Can Fiscal Policy Explain Technical Inefficiency of Privatised Firms? A Parametric and Non-Parametric Approach

Abstract

The massive interests of economic literature about the privatisation gave a notable impulse to the discussion about this theme in the pre and post privatisation firms performance. Basically in every case after privatisation the level of profit increases. Does this mean that privatisation is certainly able to increase efficiency? In this field a large part of the literature leave out the complex problem that public firms usually are subject to objectives and constraints that differently from private firms can affect the overall economic efficiency. Unfortunately many authors ignore the effects of taxation during the process of privatisation, but in real term there are significant tax issues that must be considered by public and private decision maker. In this paper we concentrate the attention on the efficiency measures with the purpose to identify and measure sources of successful performance that can be used in policy planning and allocation of resources. Several techniques to calculate these frontier functions have been used, some of them parametric, others non-parametric to empirically investigate the relationship between taxation on firm's income and efficiency in the period pre and post-privatisation. In this work we use both econometric and mathematical programming approaches for measuring efficiency. The econometric tool provide maximum likelihood estimates of a stochastic production and cost functions to distinguish noise from inefficiency. Instead, the mathematical programming approaches are non-stochastic and they do not make strict assumptions on the functional form of production and the statistical properties of the data. The general results obtained from the 3 different tools (Stochastic Frontier, Data Envelopment Analysis and Neural Network) are consistent. In fact, we see that privatization enhanced efficiency in three out of four sample firms.