Purpose and summary
Guide to unemployment statistics provides guidance to our data customers on the different features of various unemployment measures. It will help them choose the correct unemployment measure to suit their needs.
This third edition reflects changes introduced after we redeveloped the Household Labour Force Survey (HLFS) in 2016. Specifically, the introduction of ‘underutilisation’ indicators, and a subsequent move away from reporting on ‘jobless’ people. It also acknowledges more recent changes to specific benefits around the work obligations of benefit recipients.
This guide contrasts the HLFS unemployment measure (‘persons unemployed’) with other administrative data sourced from the Ministry of Social Development, particularly counts of people receiving Jobseeker Support – Work Ready and the now obsolete job seekers register. Each of these indicators has a different purpose, and different definitions, timings, and qualities.
This guide gives an overview of each indicator, what it measures, and how each can best be used.
- The HLFS unemployment measure is best used when an official (and internationally comparable) measure of unemployment in New Zealand is required.
- The other measures are more useful when looking at specific aspects of unemployment.
A key message throughout this guide is that each unemployment indicator is different, conceptually and methodologically. They are not simply different ways of measuring the same thing.
Readers should also note that unemployment is only one indicator of the state of the labour market. We encourage customers to look at other indicators available to researchers to get a more holistic view of labour supply, labour demand, and wage pressures.
The guide also gives an overview of a related measure now produced by the HLFS – underutilisation. This is a more expanded measure of people’s unmet needs for work, including the unemployed, underemployed, and people who are considered part of the ‘potential labour force’ but who do not meet certain criteria to be officially unemployed.
Introducing the unemployment measures
Many labour market indicators can provide an understanding of unemployment in New Zealand. This guide explores measures produced by Stats NZ and the Ministry of Social Development (MSD).
Stats NZ’s Household Labour Force Survey (HLFS) provides New Zealand’s official unemployment measure. In addition, the HLFS ‘underutilisation’ series measures a wider group of people with an unmet need for work but who do not fit the specific criteria required to be counted as unemployed. This also includes people who are considered employed but specifically ‘underemployed’.
Occasionally, commentators suggest looking at alternative indicators of the health of the labour market, such as benefit data produced by MSD. ‘Jobseeker Support – Work Ready’ numbers are currently the most comparable to HLFS unemployment. However, this guide shows that ‘comparability’ comes with several caveats.
Commentators sometimes refer to the job seekers register. This was used historically as a measure of unemployment, before the introduction of the HLFS, but it has limited use as a current labour market indicator.
Each measure of unemployment has a different purpose and definition.
Household Labour Force Survey
HLFS statistics help customers understand labour market trends across time. Before 1985, the only quarterly measure of unemployment in New Zealand was derived from the job seekers register, which was maintained by the Department of Labour. However, this measure had shortcomings, which led to the HLFS being developed.
The HLFS, a sample survey, was specifically designed to provide a more consistent and accurate measurement of employment and unemployment trends in the labour market. Additionally, by following the International Labour Organisation (ILO) standards we ensured international comparability.
In 2016, the HLFS underwent a major redevelopment. One change was to introduce questions that allow us to measure underutilisation, in line with ILO standards.
The HLFS ‘Persons unemployed in the labour force’ series (referred to in this guide as ‘HLFS unemployment’) is the official measure of unemployment in New Zealand.
To count as unemployed in the HLFS, a person must be:
- part of the working-age population: the usually resident, non-institutionalised population of New Zealand aged 15 years and over
- without a job – neither in paid work nor unpaid work in a family business
- both actively seeking and available for paid work, or have a new job to start within the next four weeks.
People can be actively seeking work in many ways. Examples include contacting or applying in person to an employer, placing an advertisement to find a job, or taking steps to set up your own business.
However, if someone only looks at job advertisements (and takes no other action) they do not qualify as ‘actively seeking’.
The HLFS also produces a suite of series under the umbrella ‘underutilisation’. This measure is much broader than HLFS unemployment – we include it in this guide because it covers individuals who meet some of the criteria of being unemployed, and because they can be considered part of the ‘potential labour supply’.
In addition to the unemployed, underutilised people include:
- The underemployed – people who are working part-time (less than 30 hours a week) who would like to work more hours, and are available to do so
- Unavailable jobseekers – people actively seeking work but who were not available to start work in the reference week but would become available within the next four weeks
- Available potential jobseekers – people not actively seeking work but who would like a paid job and were available in the reference week.
The latter two groups collectively make up the potential labour force, sometimes referred to as people who are ‘marginally attached’ to the labour force.
