This chapter discusses common methodological, theoretical and empirical choices that scholars face when undertaking productive and economic efficiency analyses. After summarizing the main results of duality theory in Section 2, we outline in Section 3 the most popular empirical methods available to undertake efficiency analyses, namely nonparametric data envelopment analysis (DEA) and parametric stochastic frontier analysis (SFA). We discuss in Section 4 several strategies aimed at reducing the dimensionality of the analysis, either by relying on dimension reduction techniques that aggregate the original variables into a smaller set of composites, or by selecting those that better characterize production and economic processes. Section 5 discusses how to control for environmental or contextual z-variables that do not fall within managerial discretion, as well as the implications that each option has for researchers, managers and policy makers. Section 6 presents a series of recent models addressing endogeneity issues in the DEA and SFA approaches. In this section, we also discuss the endogenous nature of the distance function when assessing firms’ efficiency. Finally, Section 7 summarizes the guiding principles of the chapter and draws the main conclusions.

Performance, Productivity, Efficiency, Types and numbers of variables, Endogeneity
Semiparametric and Nonparametric Methods (jel C14), Classification Methods • Cluster Analysis • Principal Components • Factor Models (jel C38)
Department of Technology and Operations Management

Orea, L., & Zofio Prieto, J.L. (2019). Common methodological choices in non-parametric and parametric analyses of firms’ performance. In Palgrave Handbook of Economic Performance Analysis. doi:10.1007/978-3-030-23727-1_12