Estimation of long-run inefficiency levels: a dynamic frontier approach

Abstract

Cornwell, Schmidt, and Sickles (1990) and Kumbhakar (1990), among others, developed stochastic frontier production models which allow firm specific inefficiency levels to change over time. These studies assumed arbitrary restrictions on the short-run dynamics of efficiency levels which have little theoretical justification. Further, the models are inappropriate for estimation of long-run efficiencies. We consider estimation of an alternative frontier model in which firmspecific technical inefficiency levels are autoregressive. This model is particularly useful to examine a potential dynamic link between technical innovations and production inefficiency levels. We apply our methodology to a panel of US airlines.

Estimating labor force joiners and leavers using a heterogeneity augmented two-tier stochastic frontier

“…17 Polachek, Das and Apiroam-Thamma (2015) show the importance of individual-specific heterogeneity by illustrating inherent biases in estimating population-wide persistence of permanent and transitory shocks. Studies that examine persistent and transitory effects within the context of frontier estimation models are Ahn and Sickles (2000), Colombi (2010), Tsionas and Kumbhakar (2014), and Filippini and Greene (2016). 2015. This computation first entails estimating bias corrected ̂,̂ and ̂ from the 1977-2012 data.…”

Model specification, estimation and inference in studying economic output

“…We observe a high persistence in the adjustment of Portugal's efficiency while it is not the case in the non-dynamic model. A high persistence in the adjustment of efficiencies is a feature that many researchers often wish to capture the dynamic pattern of efficiencies (Ahn and Sickles (2000), Tsionas (2006)). A rationale behind this is that efficiencies might take time to be adjusted due to the potentially large adjustment costs associated with the inputs, or persistent change in management skills.…”

“…Although the studies might provide approximation for the dynamic of inefficiencies, the specifications are arbitrary approximations with little theoretical justifications. Many researchers aim to provide some economic interpretations regarding the dynamic patterns using an AR(1) process (see Ahn and Sickles (2000), Emvalomatis et al (2011) and Tsionas (2006)). Ahn and Sickles (2000) propose an AR(1) of an technical inefficiency as follows u it = ρu it−1 + ζ it , where 0 < ρ < 1 measures a firm's ability to adjust its past period inefficiency level.…”

“…Many researchers aim to provide some economic interpretations regarding the dynamic patterns using an AR(1) process (see Ahn and Sickles (2000), Emvalomatis et al (2011) and Tsionas (2006)). Ahn and Sickles (2000) propose an AR(1) of an technical inefficiency as follows u it = ρu it−1 + ζ it , where 0 < ρ < 1 measures a firm's ability to adjust its past period inefficiency level. Ahn and Sickles (2000) and Emvalomatis et al (2011) models are not designed for incorporating the variables affecting inefficiency, while Tsionas (2006) does.…”

“…Ahn and Sickles (2000) propose an AR(1) of an technical inefficiency as follows u it = ρu it−1 + ζ it , where 0 < ρ < 1 measures a firm's ability to adjust its past period inefficiency level. Ahn and Sickles (2000) and Emvalomatis et al (2011) models are not designed for incorporating the variables affecting inefficiency, while Tsionas (2006) does. Our model specification differs from Tsionas's (2006) in the sense that the base inefficiencies are not identical across DMUs, and the effects of environmental variables are not time-invariant and the same across DMUs.…”