Intersectoral Labor Immobility, Sectoral Comovement, and News Shocks

Abstract

Sectoral comovement of output and hours worked is a prominent feature of business cycle data. However, most two-sector neoclassical models fail to generate this sectoral comovement. We construct and estimate a two-sector neoclassical Dynamic Stochastic General Equilibrium (DGSE) model generating sectoral comovement in response to both anticipated and unanticipated shocks. The key to our model's success is a significant degree of intersectoral labor immobility, which we estimate using data on sectoral hours worked. Furthermore, we demonstrate that imperfect intersectoral labor mobility provides a better explanation for the sectoral comovement than an alternative model emphasizing the role of labor-supply wealth effects.JEL codes: E13, E32

Uncertainty shocks and the relative price of investment goods

“…Imperfect intersectoral labor mobility is consistent with persistent sectoral wage di↵erentials (e.g., Krueger and Summers, 1988;Neumuller, 2015). Horvath (2000) and Katayama and Kim (2015) report low estimates of intersectoral labor substitutability. Beaudry and Portier (2011) present empirical evidence that intersectoral labor mobility is not su cient, so that the returns to labor between individuals initially attached to di↵erent sectors is not equated.…”

“…We use the conventional King-Plosser-Rebelo momentary utility function adopted by Basu and Kimball (2002), Shimer (2009), and Katayama and Kim (2015):…”

“…We model inflexible intersectoral labor mobility by assuming that labor devoted to each sector is imperfect substitutes to the representative household, which inhibits labor from moving across sectors. This assumption is motivated by the findings of Horvath (2000), and Katayama and Kim (2015), who report a relatively low estimate for the elasticity of substitution of labor across sectors.…”

“…We set the elasticity of intratemporal substitution in labor supply, ✓ to 0.3030, the Frisch elasticity of labor supply ⌘ to be 1.25 and the investment adjustment cost parameter  to be 1.43. These values are based on the estimates in the two-sector model of Katayama and Kim (2015). The parameter values used in the simulations are summarized in Table 1. Under these parameter values, the steady-state share of the flexible-price investment sector is 31 percent.…”

The Macroeconomic Consequences of Stimulating Offline Consumption during COVID-19

“…We choose the same parameter values for the rate of depreciation, δ, and the weekly discount factor, β, used in Eichenbaum, Rebelo, and Trabandt (2020). In addition, we revise the coefficients in the adjustment cost for investment as suggested in Katayama and Kim (2018) on a weekly basis. Following Bigio, Zhang, and Zilberman (2020), we set σ, a parameter governing the substitutability between two types of consumption goods, as 1.8.…”