Buyers and Sellers: Should I Stay or Should I Go?
研究在货币交换中，买方和卖方谁更愿意主动搜寻交易伙伴，发现即使买方搜寻成本更高，也可能出现只有买方搜寻的均衡，且此时货币成为每笔交易的必要媒介。
This paper explores some aspects of the exchange process that should be of interest to economists who use search theory, and especially to those who use search as a foundation for monetary economics. We are mainly concerned with a question that seems basic but has not been analyzed previously: is there some way to determine endogenously which agents are willing to invest resources in the process of active search for trading partners, and which agents prefer to wait passively for trading partners to come to them? In particular, given a group of buyers with money and a group of sellers with goods, is there any reason to expect that buyers will search for sellers, rather than the other way around? One obvious factor is the relative cost of transporting goods and money; but we are interested in examining whether there is anything about the process of monetary exchange per se, and not merely exogenous search costs, that makes buyers more willing than sellers to bear the costs of seeking out trading partners. We investigate this within a generalized version of the search-theoretic model of fiat money in Kiyotaki and Wright (1991, 1993). We find that there may exist equilibria in which buyers search while sellers do not, even if the search cost is greater for the former. However, there can also exist other equilibria with different properties. Perhaps the key finding is that the situation is not symmetric: factors that determine whether to search are fundamentally different for buyers and sellers. One property of an equilibrium in which only buyers search is that money appears on one side of every transaction, because if sellers do not search they do not meet and cannot barter. This is consistent Robert W. Clower's (1967) observation that money buys goods and goods buy money, but goods do not buy goods, although here it is derived endogenously rather than assumed. Some search-based monetary models simply rule out barter, including Peter A. Diamond (1984), Douglas M. Gale (1986), Diamond and Joel Yellin (1990), Alessandra Casella and Johnathon S. Feinstein (1990), Kiminori Matsuyama et al. (1993), Victor E. Li (1995), and Alberto Trejos and Wright (1995). It is difficult to motivate this absence of barter in these papers. In a model that is similar to the one used here, except search is costless, S. Rao Aiyagari and Neil Wallace (1992) prove that all equilibria involve some barter. A contribution of this paper is to show that barter may disappear once the decision to search is endogenized.