4 New Zealand’s historical productivity record
This section briefly summarises the results obtained in a number of studies that have attempted to measure productivity growth rates for New Zealand. Figure 4 summarises the growth rates found in studies that included measures of TFP growth, while Figure 5 summarises labour productivity growth rates. This section does not attempt to discuss the robustness of results obtained in any of the studies, rather it illustrates that a number of studies have attempted to measure productivity growth in New Zealand, albeit over different timeframes and by using a number of different measurement techniques. Diewert and Lawrence (1999) chapter 6 reviews more comprehensively a number of the studies featuring in Figures 4 and 5.
In Figures 4 and 5, the timeframe to which the average productivity growth relates is shown by the placement and length of the horizontal bars. The number in the centre of the bar is the average annual growth rate for that period.
Razzak (2002) used an OLS method that allows for either a deterministic or stochastic trend to produce a number of TFP estimates. Razzak argued that TFP measurement in New Zealand may be affected by several factors. The first factor is whether or not there exists a structural break in the data used. Second, is the nature of the trend. Third is whether the production function exhibits constant or increasing returns to scale.
McLellan (2001) used a growth accounting approach to measure TFP and labour productivity over the classical business cycle using a peak-to-peak approach.[16] This approach attempts to measure trend output growth by removing the cyclical component of growth. TFP was calculated using a Cobb-Douglas production function with labour’s share set equal to 0.65. The data used were quarterly production GDP, Labour full time equivalents (FTEs) from the Household Labour Force Survey (HLFS), with a weighting of 0.5 on part-time employment, and a capital stock measure calculated using Statistics New Zealand’s productive capital stock data.
The Diewert and Lawrence (1999) preferred TFP series (labelled DL Preferred Series in Figure 4) was constructed using data in the Diewert and Lawrence database. A total output index was formed by aggregating 49 output components and a total input index was formed by aggregating 13 input components, both using a chained Fisher index. All outputs and inputs are expressed in producers’ prices. Both indices relate to the market sector. Their database covered the period 1972 to 1998 (March years). Their database drew on labour times series presented in the OECD’s Economic Outlook and Census based hours worked per person by occupation data from Statistics New Zealand. Their (HLFS Hours) TFP series on the other hand used data derived from the HLFS rather than the Census.
TFP numbers using the “Official” database were also constructed by Diewert and Lawrence. The database was based on Statistics New Zealand (SNZ) data and covered the period 1978 to 1998 (and consequently numbers were based on the System of National Accounts 1968 data). The official database included data on production GDP, labour and capital for 20 separate market sector industries. Chained Fisher TFP series were constructed using industry GDP data, industry hours worked data (drawn primarily from the HLFS, Quarterly Employment Survey (QES) and the Economic Survey of Manufacturing (ESM)), and industry net capital stocks for plant and equipment and building and construction, weighted by user costs with industry specific depreciation rates and asset lives based on Philpott (1992).[17]
