The Cost of Business Cycles under Endogenous Growth

Abstract

In his famous monograph, Lucas (1987) put forth an argument that the welfare gains from reducing the volatility of aggregate consumption are negligible. Subsequent work that revisited Lucas' calculation continued to find only small benefits from reducing the volatility of consumption, further reinforcing the perception that business cycles don't matter. This paper argues instead that fluctuations can affect welfare by affecting the growth rate of consumption. I present an argument for why fluctuations can reduce growth starting from a given initial consumption, which could imply substantial welfare effects as Lucas (1987) already observed in his calculation. Empirical evidence and calibration exercises suggest that the welfare effects are likely to be substantial, about two orders of magnitude greater than Lucas' original estimates.

On the Timing of Innovation in Stochastic Schumpeterian Growth Models

“…It is also distinct from previous work that argues cycles are costly because of their e¤ect on growth, e.g. Barlevy (2004), since here the e¤ect involves the timing of innovation rather than the level of innovation. The quantitative analysis above suggests that this ine¢cient timing triples the cost of business cycles relative to models in which growth is assumed to be una¤ected by cyclical ‡uctuations.…”

“…The quantitative analysis above suggests that this ine¢cient timing triples the cost of business cycles relative to models in which growth is assumed to be una¤ected by cyclical ‡uctuations. This is more modest than the costs cycles impose because of their e¤ect on the level of growth as reported in Barlevy (2004); however, the estimated cost in this paper is likely to be understated given that trend growth is more volatile than implied by the model. But even if the cost is much larger, it does not necessarily follow that stabilization is desirable, since policymakers could always avoid this cost by subsidizing R&D during recessions.…”

“…The standard deviation of trend growth in the model is about 0:2 percentage points. By contrast, when Barlevy (2004) estimates the volatility of trend growth (i.e. after accounting for changes in the level of economic activity), he …nds a standard deviation of 1:8 percentage points.…”

“…To capture the fact that productivity in the goods sector varies over time, I let z t follow a Markov switching process between two states, Z 1¸Z0 , with a constant hazard rate ¹. I treat these ‡uctuations as exogenous, although one could potentially derive them endogenously. 4 Turning next to the production for intermediate goods, I assume each good j 2 [0; 1] can be produced from labor according to a linear technology…”

Speculative Growth, Overreaction, and the Welfare Cost of Technology-Driven Bubbles

“…Given the calibrated values of  and  shown in Table 1, equation (6) can be used to recover the implied curvature parameter  1 for comparison with Barlevy (2004). Assuming an annual depreciation rate of  = 01 equation 6yields  1 = 059 when  = 04 and yields  1 = 012 when  = 10 Barlevy (2004) considers values in the range 012 ≤  1 ≤ 026 for an endogenous growth model that corresponds to the  = 10 case. As  1 → 10 the implied adjustment costs approach zero.…”