INCOME FLUCTUATION AND CONSUMPTION:THEORETICAL RESULTS AND REMARKS ON EMPIRICAL STUDIES (in Japanese)

In this paper, I survey the theories how income fluctuation affects the process of consumption, especially on the life-cycle/ permanent income hypothesis and the precautionary motive of saving. I also give some remarks on empirical studies of consumption. First I reexamine the implication of the life-cycle/ permanent income hypothesis. This theory has a concrete microfoundation, but there was no valid econometric method for verifying, until the well-known "Random-Walk Hypothesis" of Hall (1978). Hall's hypothesis was rejected statistically in two aspects, however. The one is that the current and/or past levels of income help the prediction of the future consumption. This is called "excess sensitivity" of consumption. The other is that the variance of consumption is strictly smaller than that of the permanent income. This is called "excess smoothness" of consumption. Some studies tried to explain these two facts by extending the standard "certainty-equivalence" model in some aspect, such as preference shock, durables, habit formation, and aggregation of different agents. We reviewed some of these models with the assumption that the instantaneous utility function is quadratic, and we found that some extensions can explain both excess sensitivity and excess smoothness only if the stochastic process of income is nonstationary. In these cases, the response of consumption to unexpected income change is smaller than that of the certainty-equivalence model. The rest of the paper, I survey the studies of precautionary savings. Roughly speaking, a household has a precautionary motive of saving if the third derivative of his instantaneous utility function, u'''(c), is positive. The work of Kimball (1990) on "prudence" gives the measure of the strength of the precautionary saving motive. Caballero (1990) studies the optimal consumption-saving decision of an infinitely living household with constant absolute risk aversion (CARA) utility. He shows that, in case of homoscedastic labor income, the stochastic process of consumption follows a martingale with drift, which is just like in the certainty-equivalence model. Next we review the "buffer stock" theory of saving, propounded by Carroll (1992, 1997). He shows that, if the absolute risk aversion is decreasing in consumption, a household regards its non-human wealth as a "buffer" for labor income fluctuation and chooses consumption so as to keep his non-human wealth to an appropriate level. A linear approximation of this "buffer stock" model explains both excess sensitivity and excess smoothness at the same time, even if the stochastic process of income is stationary. Finally the advices on empirical studies of precautionary savings, which is derived from the theoretical studies above, are given. Many of the previous studies neglect to examine the behavior of the stochastic processes of independent variables, and suffer serious misspecifications of the lack of non-human wealth.