DATA ISSUES IN TOTAL FACTOR PRODUCTIVITY BENCHMARKING: A CENTRAL EUROPEAN PERSPECTIVE

Total factor productivity is a method of measuring overall productivity of businesses, industries or economies. It is an approach which is recently becoming popular among government regulatory agencies when applying the so-called performance-based regulation. This principle induces regulated companies (for instance, electricity or natural gas distributors) to behave efficiently even if their industries are not exposed to competitive pressures, since they are rewarded for being more productive than other firms in the industry, and penalized in the opposite case. The aim of this article is to deal with the question what are the main data issues when comparing total factor productivity among firms or industries with focus on the Central European region, which is very heterogeneous in terms of geographic, social, economic and historic conditions. In the first part, we introduce total factor productivity and the most common methods of its measurement ˉ Malmquist and T ˉ rnqvist indexes. Consequently, we divide the data issues into separate categories and discuss them more in detail. The first category of issues is related to defining the set of comparable firms. Many factors, such as the degree of competition in the market, the extent of government regulation, economies of scale, firm size, geographical conditions and historic development have to be taken into consideration. The second category is associated with specifying the time period. TFP estimates should be based on long time series and the period should include the whole business cycle and be representative and exclude extraordinary events. The third group of issues is related to defining and measuring the inputs and outputs. Since a number of difficulties are associated with labor input measurement, it is often included into operating expenditures along with materials and services. The measurement of capital is even more contentious. The outputs should reflect performance, complexity and quality of service rendition. Finally, the fourth category of problems concerns determining the costs of inputs and outputs (cost-based indexes) or defining and measuring the reference technology (distance-based indexes), where some degree of arbitrary judgment and inaccuracy is inevitable. We conclude with the suggestion to avoid methods requiring large time series such as frontier-based methods of efficiency benchmarking (data envelopment analysis or econometric methods) or Malmquist TFP benchmarking in the Central European region due to huge differences between firms and their environment. However, the qu