costs of being in Iraq are greater than the benefits.