7  Summary and assessment

This study has confirmed the prevailing consensus that, in aggregate, New Zealand market sectors compare unfavourably with the UK on average labour productivity (ALP) - and by implication compare even more unfavourably with other countries such as the US. At total market sectors level the proximate causes for this relatively weak performance on ALP in New Zealand are lower multi-factor productivity (MFP) closely followed by lower physical capital-intensity.

However, beneath this overall story there is considerable sectoral variation. In 2002 there were five sectors in which New Zealand had an estimated lead on ALP of 5 pp or more over the UK: food, drink and tobacco, accommodation, restaurants and bars, communication services, finance and insurance, and cultural and recreational services. In addition, metal products is just ahead of the UK on ALP and there are two sectors -- mining and business services - where New Zealand is behind the UK on ALP but ahead on MFP. This suggests that, although New Zealand firms in these two sectors are on average less capital-intensive than their UK counterparts, many of them make more efficient use of existing resources. 

The same is probably true of agriculture as well but due to the uncertainties about UK agricultural employment numbers which we have discussed in this report, we are unable to present a satisfactory New Zealand/UK comparison of ALP in agriculture.

This leaves 12 out of 21 sectors where New Zealand falls short of the UK on both ALP and MFP. These sectors comprise all other manufacturing apart from metal products, electricity, gas and water supply, construction, wholesale, retail and transport services. They represented just under 58% of total hours worked in New Zealand market sectors in 2002. With the exception of wood products, electricity, gas and water and construction, these sectors are all relatively low in physical capital-intensity compared to the UK.

With some reservations we have made use of a measure of relative labour quality which incorporates wage differentials between qualification groups (assumed to proxy productivity differences) as well as differences in the mix of certified qualifications in each country. On this measure New Zealand appears to be ahead of the UK in workforce skills in all sectors except for mining. However, the New Zealand lead has declined since 1998, partly due to increased numbers of UK workers holding certified qualifications and partly due to declining returns to sub-graduate qualifications in New Zealand as compared to degree-level qualifications. In growth accounting terms, the measured advantage in New Zealand skills does serve to partly offset UK advantages in physical-capital intensity and MFP in many sectors. However, the extent of the offset is fairly small, partly because the New Zealand skills lead itself is small and partly because in growth accounting the respective contributions of different production inputs are evaluated separately and thus do not take account of potential complementarities between inputs (such as the contribution of workforce skills to the effective selection and utilisation of capital equipment).

It is worth considering how much of the New Zealand-UK gap in ALP at aggregate market sectors level is due to differences in industrial structure. For example, if New Zealand employment tends to be concentrated in sectors with relatively low absolute levels of ALP (typically less capital-intensive sectors), this might help to explain the overall weakness in ALP relative to the UK. To examine this we follow van Ark et al (2002) in using a shift-share method which decomposes the UK-New Zealand ALP gap into two components with the UK as the base country:

(9)    

where LP refers to the average labour productivity level in US$ terms and  S_i  refers to the employment share of industry i in each country. If ALP levels are the same in each country, then the first term on the right-hand side of Equation 9 is zero and the ALP gap is entirely due to differences in employment structure. Conversely, if employment shares are the same in each country, then the second term is zero and the ALP gap is solely attributable to productivity differences between the two countries at sector level.

The results shown in Table 13 suggest that roughly a quarter of the New Zealand-UK gap in ALP for aggregate market sectors is attributable to differences in employment structure such as the relatively high shares of UK employment in sectors with comparatively high value added per employee such as financial and communication services (even though New Zealand is actually ahead on ALP in those two sectors). The other side of this coin is the relatively high concentration of New Zealand employment in comparatively low value added sectors such as agriculture.

However, the remaining three quarters of the ALP gap are accounted for by within-sector productivity differences. To understand the sources of these differences, further research needs to go beyond growth accounting and investigate the factors underlying the proximate causes of the ALP gap which are emphasised in the growth accounting framework, namely, differences in accumulated stocks of physical and human capital and in MFP. For example, how do the cost and incentive structures confronting businesses making investment decisions in New Zealand compare with those in other countries? How well placed are New Zealand businesses to benefit from international knowledge spillovers which might help improve MFP performance through faster innovation and improved efficiency of resource utilisation? What are the effects on incentive structures and innovation performance of New Zealand-specific characteristics such as small population size, low population density and distance from large markets?