Before the June 2016 quarter, the HLFS produced a measure of the jobless. This series counted the official HLFS unemployed along with people who were actively seeking work but unavailable, and people available to work but not actively seeking it. From the June 2016 quarter, we discontinued the jobless series and replaced it with the more relevant and internationally comparable underutilisation series.
Jobseeker Support – and other benefits
Unemployment benefit numbers, produced by MSD, were historically used as a measure of unemployment. However, MSD no longer produces these figures following major changes to benefit categories.
In July 2013, the unemployment benefit transferred to Jobseeker Support as part of the government’s programme of welfare reform.
Jobseeker Support is designed to provide short-term financial assistance to people who are looking for work, training for work, or temporarily unable to work due to a health condition or disability.
An individual’s eligibility for Jobseeker Support is affected by their eligibility for other benefits, and the economic activity of other household members. For example, people with a long-term health or disability could be eligible for the Supported Living Payment, whereas sole parents caring for children under 14 years could be eligible for Sole Parent Support. Additionally, people living with partners who earn above a specified income could be ineligible, whether or not they are seeking work.
‘Jobseeker Support – Work Ready’ is a subset of the Jobseeker Support benefit. The number on the work ready benefit is most appropriately used as a measure of people actively seeking full-time work who are unable to support themselves financially.
Jobseeker Support – Work Ready compares best with the previous unemployment benefit and with current HLFS unemployment. However, note that it includes some people who previously received other benefit types – as a result, the number of recipients is higher than the number previously on the unemployment benefit. Changes in work obligations for other benefit types are also relevant to making comparisons with HLFS unemployment.
See Guide to unemployment statistics (2014) for more information on the modifications to benefit types following the July 2013 welfare reform.
How work obligations are relevant when using benefits as an unemployment indicator
While this guide focuses on Jobseeker Support – Work Ready as the measure most comparable with previous unemployment benefit numbers, and the HLFS unemployed, we have caveats around what should be considered ‘comparable’. Customers need to note the 2016 changes to the work obligations of benefit recipients before comparing benefit data with official unemployment estimates.
While Jobseeker Support – Work Ready figures exclude a group of jobseekers that may be exempt from finding full-time employment (Jobseeker Support – Health Condition or Disability), they do not capture all those on benefits who have active work obligations. People receiving Sole Parent Support who have dependent children aged 3–13 years have part-time work obligations, as do some Jobseeker Support – Health Condition or Disability recipients. (See figure 1.)
Changes to benefit names (2013), including work obligations (2016)
Both full-time and part-time work obligations are relevant when comparing with HLFS unemployment, as to be counted as ‘employed’ a person need only be working one or more hours a week.
Following this logic, the full-time work obligations for Jobseeker Support – Work Ready recipients, and part-time obligations for Sole Parent Support recipients, would create the group that is most comparable to the HLFS unemployment measures.
We recognise that not all this group fits the official unemployment definition. Both benefit groups will have some recipients who are currently working part-time and therefore are not officially unemployed (although they may be underemployed). Some may also not be actively seeking work, as specifically defined in the HLFS. However, this combined approach better reflects the more active management of work obligations, especially for the Sole Parent Support group, and relates to the HLFS unemployment, underemployed, and underutilisation measures.
Note that this advice is for the primary benefit recipient rather than for partners who may also have work obligations, but are not counted in MSD’s official reports (therefore they are an undercount). Also, not all unemployed people are eligible for a benefit and not all people who are eligible will apply.
Job seekers register
Historically, MSD maintained the register of unemployment – which was the official measure of unemployment – as the ‘job seekers register’. Stats NZ previously released figures from the register but these are no longer produced.
The job seekers register is no longer used for reporting on unemployment. Policy changes over the last 10 years mean the numbers are not comparable with the present time.
Differences between measures
The HLFS unemployment and underutilisation statistics are quarterly averages. Jobseeker Support – Work Ready figures are counts of people receiving this assistance at the end of the month. This timing difference is particularly important when there are substantial seasonal rises in unemployment; for example, in the December quarter when more students enter the labour force.
It is not always clear what people mean when they talk about ‘unemployment’. The following examples illustrate how a person may fit into the different measures of unemployment:
- Mr Brown works part time. He earns very little and is looking for full-time employment. This qualifies him for Jobseeker Support, but he is considered employed in the HLFS. However, we may consider him underemployed if he is available and willing to work more hours.
- Ms Grey is not in paid work, but is available and is actively seeking work by contacting various employers. Her spouse is employed and receives income above a specified amount, so Ms Grey is not eligible for Jobseeker Support. She is considered unemployed in the HLFS.
