Improving the econometric precision of regulatory models

Regulatory regimes often attempt to introduce quasi-competitive pressures by undertaking comparative efficiency assessments between the regulated companies and setting company-specific cost reduction targets based on those comparisons. The UK water industry is one example of such a regime—indeed, it has emphasized the importance of maintaining the number of independent companies in order to preserve the robustness of the modeling. For example, in 2007, the Competition Commission considered whether the merger between Mid Kent Water and South East Water might prejudice the ability of the regulator (Ofwat) to make comparisons across water companies for the purposes of assessing performance and setting price controls. In this paper, we examine this issue and provide specific recommendations to regulators. Our cross-sectional results show that the impact of this merger is not significant. We demonstrate that joint estimation of all the sub-models using the ‘seemingly unrelated regression’ (SUR) procedure in a cross-section and/or panel data framework can dramatically improve the accuracy of the modeling. Moreover, the merger does not affect the confidence intervals significantly under such approaches, which still remain far superior to those under Ofwat’s cross-sectional approach. Based on these results, we recommend that Ofwat and other regulators adopt SUR and/or panel data analysis and thereby reduce their reliance on having sufficient numbers of independent companies.     

Joint estimation of technology choice and technical efficiency: an application to organic and conventional dairy farming

This paper proposes an econometric framework for joint estimation of technology and technology choice/adoption decision. The procedure takes into account the endogeneity of technology choice, which is likely to depend on inefficiency. Similarly, output from each technology depends on inefficiency. The effect of the dual role of inefficiency is estimated using a single-step maximum likelihood method. The proposed model is applied to a sample of conventional and organic dairy farms in Finland. The main findings are: the conventional technology is more productive, ceteris paribus; organic farms are, on average, less efficient technically than conventional farms; both efficiency and subsidy are found to be driving forces behind adoption of organic technology.     

Closed-skew normality in stochastic frontiers with individual effects and long/short-run efficiency

This paper considers the estimation of Kumbhakar et al. (J Prod Anal. doi:10.1007/s11123-012-0303-1, 2012) (KLH) four random components stochastic frontier (SF) model using MLE techniques. We derive the log-likelihood function of the model using results from the closed-skew normal distribution. Our Monte Carlo analysis shows that MLE is more efficient and less biased than the multi-step KLH estimator. Moreover, we obtain closed-form expressions for the posterior expected values of the random effects, used to estimate short-run and long-run (in)efficiency as well as random-firm effects. The model is general enough to nest most of the currently used panel SF models; hence, its appropriateness can be tested. This is exemplified by analyzing empirical results from three different applications.     

Management in production: from unobserved to observed

Are productivity estimates good proxies for unobserved management? And, does management affect production in a neutral and monotonic fashion as assumed by these proxies? We use Bloom and Van Reenen’s management data to show that two popular proxies, fixed effects and inefficiency scores, correlate with observed management practices. We find that the correlations are positive but weak. Also, management explains only a fraction of the proxies’ variances. The data rejects the assumptions of neutrality and monotonicity. Last, our results suggest that management has characteristics both of a technology and an input.     

Nonparametric estimation of the determinants of inefficiency

We consider the benchmark stochastic frontier model where inefficiency is directly influenced by observable determinants. In this setting, we estimate the stochastic frontier and the conditional mean of inefficiency without imposing any distributional assumptions. To do so we cast this model in the partly linear regression framework for the conditional mean. We provide a test of correct parametric specification of the scaling function. An empirical example is also provided to illustrate the practical value of the methods described here.     

Estimation of productivity and markups with price dispersion: Evidence from Chinese manufacturing during economic transition

We estimate firm productivity and markups in the presence of heterogeneous output and input prices. We assume that firms are monopolistically competitive in the output market and that state firms are favored and pay lower prices for inputs due to political connections. The proxy-variable (or called the control-function-based) approach of structurally identifying production functions and productivity is customized to our new setting. Our approach solves the underidentification problem inherent in the standard proxy-variable method. We investigate the Chinese transportation equipment industry and find evidence against perfect competition and price homogeneity. We also find that productivity and markup differentials and dynamics are consistent with the market-oriented reforms in China.     

Parametric Approaches to Productivity Measurement: A Comparison among Alternative Models

This paper deals with modeling total factor productivity (TFP) growth in a flexible manner using panel data. Several competing parametric models are used to explore whether there are any similarities in the estimates of TFP growth and technical change among these models. Using a primal approach, we decompose TFP growth into different components. The models are then used to measure productivity and technical change in the Swedish cement industry. In general, the results are found to be model dependent and often conflicting, although much less so for returns to scale and overall productivity growth.
JEL classification: O 30; C 33