Assessing Economic Performance among North American Manufacturing Establishments, 1870/71: Data, Methodology and Measurement Issues

A number of conceptually robust and empirically practical approaches are available to assess relative economic performance among producers who operate on either side of an international border. In this paper we discuss the impact that data compilation, methodological choice, and variable definitions may have on the quantitative and qualitative assessment of cross-border performance comparisons. As an illustrative example we use manuscript census data from 1870/71 to compare total factor productivity (TFP) among a sample of manufacturing establishments located along the Canada-US border. We briefly discuss issues associated with the preparation of manuscript census data for the measurement of cross-border TFP differentials and the establishment of industry selection criteria. We also review TFP measurement techniques, such as growth accounting calculations, cost and production function index number approaches, and econometric estimation. However, the central focus of the paper is an investigation of the impact that variable definitions have on our assessment of TFP performance. In particular, we probe the relationship between the size of cross-border TFP differentials and the reliance on a variety of common definitions for labour, capital, output, input weights, and prices.