Data Collection
Labour Market Statistics: June 2016 quarter
#Period-specific information
##Response Rates
| Survey | Reference period | Response rate | Achieved Sample rate |
|---|---|---|---|
| HLFS | Each week during the quarter (3 April 2016 – 2 July 2016) | Target: 90 percent Achieved: 80.1 percent |
Target: 76 percent Achieved: 70.9 percent |
| QES | The pay week ending on, or before, 20 May 2016 | Target: 89 percent Achieved: 89.9 percent |
N/A |
| LCI | Pay rates at 15 May 2016 | Target: 94 percent Achieved: 95.5 percent |
N/A |
See New quality measures for the Household Labour Force Survey for more information on the sample rate and response rates.
##HLFS ###Outliers During the seasonal adjustment process, X-13-ARIMA-SEATS can give less weight to the irregular component. Specifically, if the estimated irregular component at a point in time is sufficiently large compared with the standard deviation of the irregular component as a whole, then the irregular component at that point can be downweighted or removed completely and re-estimated. We refer to such observations as partial- and zero-outliers, respectively. In practice, the downweighting of outliers does little to seasonally adjusted data, but the impact of the outliers on the trend series will generally be reduced. However, if an outlier ceases to be an outlier as more data becomes available, then significant revisions to the trend series become possible.
The table below shows any partial (P) and zero (Z) outliers for the last year of each time series
| Outliers | ||||||
|---|---|---|---|---|---|---|
| Quarters | Male employed | Female employed | Male unemployed | Female unemployed | Male not in the labour force |
Female not in the labour force |
| Sep 2015 | P | |||||
| Dec 2015 | P | |||||
| Mar 2016 | ||||||
| Jun 2016 |
###HLFS pre- and post-calibration weight The following figure shows that while the distribution of the pre- and post-calibration weights differs within a quarter, the difference between the weights typically does not change from quarter to quarter.
The undercoverage rate indicates how representative the pre-calibrated sample is. The higher the undercoverage rate, the less representative the pre-calibrated sample.
Usually the undercoverage rate in the HLFS is around 20 percent. The overall undercoverage rate for the HLFS in the June 2016 quarter was 18.2 percent. This compares with 19.2 percent in the March 2016 quarter and 20.3 percent in the June 2015 quarter.
###Inclusion of armed forces in the Household Labour Force Survey
The target population of the Household Labour Force Survey (HLFS) was modified between March and June 2016. The following groups are now included in the target population, whereas previously they were excluded:
- Members of the New Zealand permanent armed forces
- Non-New Zealand diplomats and non-New Zealand members of their staff and households
- Members of non-New Zealand armed forces stationed in New Zealand and their dependents
As a result of the changes in the target population, individuals in the armed forces were included in the working age population (WAP) estimate. Of the changes to the target population, it was only the inclusion of armed forces that was reflected in the WAP estimate because this group was easily identifiable in the population estimates and was deemed to have a relatively large contribution to the WAP in contrast to the other groups. As a result of changes in the target population, armed forces members that reside in private dwellings are no longer out scope (i.e. they are eligible to be interviewed).
Due to the changes in WAP composition, the HLFS presents a more accurate estimate of labour market outcomes. The limitation of introducing this change is that the comparability of figures is reduced between previous quarters and the June 2016 quarter. Changes to the target population and scope rules for the HLFS were applied in order to achieve consistency with other household surveys. The primary reason for including armed forces in the target population was to achieve alignment with the International Labour Organisation’s (ILO) plans to promote coverage of armed forces members in national labour force surveys. It was of interest to explore whether the inclusion of New Zealand armed forces in the target population, and changes in scope rules, would influence changes in the labour market series between March and June 2016.
The inclusion of the armed forces in the target population does contribute to some of the changes in employment. Table 1 presents labour market estimates under three scenarios: 1) armed forces not included in the WAP and not interviewed, 2) included in the WAP and not interviewed and 3) included in WAP and interviewed (current approach). Under the first scenario, the estimated increase in employment over the quarter is 35,300 individuals. When armed forces members are part of the target population and survey population, the estimated increase in employment over the quarter is 45,300 individuals. Therefore, the combined effect of incorporating armed forces in the WAP estimate and interviewing those that are in private dwellings is estimated to be contributing to an inflow of roughly 10,000 individuals into employment.
In contrast, the impact of interviewing armed forces is relatively small on the changes in employment. Comparing the second (included in the WAP and not interviewed) and third (included in the WAP and interviewed) scenarios provides some insight into the changes in employment that are a result of including armed forces members (in private dwellings) within scope of the HLFS. Under the second scenario, the estimated increase in employment over the quarter is 42,500, which compares to the actual change of 45,300. Therefore, the change in scope rules is estimated to be contributing to an inflow of roughly 2,800 individuals into employment.
| | Labour Market Outcomes
(000s) | | | | | Rates | |
|---------------------------------------|--------------------------------|----------------|-------------------------------|--------------------------|------------------------------------|-----------------------|-------------------------|
| | Not in the labour force
(NILF) | Employed | Unemployed | Total Labour
Force | Working Age Population
(WAP) | Employment
rate | Unemployment
rate |
| Not included
in WAP & Not
interviewed | 1134.9 | 2444.3 | 126.4 | 2570.7 | 3705.6 | 66.0 | 4.9 |
| Included in
WAP & Not
interviewed | 1136.6 | 2451.5 | 126.9 | 2578.4 | 3715.0 | 66.0 | 4.9 |
| Included in
WAP &
Interviewed | 1134.4 | 2454.3 | 126.3 | 2580.6 | 3715.0 | 66.1 | 4.9 |
Table 1. Table presenting the labour force estimates derived by 1) not including armed forces in the WAP and not interviewing them, 2) including armed forces in the WAP but not interviewing them and 3) including armed forces in the WAP and interviewing them.
