Monetary and Financial Interactions in the Business Cycle
Fuerst, Timothy S., Evans, harles L., Gertler, Mark, Journal of Money, Credit & Banking
Many students of the business cycle view the cycle as composed of two logically distinct forces: a random shock or impulse to the environment, and, a propagation mechanism by which these shocks are transformed into persistent movements in economic time series.(1) A propagation mechanism that business cycle theorists often appeal to is capital accumulation. The familiar story goes something like this. Suppose that an impulse leads to a boom in current output. This output boom increases expenditures on capital accumulation, which in turn propagates the impulse forward by expanding future technological possibilities. One dividend of the real business cycle (RBC) research program is the demonstration that this propagation mechanism is remarkably weak. To be precise, in a standard RBC model a serially uncorrelated technology shock delivers a trivial amount of propagation in economic aggregates (for example, King, Plosser, and Rebelo 1988, Cogley and Nason 1993). This absence of a significant internal propagation mechanism is not necessarily a problem for the RBC model because of empirical evidence in support of highly persistent technology shocks (for example, Prescott 1986). However, this is a problem for monetary models of the business cycle that layer a monetary economy onto a standard RBC framework. In many of these models, optimizing behavior on the part of economic agents results in the monetary impulse being serially uncorrelated.
The trivial amount of propagation in the basic RBC model has led to the search for other sources of persistence. This paper explores the role of financial factors in accentuating the capital accumulation channel. In particular, the paper develops a model in which an assumed informational asymmetry creates a moral hazard problem in the construction of new capital. These agency costs endogenously become less severe in periods of high output and thus lower the cost of capital accumulation. This is a supply-side force that complements the demand-side force already present in the standard RBC framework. Investment and consumption goods are perfect substitutes in the RBC model so that the supply curve for capital is perfectly elastic at a price of unity. Hence, a positive technology shock alters investment behavior in an RBC economy solely by increasing the demand for capital (because of households' consumption-smoothing motive). In the agency cost model developed below, this demand force is still present, but there is also a supply-side force: Increases in output lower agency costs and thus lower the cost of creating capital, effectively shifting the upwardly sloped supply curve to the right. (The upward slope to the capital supply curve is also a natural result of agency costs because increases in capital production exacerbate agency problems by increasing the need for external finance.) Hence, in the model developed below, a technology shock increases capital accumulation through both a demand-side and a supply-side channel. This supply-side force propagates forward as the increased capital accumulation increases output in future periods and thus also lowers agency costs in the future. The basic issue the paper addresses is the quantitative extent to which these agency costs expand the propagation mechanism in the basic RBC framework.
This paper is part of the rapidly growing literature on financial factors and the business cycle. Excellent surveys are provided by Gertler (1988) and Jaffee and Stiglitz (1990). The subset of papers closest in structure to the model presented below include Bernanke and Gertler (1989), Fisher (1994), and Fuerst (1994b). Bernanke and Gertler (hereafter denoted by BG) developed a Diamond (1965)-type overlapping generations model in which each generation consists of agents of two types: entrepreneurs and lenders. Along with supplying labor to firms engaged in the production of the consumption good, the entrepreneurs have access to a stochastic technology that transforms the consumption good into the capital good. …