The Review of Economics and Statistics Vol. 51 No. 2 1969
Aggregation, Index Numbers and the Measurement of Technical Change
Abstract
M OST of the problems of measurements of productivity are related in one way or another to problems of aggregation. This statement also covers the measurement of technical change. There are few, if any, conceptual problems in measuring a production process which consists of a single factor and a single product. As soon as more factors or more products are included, complications arise. Some of these are examined in this paper. In the absence of technical change (TC) the main finding is our claim that the well-known concept of index number bias has no empirical validity and is therefore irrelevant in answering some of the questions it has been supposed to answer. In introducing TC, a natural definition of neutral TC (NTC) is suggested so that changes in output can be allocated to NTC, differential TC (DTC), and changes in inputs. It is shown how in general DTC is measured as NTC. An important role in our discussion is played by aggregates derived from the underlying production function. To illustrate the use of such aggregates a multiproduct production function is partly estimated. We start this paper by presenting the empirical production function which serves as illustration in subsequent discussion. In section III we take up the question of index number bias. In section IV we define the two forms of technical change and deal with the effects of NTC and changes in inputs on output. In section V the measurement of DTC is discussed.
- DOI
- 10.2307/1926726
- Volume
- 51
- Issue
- 2
- Pages
- 166
- Sources
- openalex crossref