PIMP1. “Volatile Policy and Private Information: The Case of Monetary Shocks,” with R. E. Manuelli), Journal of Economic Theory , 99, 265-296, 2002. Download a PDF copy of this paper.
PIMP2. "Crises, Private Information and the Collapse of (Non-tradeable) Asset Markets," 1999, with R. E. Manuelli.
PIMP3. "Increases in Risk and the Probability of Trade: A Note," 1999, with R. E. Manuelli.
(You have to love this acronym...) PIMP1 was built around a common anecdote from hyperinflations. This is that it is very common for trade to break down completely. In the Argentine hyperinflation of the 1980's for example, many stores closed posting in their windows a sign reading "Closed for Lack of Prices." What this paper does is model a 'Lemon's' effect in money when agents are differentially informed about recent inflation. Uninformed parties are concerned about nominally denominated trades since it may be that their trading partners have better information than they do about the 'value of money.'
Formally, we analyze a model of the effects of private information about the size of monetary expansions on the outcome of bargaining. We show that the social cost of this private information depends on the volatility of monetary policy and that as this volatility increases, the gains from trade decrease to zero. We endogenize the choice of information purchases and show that for moderate volatility, there are two sources of welfare loss. The first is that described above when traders are asymmetrically informed. In addition to this, there is wasteful generation of information that is only useful because the policy is volatile.
PIMP2 and PIMP3 are two further applications of this set of ideas, PIMP2 deals with financial crises, while PIMP3 deals with the technical question of when increases in risk decrease the probability of trade in an auction.