3  Productivity, capital-intensity and labour quality in the aggregate market economy

Before going on to present sector-level productivity estimates, we first present an overview of our findings at aggregate market economy level. ‘Market sectors’ are here defined to include agriculture, forestry and fishing; mining and quarrying; manufacturing; electricity, gas and water supply; construction; wholesale, retail and other distribution activities; hotels and restaurants; transport, storage and communications; financial services; business services (excluding property services); and cultural and recreational services. They equate to what Statistics NZ (2006a) defines as the New Zealand ‘measured sector’ plus Business Services (ANZSIC Division LC). The excluded sectors are those which are dominated by public ownership (eg, public administration, education and health) and/or where no satisfactory measures of output exist in one or both countries (eg, property services, including residential buildings, and some personal, social and community services). A full list of market sectors with matching industrial classifications is shown in Table 1.

Using a market sectors PPP exchange rate derived by aggregating up from sector-level estimates (see Section 4.1), we estimate that in our chosen benchmark year (2002), average value added per hour worked in New Zealand market sectors was some 77% of the UK market sectors level, down from 82% in 1995 (Table 2). This differential may be compared against the estimate in Schreyer (2006, Table 1) that New Zealand GDP per hour worked in 2002 was roughly 69% of the UK level. The 8 pp difference between these two results is not large considering that (1) non-market sectors are excluded in our case and (2) our results are based on sector-specific PPP exchange rates rather than on the GDP PPP exchange rates used by Schreyer (2006).

Notes:   All estimates are for calendar years in contrast to the typical presentation of National Accounts data for years ending March 31st in New Zealand (see Appendix Section A1 for details of the assumptions made in order to convert March year data to a calendar year basis). 

2002 benchmark estimates of average value added per hour worked have been extrapolated back to 1995 and forward to 2004 on the basis of movements in constant price value added and labour inputs in each country. See Appendix Section A2 for a discussion of the advantages and disadvantages of this use of ‘constant PPPs’ as compared to an alternative ‘current PPPs’ approach. For details of the aggregate market sectors PPP exchange rate in 2002 see Table 4 below.

As will be shown in Section 5 below, average physical capital per hour worked in New Zealand market sectors was some 69% of the UK level in 2002 while average labour quality, on the measure used in this report, was an estimated 7% above the UK level.[4] ;When account is taken of these New Zealand-UK differences in relative capital-intensity and labour quality, then the average MFP level in New Zealand market sectors in 2002 was some 87% of UK levels, much the same as in 1995 (Table 2, Column 2). Between 2002-04 our estimates show a small relative decline in the New Zealand position on the ALP measure but not on MFP.

Using growth accounting methods described in Section 2.1, we can decompose the New Zealand-UK differences in relative ALP levels into three components:

1. The proportion explained by differences in relative physical capital-intensity.
2. The proportion explained by differences in relative labour quality.
3. The residual MFP component.

Columns 3-5 in Table 2 show that higher average physical capital per hour worked in the UK accounted for roughly 57% of the New Zealand-UK gap in ALP in 2002 and was partly offset by the higher average level of labour quality in New Zealand (-14%). The residual MFP contribution to the ALP gap was thus 57%. Trends over time in the relative capital-intensity and labour quality measures are shown in Figure 1 alongside the trends in relative ALP and MFP. While New Zealand MFP increased relative to the UK between 1999-2002, it is still no higher than in 1995 and in recent years appears to have been largely offset by declines in relative capital-intensity and a narrowing of the NZ lead over the UK in labour quality.

In comparing New Zealand with the UK it should be borne in mind that the UK is by no means a productivity leader among advanced industrial nations. Recent NIESR estimates for 2002 show that the UK lagged some 40 pp behind the US in terms of ALP, 32 pp behind France and 22 pp behind Germany. As shown in Figure 2, these gaps in ALP were associated in large part with lower physical capital-intensity in the UK compared to the other three countries and to a much lesser extent with gaps in relative labour quality.

Figure 1 - Relative ALP, MFP, physical capital per hour worked and labour quality, aggregate market sectors, New Zealand/UK, 1995-2004

Notes: For details of calculations underlying the relative capital-intensity and labour quality measures, see Sections 5.1 and 5.2 respectively.

Another perspective on recent trends in productivity growth in New Zealand and the UK is provided by decomposing average annual rates of ALP growth in aggregate market sectors, firstly, between growth in output and hours worked; and secondly, between the respective contributions of growth in physical capital, labour quality and MFP. Over the whole period from 1995-2004, average annual growth in output in the two countries was much the same. However, New Zealand experienced faster growth in labour inputs over this period. Although this is a positive outcome in terms of job creation, the net outcome was slower growth in ALP (averaging 2% per annum in New Zealand compared to 3% in the UK (Table 3). Average annual growth rates in MFP were the same in both countries over this period but the UK also benefited from faster growth of physical capital-intensity and a positive (albeit small) impact of growth in measured labour quality which narrowed the NZ lead on this measure between 1998-2004. The bulk of the growth in labour inputs in New Zealand occurred during the period 2000-04 when output growth was considerably faster than in the UK but productivity growth in the UK was similar to New Zealand due to a reduction in annual hours worked and continued growth in physical capital.

Figure 2 - Relative ALP, physical capital-intensity and labour quality in the UK, US, France and Germany, aggregate market sectors, 2002 (Index numbers: UK=100)

Notes: Derived from Mason, O’Leary, O’Mahony and Robinson (2006).

Some differences between this 4-country study and the present study need to be noted:

1. Total market sectors in the 4-country study includes some real estate activities (part of SIC 70) and social, community and personal services (UK SIC 90-91) which are not included in the definition of market sectors in the present study. In total these sectors account for just under 4% of total market sectors employment in the UK in the 4-country study.
2. Average annual hours worked per employee in the 4-country study were derived from GGDC (2005) for all four countries, including the UK. However, for the present UK-New Zealand comparison new estimates of UK hours worked have been derived from Labour Force Survey data in order to achieve greater comparability with New Zealand hours data. This makes little difference to estimates of relative ALP at total market sectors level.
3. Relative capital per hour worked in the four-country study is based on measures of three different types of capital asset: structures, plant and machinery and vehicles. For the present study we distinguish five different capital assets: structures, computers, other plant and machinery, vehicles and intangibles (principally software).

Note: All estimates are for calendar years in contrast to the typical presentation of National Accounts data for years ending March 31st in New Zealand. Hence the estimated growth rates for New Zealand are not directly comparable with those produced by Statistics New Zealand(see Appendix Section A1 for details of the assumptions made in order to convert March year data to a calendar year basis).