PublicationWorking paper

A penalization approach for estimating inefficiency in stochastic frontier panel models

Abstract

Efficiency analysis is essential for evaluating the performance of entities that deliver essential or standardized services. The estimator proposed by Jondrow et al. (1982) is widely used in this context, but it has been criticized for several shortcomings: it tends to bias inefficiency estimates toward the mean, distorts the distribution, and misrepresents the conditional distribution of inefficiency—especially in cross-sectional data.

Zeebari et al. (2023) propose a regularization-based alternative that aligns sample and theoretical moments; however, this method is primarily designed for cross-sectional applications and does not extend naturally to panel data.

In response, this paper introduces a penalized mode estimator for unit inefficiency in panel data. The estimator accounts for heteroskedasticity in both inefficiency and idiosyncratic errors. A closed-form expression is derived, and Monte Carlo simulations demonstrate its superior performance compared to existing methods. An empirical application using data from electricity providers in Australia, Canada, and New Zealand highlights the practical advantages of the proposed approach.

Tchatoka, F. D., Söderberg, M., & Hakeem, M. A. (2025). A penalization approach for estimating inefficiency in stochastic frontier panel models. University of Adelaide, School of Economics and Public Policy Working Paper.

Details

Author: Tchatoka, F. D., Söderberg, M., Hakeem, M. A.
Publication year: 2025
Published in: University of Adelaide, School of Economics and Public Policy Working Paper.

Related

Professor
Magnus Söderberg
magnus.soderberg@ratio.se

Similar content

Article (with peer review)

Scale properties and efficient network structures in the Swedish electricity distribution market

Söderberg, M., Vesterberg, M.

Publication year

2025

Published in

Journal of Regulatory Economics

Abstract

This paper examines the Swedish electricity distribution sector to highlight three key findings. First, we identify significant economies of scale among electricity distribution firms, indicating that larger firms operate more efficiently. Second, we explore alternative market structures and demonstrate that these can substantially reduce the aggregated costs of electricity distribution. Third, we use novel survey data to show that firms perceive the economic incentives for mergers to be insufficient. These findings suggest that policymakers should consider creating a regulatory environment that encourages consolidation and enhance efficiency in the sector.

Article (with peer review)

Are CEOs judged on how cost efficient their firms are?

Månsson, K., Qasim, M., Söderberg, M.

Publication year

2025

Published in

Energy Economics

Abstract

This paper investigates whether executive boards consider firm-specific inefficiencies when they change CEOs in the Swedish electricity distribution sector. Firm-level inefficiencies are calculated using data from all Swedish electricity distributors from 2001 to 2022 and a data envelopment analysis (DEA) approach.

DEA has advantages over standard financial key performance indicators since it controls for heterogeneity in inputs and outputs. It is also frequently employed by energy regulators to calculate relative cost inefficiencies.

Our baseline approach uses a multilevel model and investigates the relationship between inefficiency and CEO between-effects. This analysis shows that 9–15 % of the variation in inefficiency can be attributed to the CEO effect.

The second modeling approach quantifies the CEO effect using a synthetic difference-in-differences approach, focusing on firms that have changed CEOs. The results reveal that new CEOs reduce cost inefficiency more when they succeed CEOs who were forced to leave.

Electricity prices and preferences for climate and energy policy

Bonev, P., Söderberg, M., & Vesterberg, M

Publication year

2025

Published in

Griffith University Working papers.

Abstract

We analyse the impact of high electricity prices on individual preferences and attitudes toward climate and energy policy in Sweden. Our identification strategy leverages key features of the data and study design. First, consumers are distributed across four electricity price areas, each characterized by varying and often divergent electricity prices. Second, surveys were conducted in four distinct waves. Notably, between the second and third waves, electricity prices were nearly identical across price areas, whereas after the third wave, significant price differences emerged. Using multiple estimators that exploit these features, we find that higher electricity prices: (1) reduce acceptance of a carbon tax, (2) increase support for nuclear power, (3) diminish concerns about climate change, and (4) have no significant impact on other political attitudes or food consumption preferences.