1  Introduction

This paper provides a review of conceptual and methodological issues in measuring productivity. Attention is given to the concept of productivity and the relationship between productivity and technological change. The paper also surveys four approaches to measuring productivity growth and reports productivity estimates for the New Zealand economy. The availability of appropriate input and output data is essential for the accurate measurement of productivity and therefore this paper also discusses some important data issues that influence productivity measurement.

Economists and policy makers are interested in productivity measurement for several reasons. First, although TFP does not measure technological change, it is related to technological change to the extent that it measures free lunches associated with technological progress.[1] Second, even though productivity measures may imperfectly measure the free lunches associated with technological change, they still exhibit persistent and systematic time series properties. These properties point to relationships that are not yet independently measured in the growth accounting methodology and therefore suggest that current understanding of what drives economic growth requires refinement. Third, productivity numbers provide information about how much measured output is being produced in an economy relative to measured inputs. This is an indication of how the economy is performing in terms of its productive efficiency with respect to available resources. Fourth, growth accounting, growth theory and productivity measures form an intellectual paradigm that has engaged a vast number of researchers who are trying understand the growth process (Hulten, 2000). Productivity analysis has been a core of this research agenda and discovering what it does not measure has been as important as determining what it does measure.

Taken together with other indicators of economic activity (including independent measures on R&D and technological investment and diffusion), economists and policy makers can use productivity measures to develop an idea of how well or poorly an economy is doing in terms of the return it is obtaining on its investments. Uniform measures taken across time, and to the extent possible across production units, can be used to examine whether productivity is rising or falling. These numbers can then be compared with other measures of economic performance, including per capita output growth, and direct measures of technological change to determine what are the correlations between measured productivity, output growth and technological change.

What is probably of most importance in terms of the research agenda on productivity is an answer to Prescott’s (1998) call for a theory of productivity. Working out the causal linkages from fundamentals such as technological change through to measures such as GDP per capita growth and productivity change will provide economists and policy makers with a clearer picture of how productivity measures relate to underlying economic change and growth. To this end, research into productivity and its measurement are essential.

The paper is organised as follows. Section 2 discusses the concept of productivity and its relationship to technological change. Section 3 surveys four different approaches to productivity measurement. Historical productivity estimates for the New Zealand economy are reported in Section 4. Given data issues that complicate the construction of accurate productivity series, Section 5 discusses data issues in productivity measurement and research. Section 6 concludes.