- John Smith is a 17-year-old school student. He is younger than 18 years so is not eligible for Jobseeker Support. He is actively looking for a part-time job in the school holidays. Because he is not available at the moment, we do not consider him unemployed in the HLFS. However, he is counted as an unavailable potential jobseeker (and is ‘underutilised’).
Figure 2 shows the relationship between HLFS unemployment and benefit recipients.
The definitional characteristics section (below) provides a more comprehensive summary of the differences between the two HLFS unemployment and Jobseeker Support – Work Ready measures.
People who are only in figure 2’s left section are represented in HLFS unemployment statistics; people only in the right section are counted in Jobseeker Support and other benefit types. People in the middle section are represented in both measures.
People in the right section may feature in the underutilisation measures, including as underemployed and as unavailable potential jobseekers.
Because of timing and definitional differences, HLFS unemployment numbers and the numbers receiving Jobseeker Support – Work Ready or other benefits with work obligations do not always parallel one another.
Figure 3 shows a time series of the two unemployment measures. Note how the measures generally move in a similar direction, but have different levels.
Below we list the differences between the HLFS unemployed series, and Jobseeker Support – Work Ready recipients.
Definitional characteristics for HLFS unemployment measures
Age: 15 years or over
Residency: ‘usually resident’
Work sought: one hour or more a week
Current work: less than one hour a week (in paid job or own business), and with no unpaid work in a relative’s business.
Income: not relevant
Partner’s employment status: not relevant
In tertiary study: can count as unemployed
Effort to find work: must have actively sought work (eg contacted private employment agency, written to employer, contacted friend about a job) within last four weeks
Availability for work: must be available within next four weeks
Wants income assistance from Work & Income: not relevant
Timing: quarterly average.
Definitional characteristics for Jobseeker Support – Work Ready recipients
Age: 18–64 years (generally)
Residency: lived continuously in New Zealand two years or more (as citizen/permanent resident)
Work sought: full-time employment
Current work: may work part time, subject to relevant income tests
Income: ineligible if greater than certain threshold
Partner’s employment status: makes some ineligible (through excess income)
In tertiary study: ineligible
Effort to find work: complies with job seeker agreement, work test, and any other Work & Income administrative requirements
Availability for work: may be unavailable for short periods
Wants income assistance from Work & Income: must apply and meet eligibility criteria
Timing: admin count at month’s end.
Estimates of HLFS unemployment and underutilisation statistics are based on a sample survey. A typical margin of sampling error around these estimates is plus or minus 5 percent. As Jobseeker Support numbers are administrative records, they are not subject to sampling errors.
We seasonally adjust the number of people unemployed to minimise the seasonal effects on the data. Doing this reveals the underlying trend and allows for quarter-to-quarter comparisons. Underutilisation numbers are currently not seasonally adjusted as not all the underlying components follow a seasonal pattern.
Time-series data based on Jobseeker Support – Work Ready recipient numbers are not seasonally adjusted. This means the series are more suited for annual comparisons, rather than month-to-month comparisons.
While we seasonally adjust the total number of people unemployed in the HLFS, it is not feasible to do this for every breakdown of unemployed people that the HLFS can provide. We only adjust the number of unemployed males, females, and the total. Because breakdowns by age, ethnicity, and regional council area are not seasonally adjusted, they are only suitable for annual comparisons.
The HLFS labour force estimates provide a measure of the economically active population. They are based on the principle that a person’s labour force status should be determined on the basis of their activities during a specified reference period. In the HLFS, this reference period is short (one week) to ensure the measure accurately reflects the labour supply situation at a specified moment in time.
Any estimate produced by Stats NZ that is fewer than 1,000 people is suppressed, for quality reasons. The sampling errors of these estimates are too great for most practical purposes, so we do not release them. This limits the level of detail we release on some breakdowns (eg by certain ethnicities and regions).
For confidentiality reasons, MSD suppresses data when the number of people in any breakdown below national level is five or fewer. This may affect the breakdowns of benefit numbers that MSD provides.
Changes in social welfare policy can also change benefit recipient numbers. For example, changing the eligibility criteria for a benefit may affect the number of people receiving it. This can cause inconsistencies in the time series, which is always a risk in using an administrative data source to track something over time.
Choosing between measures
HLFS unemployment statistics are used when the official measure of unemployment is required. Unadjusted numbers, broken down by age, sex, ethnicity, and region (and combinations of these), are available. The HLFS unemployment measure is particularly good for making comparisons across demographic characteristics.
Only limited demographic breakdowns are released by MSD for Jobseeker Support – Work Ready recipients.
