Intertemporal Utility Maximization and the Timing of Transactions
基于跨期效用最大化模型，区分购买（市场活动）与消费（非市场活动），解释家庭消费和购买计划的选择，对理解货币作用、短期总需求波动及货币需求存货理论有重要意义。
This paper addresses the problem of explaining a household's choice of consumption and purchasing plans on the basis of a model of intertemporal utility maximization. Until recently the intertemporal choice models that have been developed by economists have simply not distinguished between purchasing, a market activity, and consumption, a nonmarket activity.' The importance of this distinction, and of incorporating it into a model of intertemporal choice, can be seen from the point of view of three separate areas of current research. First, recent work on the microfoundations of monetary theory2 has shown the importance of transaction costs in explaining the role of money in economic activity. The absence of these costs from standard general equilibrium theory makes it difficult to account for the special characteristics, and even the existence, of money within that framework. This research has underlined the need to develop a theory of transactions on the same level of sophistication as our theories of production and consumption. The present problem may be viewed as one part of the larger problem of developing such a theory of transactions. Second, in the area of short-run aggregate analysis, purchasing decisions are of more interest than consumption decisions because they are more closely related to the level of aggregate demand. Recent empirical investigations by Michael Darby have supported Milton Friedman's conjecture that the aggregate rate of purchase of consumer durables can undergo large fluctuations even when the aggregate rate of consumption is relatively constant. It would clearly further our understanding of short-run fluctuations in aggregate demand if both of these rates could be explained on the basis of intertemporal choice. Third, the area most closely related to the present problem is the inventory theory of the demand for money. This theory has been extended in recent years by several authors into a generalized theory of the size and timing of all sorts of transactions, including wage payments, commodity purchases and sales, and various