Tracking the stepwise effects of regulatory reforms over time: a ‘back-door’ approach

Abstract

Most of the literature on the efficiency effects of regulatory reforms ignores the gradual nature of the implementation of these reforms. This article describes a new simple data manipulation that minimizes data requirements to measure econometrically the impact on efficiency of each stage of multi-stages regulatory reforms. The approach can easily be implemented using standard stochastic frontier software programs. It is illustrated with data on the Spanish port system, which went through a two-stage reform during the 1990s.

Notes

¹ There are some papers that illustrate the effects in regulatory changes (Estache et al., 2002; Anstine, 2004) or the efficiency or cost structure in regulated industries (Grosskopf et al ., 2006; Jara-Díaz et al ., forthcoming).

² For a more detailed survey about temporal variation in inefficiency see Cuesta (2001).

³ At least three software programs allow the estimation of the Battese and Coelli's model (1988): FRONTIER (Coelli, 1996), LIMDEP and STATA.

⁴ For more details see González (2004).

⁵ This methodology has increasingly common in empirical papers on infrastructure industries, e.g. (Coelli and Perelman, 1999, 2000; or on regulated industries, e.g. Morrison et al ., 2000, among others).

⁶ As stated by Cuesta and Orea (2002) the chosen output does not influence the results.

⁷ For more details about the data and variables see Gonzalez (2004) and Gonzalez and Trujillo (2005).

⁸ See Coelli et al . (2003).