5.2  Labour

Accurate measurement of the labour input into the production process is essential if meaningful productivity statistics are to be produced. Information is required on both the quantity and price of labour. To measure the quantity of labour there are several possible measures. These include total employment (in an industry or economy), full-time equivalent employment and total hours worked.

It is important that in each case an adjustment is made to incorporate the labour contributions of self-employed individuals if they are not included in the data. Total employment measures the total number of people who are employed in the economic unit of interest and includes both full-time and part-time workers. Such measures will not accurately represent changes in the amount of labour used in the production process when the number of average hours worked changes or the mix of full- and part-time employment changes.

Full-time equivalent employment is an improvement over total employment as it incorporates changes in the mix of full- and part-time employment.[18] However, full-time equivalent measures will not reflect changes in average full-time employment hours. Total hours worked is the preferred labour quantity measure.[19] The choice of approach to measure the quantity of labour is important. Typically the different measures result in different growth rates in the amount of labour over time. The higher the measure of labour growth, the lower the productivity measure as more of the growth in output will be accounted for by the growth in the labour input.

The Household Labour Force Survey (HLFS), Quarterly Employment Survey (QES) and Economic Survey of Manufacturing (ESM) are different survey based sources of hours data produced by Statistics New Zealand. The QES is a firm based survey and provides information of the number of hours paid. The HLFS is a private household survey and provides information on the number of actual hours worked. The QES measure of hours paid does not exclude sick leave and holidays, and is unlikely to reflect increases/decrease in the number of hours worked by salaried workers during peak/slack periods. Furthermore, the QES industry coverage is not as great as the HLFS industry coverage. The QES does not cover agriculture, hunting, and fishing industries. The QES also excludes work without pay in the family business and causal employment (eg, cash jobs, cottage industries) and geographical units with less than one full time equivalent (FTE) worker are excluded from the QES hours.[20]

Sample surveys produced by Statistics New Zealand are redesigned periodically to ensure that the sample adequately reflects the contemporary composition of the population. Changes in sample design may result in some discontinuity in the statistical time series. The “Official” database provided to Diewert and Lawrence (1999) had HLFS and QES data from 1978. In Diewert and Lawrence (1999) the choice of which survey to use had an important influence on the measured growth of the labour input over time.

When hours worked is the quantity measure of the labour input, the corresponding price is the average hourly compensation received. To accurately reflect the cost of labour to the producer, it is important that the compensation data used includes all supplements to wages and salaries (for example, employer contributions to superannuation schemes). The OECD’s Productivity Manual notes several issues that relate to the measurement of labour compensation. This includes the allocation of mixed income.

“Mixed income” is the income that accrues to unincorporated enterprises owned by members of households, ie, the self-employed. It is necessary to determine an appropriate allocation of the operating surplus of proprietors and the self-employed into labour and capital components. There are two broad approaches to doing this. The first is to assume that the average hourly compensation of a self-employed person equals that of the average wage or salary earner. From this an imputed wage bill can be constructed and subtracted from the operating surplus, with the residual being assigned to the capital component. The second approach involves assuming a rate of return on capital to construct a cost of capital bill that can then be subtracted from the operating surplus with the residual being assigned to the labour component.

Up until now we have not taken into account the implications of heterogeneity of the labour inputs. In other words, no allowance has been made for differences in the level of human capital for different workers. When the hours worked by different workers are treated as homogenous, productivity measures are subject to measurement error. Where the average quality of labour is changing over time it is preferable to construct quality-adjusted measures of labour input. The quality of labour has generally been increasing over time.[21] The OECD Productivity Manual states:

When quality-adjusted measures of labour input are used in growth accounting instead of unadjusted hours worked, a larger share of output growth will be attributed to the factor ‘labour’ instead of the residual factor ‘productivity growth’. In other words, substituting quality-adjusted labour input measures for simple ones can shift the appreciation of the sources of growth, from externalities or spill-overs captured by the productivity residual to the effects of investment in human capital.

OECD, 2001: 46