This article is a brief summary of a lecture given by Professor Shapiro on 8 April 2003 at Victoria University in Wellington. Professor Shapiro was a visiting professor under the Professorial Fellowship (2) programme sponsored by the Reserve Bank.
In order to implement monetary policy appropriately, the Reserve Bank must continually update and evaluate its view of trend productivity growth, so that it can assess how much of New Zealand's economic growth in any period is attributable to productivity changes and how much is due to increases in factor inputs (such as labour and physical capital). This distinction is important because it enables the Reserve Bank to assess the extent to which economic growth over a given period may be placing pressure on the productive capacity of the economy, and therefore potentially creating inflationary pressures. All else being equal, the higher the growth in productivity in the economy, and the greater its contribution to total economic growth, the faster the economy can grow without placing excessive pressure on productive capacity and the lower the risk of inflation rising as a result of rapid growth in the economy.
However, assessing trend productivity for any given period, and over time, is fraught with difficulties. One source of that difficulty is the inadequacy of economic data, which restricts the ability to assess precisely the extent to which productivity growth has contributed to total economic growth over a given period. Moreover, there are difficulties in clearly determining the stage of the business cycle, and hence the cyclically-adjusted level of economic growth in any given period. This creates further obstacles to the accurate assessment of trend productivity growth. These difficulties create uncertainty about the extent to which the economy's growth over a given period may be placing pressure on resources in the economy and creating underlying inflationary pressures.
Professor Shapiro has studied the apparent acceleration of trend productivity growth in the United States in the 1990s. He has developed some techniques for better understanding and measuring the component of measured growth that is attributable to the business cycle, as opposed to being attributable to the trend. (3) Most attempts to measure productivity growth trends allow for cyclical factor utilisation, such as variations in overtime or capacity utilisation. These variations in factor utilisation can look like variations in productivity, but they are inherently related to the business cycle, and are not lasting. Accordingly, they can distort one's view of trend productivity growth.
On the other hand, there is another component that is not normally allowed for in attempts to differentiate trend from cycle. This concerns adjustment costs. Adjustment costs are incurred, for example, when businesses invest, which often happens in accelerated fashion late in the upswing of a business cycle. The investment diverts people and sometimes existing physical capital away from current production in order to put new productive capacity in place. During this phase, productivity appears to drop, since less current output is being produced by the same volume of inputs. But not only is this drop temporary, it results from the actions that are needed to increase trend productivity.
Conversely, during downswings, businesses tend to invest less, using proportionately more of their current resources to produce current output, which appears as increased productivity. Again, this is attributable to the cycle rather than the trend.
In the research described in his lecture, Professor Shapiro asked whether there was any evidence that trend productivity growth in New Zealand had also accelerated in the 1992 to 2002 period, and he explored the nature of the cyclical adjustments relevant to the measurement of trend productivity growth in New Zealand.
The key …