Efficiency analysis and benchmarking

Economic benchmarking integrates a group of methods used to compare the performance of one company to other companies, or alternatively to a theoretical business model. In doing so, regulators and companies can discover relative inefficiencies and potentials for improvements.

Regulatory authorities usually look for aggregated efficiency assessments that support the setting of improved efficiency targets (regulatory benchmarking). For the regulated companies, on the other hand, the assessment of total inefficiency is just a predecessor of a complex set of internal measures to improve performance. In order to define the measures, the regulated companies investigate the performance in their business areas and business processes (process benchmarking).

There are various mathematical techniques used to assess efficiency. Partial (one-dimensional) measures of performance (performance indicators) are the simplest forms of performing comparisons between different companies. Multi-factor (total) models are used to account for the relationships between different input and output factors.

These multi-factor models can apply parametric or non-parametric mathematical techniques. Non-parametric methods do not impose any functional form on the relationship between inputs and outputs. The most used non-parametric approach is Data Envelopment Analysis (DEA). Parametric methods impose a functional form on the frontier using estimation for production or cost functions. They require more knowledge about the production or cost functions and also about the distribution of errors. Parametric frontiers can be estimated by some variant of Ordinary Least Squares (OLS) or by Corrected Ordinary Least Squares (COLS). Stochastic Frontier Analysis (SFA) attempts to estimate an efficient cost frontier that does incorporate the possibility of measurement error or chance factors in the estimation of the efficient frontier.

We advise regulated companies and regulators worldwide on the conceptual properties and practical application of regulatory and process benchmarking models. In addition we support our clients to understand and interpret the results, and define internal measures to improve their performance. Our services cover the following areas:

  • definition of data requirements

  • data validation

  • model specification

  • determination of efficiency scores

  • interpretation of the benchmarking results

  • integration of efficiency scores in the price control

  • consultation support

  • definition of strategies for performance improvement.