Changes in the way New Zealanders are currently working won’t be immediately reflected in the labour market statistics we produce, Stats NZ said today.

The COVID-19 lockdown resulted in an uncertain future for some industries, widespread wage subsidies, and many people working from home.

Labour market statistics: March 2020 quarter, due to be published on 6 May 2020, will largely reflect the state of the labour market before the five-week alert level 4 lockdown, imposed to stem the spread of COVID-19.

However, we expect Labour market statistics: June 2020 quarter (due to be published 5 August 2020) will show more clearly how the COVID-19 lockdown has changed New Zealand’s labour market. A wide range of indicators, such as underemployment, hours actually worked, and hours usually worked, will provide a clearer picture of how restrictions to stem the spread of COVID-19 are affecting New Zealand’s workplaces and workforce, alongside the unemployment rate.

“Some people may be getting a wage subsidy, but are not actually working because firms have been forced to shut temporarily,” labour market and household statistics senior manager Sean Broughton said.

“Others in essential industries may have been working much longer hours than they typically would.”

“It will likely take some time to measure the full effects of the COVID-19 lockdown. The March 2020 quarter will mostly show how the labour market looked before COVID-19 hit.”

Data collection slowed by COVID-19

As well as affecting work in New Zealand, the COVID-19 lockdown has limited our ability to collect survey data about the labour market, especially in the final weeks of the March 2020 quarter.

New Zealand’s COVID-19 alert level was lifted to level 3 on 23 March 2020. It was raised to level 4 from 11:59pm on 25 March 2020 and remained at this level until 11:59pm on 27 April 2020. Widespread workplace restrictions to slow the spread of COVID-19 remain in place at the time of publishing. These government actions affect both the ability of many people to work and our ability to collect data about work.

The quarterly labour market statistics release combines data from three surveys: the household labour force survey (HLFS), the quarterly employment survey (QES), and the labour cost index (LCI).

The March 2020 quarter surveys have been partially limited by measures taken to slow the spread of COVID-19. Further effects on these surveys are anticipated for the June 2020 quarter (we plan to publish more detail on the impact of COVID-19 on our statistics later in May 2020).

March 2020 quarter data from household labour force survey interrupted

The HLFS provides a picture of New Zealand's labour force from a household perspective. These statistics relate to employment, unemployment, underutilisation, and people not in the labour force.

Estimates from the HLFS are based on a representative sample of 15,000 households and about 30,000 individuals throughout New Zealand. When a household is selected for the sample, they are first interviewed face-to-face at their place of residence, then interviews in subsequent quarters are carried out by telephone where possible. In a typical quarter, about a quarter of interviews are carried out face-to-face.

• The HLFS estimates for the March 2020 quarter will not indicate major changes following the introduction of steps to slow the spread of COVID-19 for two main reasons:
  The HLFS data is collected over 13 weeks, each quarter. This means estimates are designed to reflect the state of the labour market over the full quarter, not at a single point in time. For the March 2020 quarter, the majority of the quarter had played out prior to the introduction of steps to slow the spread of COVID-19.
• The interrupted data collection (outlined below) means that Stats NZ will not have fully captured activities during the period in which New Zealand shifted from alert level 2 to alert levels 3 and 4.

On 20 March 2020, we suspended face-to-face survey collection, following advice from the Government Chief Science Advisor and the Ministry of Health. We continued telephone-based interviews where possible.

From 26 March 2020, we had to reduce the amount of telephone interviewing, as staff in our contact centre were redeployed to urgent work answering public calls about the Government’s COVID-19 website.

Achieved sample rate for March 2020 quarter household labour force survey

The achieved sample rate for the HLFS in the March 2020 quarter was 70.5 percent, which is below Stats NZ’s target of 76 percent. Responses were also spread unevenly over the quarter, notably declining in the later weeks, when the government lockdown came into force (see figure 1). While these rates are below target, the data is still considered fit for purpose and reliable. In comparison, the achieved sample rate for the March 2019 quarter was 74.7 percent, and the achieved sample rate for the December 2019 quarter was 77.4 percent. For further comparison, in the few years prior to 2019 response rates ranged between 70 and 74 percent.

