Money, intermediation and coordination in decentralised markets
Overview: This thesis studies the coordination of individuals' transactions in a large, decentralized market. The first half of the thesis examines the role of market institutions in an environment with frictions. In particular, it studies the interaction between "intermediaries" (banks, shops) and decentralized "equilibrium arrangements" such as money or credit. The second half of the thesis studies the role of public and private information in coordinating individual actions, as well as the macro-economic effects of such coordination. First Half: I study a search economy in which intermediaries are the driving force co-ordinating the economy on the use of a unique, common medium of exchange for transactions. If search frictions delay trade, intermediaries offering immediate exchange opportunities can make arbitrage gains from a price spread, but they have to solve the search market's allocation problem. Intermediaries solve this problem best by imposing a common medium of exchange to other agents, and a Cash-in-Advance constraint arises in equilibrium: Agents trade twice in order to consume, once to exchange their production against the medium of exchange, and once to purchase their consumption. I extend my analysis to the study of fiat currencies and, in particular, free banking regimes. Second Half: The second half consists of two essays studying the role of public and private information in coordination games. In the first, I relate the convergence of equilibria to the convergence of higher-order beliefs. The central result of this essay relates the convergence of players' higher-order beliefs (and hence equilibrium convergence) to the parameters of the signal structure. This provides a limit condition determining the uniqueness or multiplicity of equilibria in the coordination game. Building on the previous chapter, the second essay studies the effects of information policies on output and inflation, when price-setters face higher-order uncertainty concerning the money supply. I show that this may lead to substantial delays in price-adjustment following a shock. To the extent that public announcements coordinate expectations, they reduce this delay, and thereby reduce the effect and persistence of monetary shocks on output in the short-rim. On the other hand, public announcements introduce a second component of noise, and may therefore increase short-run volatility.