ESTIMATING INVESTMENT RIGIDITY WITHIN A THRESHOLD REGRESSION FRAMEWORK: THE CASE OF U.S. HOG PRODUCTION SECTOR

As the U.S. hog production sector becomes ever more specialized, the importance of capital inputs has heightened. Given that it is costly to adjust the capital stock and that the associated adjustment cost function may exhibit cost asymmetries between investment and disinvestment, profit-maximizing producers may find themselves trapped in a situation where it is neither profitable investing nor worthwhile disinvesting. This article addresses two issues related to the employment of quasi-fixed input in the U.S. hog production sector: does an inaction or sluggish regime exist in the demand for quasi-fixed input, and, if so, to what extent has this impeded adjustment in quasi-fixed input stock and, hence, hog output supply toward the long-term equilibrium levels? The conceptual framework is based on the work by Abel and Eberly and allows for the existence of an inaction/sluggish regime, alongside an investment regime and a disinvestment regime. Quarterly data from 1976 through 1999 are used to estimate the three-regime investment demand equation, treating breeding sows as the quasi-fixed input. The threshold estimation procedure recently advanced by Hansen is adopted. To provide a linkage between breeding herd investment and hog output supply, a hog supply equation, specified in part as a function of lagged breeding stock, is estimated by a least squares procedure. The dynamic recursive system of investment demand and hog supply is used to simulate the effects on breeding stock and hog supply of changes in the magnitude of investment rigidities. The econometric results strongly support the three-regime breeding herd investment model. More than 10 percent of the observations fall into the sluggish regime, indicating that this regime has occurred sufficiently often to warrant attention. The estimated rate of adjustment toward the long-run equilibrium breeding stock is 2.7 percent per quarter. The existence of a linkage between lagged breeding stock and hog supply is confirmed. Thus, the results suggest that it is important to account for investment rigidity when estimating breeding herd demand and hog supply. Simulation results indicate that the effects on breeding stock and hog supply of continued specialization in the hog production sector may not be as significant as what the hog production sector has experienced in the past decades. More importantly the simulations suggest that the impact of increasing investment rigidity is rather modest, about 3 percent at most and, thus, no policy intervention appears to be needed. However, the econometric results clearly indicate that estimates will be biased if investment rigidity is not explicitly accounted for when estimating breeding herd demand and hog supply.