Labour
Statistics New Zealand produces data from two surveys that could be used to form industry labour input series: The Quarterly Employment Survey (QES) – a firm based survey – and Household Labour Force Survey (HLFS) – a private household based survey. In addition to other labour market data, the QES provides industry information on the number of paid hours, whereas the HLFS provides industry information on the number of hours worked.
HLFS hours worked data are the preferred measure of the labour input for two reasons. First, the HLFS hours worked series sets out to capture the number of actual hours worked. In contrast, 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. Second, the QES industry coverage is not as great as the HLFS industry coverage. The QES does not cover agriculture, hunting, and fishing industries. Furthermore, 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. Nonetheless, some researchers prefer to use the QES hours paid data as it does not involve human recollection error and because the industry data are considered to be more robust as the QES classifies firms into industries rather than relying on individual employee survey records. Appendix Table 2 summarises they key differences between the HLFS and QES.
| Characteristics | HLFS | QES |
|---|---|---|
| Timing | Every week in a year. | Pay week ending on or immediately before the 20th of the middle month. |
| Coverage | Individuals in private households. | Businesses with more than 2 FTEs and with turnover GST registered > $30,000 per year. |
| Period | Average over a quarter. | 1 week in a quarter. |
| Type of interviewing | Interview administrated. First is face to face, respondents are then telephoned from then on. | Form is posted. Respondent posts/faxes/telephones data through. |
| Seasonal | Students working in Feb but studying in March. | Captures information in the survey week only. |
| Age | 15+ | All ages |
| Weighting | Data is re-weighted every 5 years using Census data. | Data is weighted using Business Frame survey FTE info, updated every quarter since 1999. |
| Sample Exclusions | Excludes the following:
|
Excludes the following industries:
|
| Measures | Employment | Filled Jobs |
| Survey Size | 15,000 households (30,000 people approx). | 18,000 businesses at geo level. |
| Sources | There is one source of data - the survey. | Primary data is collected from the QES and secondary data is collected from the Business Frame. |
| Unit non-response | Weights up by age and sex to get a representative of the population. | Imputed for item & unit non-response. Weights not updated since 1989. |
| Non-respondents | Non-responding HHs are excluded from the survey results. | Imputed if key firm. |
| Questionnaire | ||
| Forms & Questions | Forms & questions have remained consistent since Dec 1985. | Forms & questions has remained consistent since taking over from DOL in 1989. |
| Survey Established | Dec 1985. | Since 1952 - Stats took over in Feb 1989. |
| Definitions/Classifications | ||
| Definitions/Classifications | Different definitions/classifications are used in both surveys | |
| Employment | A person may have 2 jobs (part time) but this counted as 1 employed person. | Measures filled jobs so a person with 2 jobs will count twice. |
| Place of employment | Primary industrial place. | Primary & secondary industrial place are recorded. |
| Who is employed? | Unpaid family helper are measured as employed i.e. in the labour force. | May not count unpaid family workers. |
| Casual/Voluntary | More likely to be measured. LF status - in the labour force. | Less likely to be measured. |
Source: Taken from a Statistics NZ table
Statistics New Zealand changed the HLFS industrial classification during the time period used in this paper. Hours worked data using the old New Zealand Standard of Industrial Classification (NZSIC) cover the period 1988 to 2002, whereas the HLFS ANZSIC hours worked data cover the period 1998 to 2002. To form comparable industry productivity data, ANZSIC hours worked data were backdated using the NZSIC hours worked data. Backdating was done by taking the level change in the industry NZSIC hours worked series and using this to rate back the corresponding industry ANZSIC hours worked data, beginning in 1998.
Industry labour cost data were derived by dividing adjusted industry compensation of employees by total hours worked. Because the System of National Accounts classifies the labour income of sole proprietors as operating surplus, an adjustment needs to be made to industry compensation of employees to more accurately reflect industry compensation for total hours worked. This adjustment was done by adding an estimate of sole proprietors’ labour income to industry compensation of employees from the income GDP accounts.
Industry compensation of employees data for the period 1988 to 1999 were drawn from Statistics New Zealand’s income GDP accounts. Industry compensation of employees data were extended to 2002 using the annual percentage changes in compensation of employees for the entire economy to rate forward the industry data. Estimates of industry sole proprietors’ labour income were derived by multiplying industry sole proprietors’ total hours worked by the private sector ordinary time wage rate. One digit industry level data on total hours worked by sole proprietors were obtained from Statistics New Zealand. The private sector ordinary time wage rate was sourced form the QES.
Capital
Industry capital stock data were sourced from Statistics New Zealand’s productive capital stock series. These industry capital stock series are available for the period 1988 to 1999. To complete the capital stock series up to 2002, the industry capital stock series were projected forward for the remainder of the period using annual percentage change in the economy wide capital stock for the period 2000 to 2002.
Industry capital cost data were constructed by taking the ratio of implicit operating surplus by the capital stock. The implicit operating surplus for each industry was calculated by subtracting compensation of employees from nominal GDP. This method assumes capital is the residual claimant in the revenue from production (net of intermediate material costs).
