Using Stochastic Frontier Analysis to Analyze Adjustment Costs and Investment Utilization

Publikation: KonferencebidragPaperForskning

Based on a theoretical microeconomic model, we develop an empirical framework for analyzing the size and the timing of adjustment costs and investment utilization. We show that adjustment costs and investment utilization result in technical inefficiency, because adjustments require the use of additional inputs and an initially incomplete investment utilization results in an output level that is temporarily not at its maximum. We estimate an output distance function as a stochastic "Efficiency Effects Frontier" model (Battese & Coelli 1995), where the estimated technical inefficiencies are explained with current and lagged investments, farm size, age of the farmer, and interaction terms between these variables. Furthermore, we derive the formula for calculating the marginal effects on technical efficiency for "Efficiency Effects Frontier" models so that we can calculate the (marginal) effect of current and past investments on technical efficiency, which we interpret as adjustment costs and temporary incomplete investment utilization. We apply this methodology to a large panel data set of Danish pig producers with 9,281 observations between 1996 and 2008. The results show that investments have a negative effect on farm efficiency in the year of the investment and the year after accruing from adjustment costs and incomplete investment utilization. In contrast, there is a large positive effect on efficiency two and three years after the investment. The farmer's age and the farm size significantly influence technical efficiency, as well as the effect of current and past investments on adjustment costs and investment utilization. These results are robust to different model specifications and different ways of measuring capital.
StatusUdgivet - 2012
BegivenhedAsia-Pacific Productivity Conference - Bangkok, Thailand
Varighed: 24 jul. 201227 jul. 2012


KonferenceAsia-Pacific Productivity Conference

ID: 41812591