Figure 1

Response rate for March 2020 quarter quarterly employment survey lower than target

Statistics from the QES for the March 2020 quarter reflect conditions in the pay-weeks ending on, or immediately before, 20 February 2020. Based on the survey period, these statistics may indicate early impacts as COVID-19 spread internationally. They will not reflect the lockdown period at the end of March 2020.

The response rate for the QES in the March 2020 quarter was 83.8 percent. While this is below Stats NZ’s target response rate of 89 percent, the data is still considered fit for purpose and reliable. In comparison, the response rate for the March 2019 quarter was 86.6 percent, and the response rate for the December 2019 quarter was 86.0 percent.

Labour cost index rates for March 2020 quarter captured before lockdown periods

LCI wage and salary rates for the March 2020 quarter were captured as at 15 February 2020. Based on the survey period for the LCI, Stats NZ does not expect to see the effects of steps to reduce the impact of COVID-19 reflected in the March 2020 quarter. Target response rates were met for the LCI in the March 2020 quarter.

June 2020 quarter responses likely to reflect COVID-19 lockdown actions

We expect to see the impact of COVID-19 lockdown actions reflected in the number of responses to the following surveys in the June 2020 quarter.

• Household labour force survey: face-to-face interviewing remains suspended until further notice.
• Quarterly economic survey and Labour cost index, which collect data from businesses: some of the businesses we select to survey in April 2020 may close, temporarily or indefinitely, before the survey reference periods in mid-May 2020. Additionally, many businesses that are operating will be doing so differently than usual, so it may be more difficult to contact them, or for them to respond.

Addressing challenges presented by COVID-19 lockdown

We have and will take the following actions to mitigate potential impacts on data quality caused by the COVID-19 lockdown restrictions.

Overcoming challenges answering household labour force survey questions

We acknowledge that respondents may find it difficult to answer some labour market survey questions, for example questions about their availability to work, working fewer hours, being away from work, being on paid leave, and reporting of wages and salaries.

If a respondent is unsure how to report their usual hours, our survey interviewer will guide them to answer based on the hours they worked prior to lockdown, and record their actual hours, which are the hours they worked in the reference week.

As another example, if a respondent is not working any hours and is receiving the wage subsidy from their employer, then they have a paid job. The survey interviewer will record them as employed but away from work during the reference week.

Changes to the survey sample

Additionally, to maintain a sufficient sample size for the HLFS, we have retained respondents who were scheduled to exit the survey after the March 2020 quarter survey.

Ensuring reliable data

We are preparing for a decrease in responses from households and businesses, especially from the industries that are heavily affected by COVID-19, such as hospitality and tourism. As with any quarter, we will ensure the statistics are as reliable as possible and be transparent about response rates and the ways we may need to fill any data gaps.

Other useful statistics about the labour market

For further insight into the labour market during this period of uncertainty due to the COVID-19 lockdown, the following options are available:

• Stats NZ’s Monthly Employment Indicator series; these indicators use a combination of data from two different Inland Revenue sources: the employer monthly schedule and payday filing. The filled jobs and gross earning indicator series are published four-to-five weeks after the end of the reference month.
• The Ministry of Social Development publishes Statistics on benefits and the wage subsidy assistance. Please note that these statistics are not directly comparable with Stats NZ’s labour market statistics. Figure 2 shows the differences between household labour force survey unemployment measures and people included in jobseeker support. See Guide to unemployment statistics (third edition) for more information.

Figure 2

Text alternative for Comparing Stats NZ’s household labour force survey unemployment measures (left) and Ministry of Social Development’s jobseeker support (right)

The diagram shows the overlap between Stats NZ’s unemployment measure and Ministry of Social Development’s jobseekers support measure.

People captured by both measures are benefit recipients aged 18–64 years available for and seeking full time or part time work.

Additional people captured only by Stats NZ’s unemployment data are 15–17-year-olds and people 65 years old and over, people who are not employed and are available for and seeking part time work, people available for and seeking work who are ineligible for a benefit, and people with family or personal income sufficient to support them while looking for work.

Additional people captured only by jobseekers support are benefit recipients seeking full-time or part-time work but unavailable for a short period of time, benefit recipients working part-time, and benefit recipients not working or seeking work.

© Scoop Media