Academic journal article Journal of Money, Credit & Banking

Persistent and Transitory Shocks, Learning, and Investment Dynamics

Academic journal article Journal of Money, Credit & Banking

Persistent and Transitory Shocks, Learning, and Investment Dynamics

Article excerpt

THIS PAPER INTRODUCES a new approach to understanding investment. The novelty does not lie in the optimization problem, the equation for the evolution of capital, the technology faced by the firm, or in heterogeneity across firms. The distinctive features are that we model the shocks that affect investment as consisting of both transitory and persistent components and that we take seriously the problem that firms face in disentangling transitory from persistent shocks.

Looking back from today's vantage point, it is widely agreed that the real interest rate was moderately high in the United States in the 1960s and early 1970s, close to zero in the mid-to-late-1970s and then substantially higher in the 1980s. At the time, however, these persistent tendencies of the real interest rate were frequently obscured by short-run variations. Instead of ignoring this empirical tendency of the real interest rate to persistently gravitate to a certain level, we directly model it--and the ensuing consequences for investment behavior. Because of the substantial transitory fluctuations around these persistent interest rate levels, it also seems important to recognize that firms face the difficult problem of trying to determine whether or not the real interest rate has shifted from one persistent level to another. We therefore also model this learning process and explore its implications for investment behavior.

The resulting model generates interesting dynamics. For example, Caballero and Engel (1994) have argued that time variation in the response of investment to shocks is important in understanding investment. Simulations of our model show that the response of investment to changes in the interest rate can vary widely over time. This is partly because the model shares a feature of recent S,s models, namely, that the current response of investment depends on the sequence of past shocks.

The wide variability in the response of investment to a given change in the interest rate is closely linked to another interesting feature of investment dynamics in the model. A firm's beliefs matter, where "beliefs" refer to the firm's assessment of the probability that next period's economic fundamentals will be favorable. (1) Our simulations show that investment will respond less when the firm is quite confident about its beliefs.

Because of the importance of beliefs, there is an asymmetry in the effect of shocks. For example, if the firm believes that the real interest rate will be persistently high and a shock raises the interest rate, the change in the interest rate will have little or no effect on investment. We refer to such shocks as confirming shocks because they tend to confirm existing beliefs. On the other hand, if the firm believes that the real interest rate will be persistently high and a shock lowers the real interest rate, the change in the interest rate can have a large effect. We refer to these as challenging shocks.

For decades, economists have suggested that "animal spirits" play a role in investment and have tried to understand how and why this might be the case. The model in this paper provides a possible interpretation. If a sequence of transitory shocks leads firms to believe the state has changed, either an investment boom or a crash can occur without any change in the true state of economic fundamentals. These booms and crashes arise because of the important role played by beliefs and because of the asymmetric and variable effect of transitory shocks on beliefs. (2)

In addition to these interesting features, our model is able to capture some of the features of investment behavior documented in previous empirical research. (3) First, empirical studies find that investment responds very sluggishly to shocks. Second, the estimated elasticity of the capital stock with respect to the relative price of capital is low. Third, several studies have found that the neoclassical theory of investment with adjustment costs (often referred to as the Q theory of investment) fails to explain a large portion of the variation in investment. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.