Unlike the MSD measures, HLFS unemployment is available as a seasonally adjusted series. This is useful for customers wanting to compare data quarter-to-quarter, rather than annually. Additionally, HLFS unemployment figures should be used when comparing data internationally – the survey was designed to the ILO’s internationally agreed standards.
The HLFS measure is used as a key indicator of the health of the labour market by stakeholders, policy makers, media, politicians, and economists.
The HLFS underutilised statistics both include and complement the unemployment figures, by providing a broader picture of the unmet need for work. They are particularly relevant in times of economic downturn, where unemployment may not necessarily rise but underemployment does, or people may be discouraged from seeking work at all.
Jobseeker Support – Work Ready numbers are useful for information on people who are seeking full-time work and unable to support themselves while doing so. Sole Parent Support (with part-time work obligations) numbers complement the ‘work-ready’ numbers to get a broader count of benefit recipients who are required to seek work.
While the job seekers register was used historically as the official measure of unemployment, it is no longer produced. Its use as a measure of unemployment is limited. Policy changes over the last 10 years mean the numbers are not comparable with the present time.
Information available from three measures (below) outlines the information (by topic) available from some measures discussed in this guide. Data from these measures is readily accessible on the Internet. Additional benefit information may be available on request through the MSD website.
We separate the official HLFS unemployment series into seasonally adjusted and unadjusted data, to illustrate which breakdowns are available from the HLFS. The unadjusted series tend to be more detailed, but are only suitable for annual comparisons.
Information available from three measures
We can assess the different information each measure provides by topic.
Age: total all ages (15+ years) if seasonally adjusted (SA); all ages from 15 years if unadjusted (UA)
Sex: male, female, total for both SA and UA
Ethnicity: total (SA); European, Māori, Pacific people, Asian, Middle Eastern/Latin American/African (MELAA), other, total (UA)
Region: New Zealand (SA); 12 regional council areas (UA)
Time series: quarter-to-quarter and year-to-year comparisons from 1986 (SA); year-to-year comparisons from 1986 (UA)
International: comparable with other ILO- and OECD-subscribing countries (SA); not comparable (UA).
Age: all ages from 15 years
Sex: male, female, total
Ethnicity: European, Maori, Pacific people, Asian, MELAA, other, total
Region: 12 regional council areas
Time series: year-to-year comparisons from 2004 (use caution before June 2016 when survey changed)
International: consistent with ILO standards.
Jobseeker Support – Work Ready recipients
Age: four age groups – 18–24 years, 25–39, 40–54, 55–64
Sex: male, female, total
Ethnicity: NZ European, Māori, Pacific people, other, total
Region: New Zealand
Time series: year-to-year comparisons from 2008
International: not comparable.
Unemployment as an indicator of the labour market
This guide focuses mostly on different measures of unemployment, and briefly mentions related indicators such as underemployment and underutilisation. It shows that unemployment is a complex topic and that it is important to understand the underlying concepts behind each measure.
Customers should also be aware that unemployment is just one indicator of how the labour market is developing. Commentators often refer to whether the labour market is getting ‘tighter’, and a downward trend in the unemployment rate is often seen as being a key indicator of this. However, we advise customers to seek to understand a range of information sources when interpreting the state of the labour market.
Stats NZ’s Labour Market Statistics release is a useful information source. This quarterly release brings together many indicators. In addition to unemployment and underutilisation, it includes statistics on job growth across different industries, hours worked, average earnings, and wage inflation.
Our aim in bringing all this information into one release is to tell a coherent ‘story’ about the labour market.
This guide discussed the following labour market measures:
- HLFS unemployment – the official measure of unemployment – provides a comprehensive understanding of labour market trends across time. This measure is the key indicator of the state of the labour market.
- The number of underutilized people is useful as a broad labour market indicator and for understanding the unmet need for work
- Jobseeker Support – Work Ready recipient figures are generally comparable with previous unemployment benefit numbers. Also considering Sole Parent Support (with part-time work obligations) helps to measure the number of people whose employment and financial situations meet the criteria for government assistance.
- The number of registered job seekers is no longer produced. Historically, it had limited use as a labour market indicator.
Ministry of Social Development (2014). Benefit fact sheets. Retrieved from www.msd.govt.nz.
Statistics New Zealand (2014). A guide to unemployment statistics (second edition). Retrieved from www.stats.govt.nz.
Statistics New Zealand (nd). Labour Market Statistics information releases. Retrieved from www.stats.govt.nz.
Statistics New Zealand (2016). Introducing underutilisation in the labour market. Retrieved from www.stats.govt.nz.
Work and Income (nd). Jobseeker Support. Retrieved from www.workandincome.govt.nz.
ISBN 978-0-908350-97-1 (online)
Published 15 March 2017