Customers are advised to interpret some of the change estimates between March and June 2016 with consideration for changes applied to the composition of the working age population and scope rules. That is, these changes may have a larger impact for the following lower level estimates:
- The change in estimates for Industry O (Public Administration and Safety) and O76 (Defence Force) because the armed forces are captured in these groups. All industry level estimates are susceptible to variations in the sample. That is, the HLFS is not designed to provide industry level estimates at any defined level of precision and as a result, could sample fewer or more armed forces personnel by chance. The variations for these industry estimates would be especially pronounced since individuals in this group are now in scope.
- The change in estimates for Manawatu-Wanganui because a number of sampling units selected in this region were in close proximity to army bases. The presence of armed forces does not appear to be spread out evenly across the country; and as a result the HLFS could by chance sample more or less people in this group.
###Intervention analysis for labour market outcomes There was a relatively large increase in the employed series between March and June 2016. This was in contradiction to expectations as the established seasonal pattern is a decrease between March and June. Work was carried out to explore whether questionnaire changes had resulted in a level shift of the total employed series, and whether an intervention to the seasonally adjusted employed series would be appropriate to improve the comparability and accuracy of figures over time.
While an upward level-shift is in line with the expected effects of introducing the new questionnaire, there is not yet enough information to accurately estimate the size of the effect, or decompose observed changes into real-world change and questionnaire effects. A review of seasonal adjustment diagnostics found that neither of the component series figures, used to derive total employed (male and female employed), were treated as outliers for the June 2016 quarter. Furthermore, establishing confidence in the magnitude of any possible level-shift in the series requires several observations where the observed figures are consistently larger or smaller than expected figures. With only one data point for the employed series under the new questionnaire, there is currently not enough evidence to estimate and apply a level shift for this quarter.
A forecast was applied to the male and female employed series in order to determine if these figures could be used to apply a level shift. This exercise revealed that there was a large amount of uncertainty in the forecast and as a result, a range of potential forecast estimates to be utilised in applying a level shift. Using these figures in evaluating the impact on the employed series found that neither the seasonally adjusted not trend estimates for this quarter were robust to variations in the forecast estimates. That is, the uncertainty in the forecast prevents Statistics New Zealand from applying an intervention that can improve the comparability and accuracy of the seasonally adjusted employed series at this point in time.
More quarter’s data is required to 1) determine if there is a level shift due to the new questionnaire and 2) accurately estimate it. All key labour market series will continue to be monitored in order to identify possible shifts in the series and determine when it would be appropriate to confidently apply an intervention that could improve the comparability and accuracy of seasonally adjusted estimates.
###Estimates of employment by industry
In the June 2016 quarter there were a higher than usual number of employed people classified to the ‘not specified’ industry group. A large proportion of this not specified group will be certain self-employed people from whom we did not collect the information required to assign an industry.
The new HLFS was designed to be better at identifying an individual’s employment status. This included better identification of self-employed people versus paid employees. However, there are scenarios where those who identify themselves as self-employed later in the questionnaire are not then asked to provide their main activity (which is used to determine the industry worked in). There was no means of assigning an industry group to these respondents in the June quarter, but additional changes will be made to our survey to resolve this issue in future.
Based on the occupation information that we do have on these respondents, this issue is spread across a range of different industries, not concentrated within a particular industry.
##LCI ###The impact of the minimum wage change
The adult minimum wage increased from $14.75 an hour to $15.25 an hour (3.4 percent increase) on 1 April 2016. For the June 2016 quarter, 14 percent of all surveyed salary and ordinary time wage rates increased – 2 percent of rates increased due to the minimum wage increase. If the wages that increase to the new minimum wage had not changed, the LCI including overtime would have still increased by 1.5 percent. However, the effects can be seen within some industry and skill-level breakdowns.
In the year to the June 2016 quarter, retail trade and accommodation (industry group GH) increased 2.1 percent. If we had processed the increases due to the minimum wage increase in the LCI as no change, then in the June 2016 quarter, retail trade and accommodation (industry group GH) would have increased 1.7 percent.
In the year to the June 2016 quarter, skill level 5 increased 2.1 percent. This level includes occupations that require a New Zealand Register level 1 qualification, no qualification, or a short period of on-the-job training (eg clerical and administrative workers, labourers, sales workers). If the increases due to the minimum wage increase were treated as unchanged, wage rates for skill level 5 occupations would have risen 1.7 percent for the year.
##General information and methodology
For general information and methodology on the specific surveys within the labour market statistics release, please see the following Datainfo+ pages:
en-NZ
