Conclusion
New classical models of aggregate fluctuations, such as the stochastic growth model, imply that aggregate fluctuations are caused by real disturbances, such as productivity shocks.
This is why such models are also called RBC models.New classical models, such as the stochastic growth model analyzed in this chapter, are DSGE models based on optimizing behavior by both households and firms, flexible prices, and fully competitive markets. Households maximize their intertemporal utility, firms maximize the present value of their profits, and markets function efficiently.
If the competitive general equilibrium models of this kind could explain all features of aggregate fluctuations, then there would be little need for models that stress distortions in product and labor markets, or other market inefficiencies. However, new classical models have significant weaknesses as models of aggregate fluctuations.
Before discussing these weaknesses, we first analyze a new classical model without capital, which is simpler to handle and can serve as a basis of comparisons with the new Keynesian models that will be analyzed in chapters 16 and 17.
1. The original stochastic growth model was presented by Brock and Mirman [1972] and was essentially a Ramsey model, augmented by stochastic shocks to productivity. Later models allowed for endogenous labor supply, encompassing the intertemporal substitution hypothesis of Lucas and Rapping [1969]. The approach analyzed in this chapter follows Kydland and Prescott [1982], Long and Plosser [1983], and Prescott [1986]. It is surveyed in, among others, Plosser [1989], Mankiw [1989], Stadler [1994], and King and Rebelo [1999]. As we shall see in chapters 16 and 17, this approach has also influenced the new Keynesian approach to aggregate fluctuations, which now also relies on DSGE models.
2. In many ways, the model in this section is a generalization of the two-period models analyzed in chapter 2 to allow for an infinite horizon, uncertainty, and stochastic shocks to productivity.
3. Appendix F contains an introduction to stochastic processes.
4. The assumption that the processes driving labor productivity and real government expenditure are AR(1) is made for analytical simplicity and can of course be generalized.
5. The concept of intertemporal substitution in labor supply was first analyzed in an important paper by Lucas and Rapping [1969]. For an empirical investigation of its significance for fluctuations in employment in the United States and the United Kingdom, see Alogoskoufis [1987a,b]. See also Hansen [1985] and Rogerson [1988] for the implications of a generalization of the model to the case where employment is indivisible, and households can work for a minimum number of hours or not work at all.