Data Collection
Household Economic Survey (Expenditure) 2022/23
Methodology
The information collected as part of the Household Economic Survey (HES) 2022/2023 provides data on Household income and housing cost statistics as well as household economic statistics (expenditure)
Changes made to HES expenditure questionnaire in 2022/23
Some changes were made to the survey in HES 2022/23 compared with HES 2018/19, with the intention to ease respondent burden and increase data quality. Some of the changes may have potentially impacted our data:
- Partway through collection, phone interviewing was deployed as an option for completing the survey in a COVID-19 restricted environment. After the end of COVID-19 restrictions it was kept as an option and will continue to be utilised. To accommodate this some flow changes were added to the Household Questionnaire (HQ) and the Expenditure Questionnaire (EQ).
- The response options for the question on relationships in the household were changed from gendered terms to non-gendered terms. For example, “son or daughter” was changed to “child.”
- Two questions in the Material Wellbeing Questionnaire were altered to avoid confusion (qGoodshoes and qGoodbed).
- An improvement on a KiwiSaver question was added in the Contribution to Schemes module. Specifically, a question asking whether a certain scheme was a KiwiSaver scheme was left to the discretion of interviewers.
- Some clarification on pensions and government super was added in the Private Superannuation module. To ensure respondents understood the difference between these two, some flow changes were made. Prior releases have found confusion on the part of respondents in this area.
- For information on changes in the Mortgages module, please see Child poverty statistics: Year ended June 2023 – technical appendix.
- The Power and Telecommunication modules were altered to allow for bundling and were moved to immediately after the Mortgages module.
- The Household Operations module was moved before the Household Maintenance modules to avoid double counting and confusion.
- Improved clarity in the Introduction to the Household Maintenance module.
- Fuel types for vehicles are now collected in the Transport module. Clean car rebates are now able to be collected. Question on roadworthiness improved to incorporate planes and boats.
- Expenditure on cruises is now collected separately and feeds into the overall transport expenditure group.
- The Recreation and Technology module collected expenditure on BYOD in school and educational settings as part of an investigation into how widespread these costs are. Further incorporation of this material is not guaranteed for future collections.
- The flowchart of the Fees module has been changed to improve flow, no change to content.
- Change to the flow in the optometrist section in the Medical and Health Services module to improve flow, no change to content.
- Introduction to some parts of the Credit module altered for clarity, no change to content.
- Option for digital diary in PDF format.
COVID-19 impact on collection period
The expenditure data interval between the year ended June 2019 and the year ended June 2023 is a four-year gap, rather than the typical three-year gap of this series. The difference is due to the postponement of data collection for the year ended June 2022 Household Economic Survey in response to COVID-19 restrictions. Therefore, the change in the four years may be greater than for the typical three-year gap. Apply caution when comparing trends.
HES 2018/19 rebase
We have made use of the latest population estimates based on the 2018 Census.
As a result of this the number of households was revised downwards by 0.1 percent for the year ended June 2019, with the largest decrease in Auckland and the largest increase in the Bay of Plenty. Looked at in isolation, census rebasing led to minor changes in average household expenditures nationally. The 2018/19 figures used in the 2022/23 release are the revised figures.
Recall period
We collect HES data over the course of a year, from 1 July to 30 June. At the interview, the respondent is asked about their expenditure in the previous 12 months. For example, an interview in November 2022 would collect the household’s expenditure in the 12 months from November 2021. This means that households interviewed for the HES 2022/23 survey in 2022 will include some 2021 expenditure, while those interviewed in 2023 will include some 2022 expenditure. This is referred to as the recall period.
External influences
Changes in household expenditure may be influenced by real-world events, such as policy changes or fluctuations in the labour market. Events that could have influenced the HES 2022/23 data are:
- The expenditure data interval between the year ended June 2019 and the year ended June 2023 is a four-year gap, rather than the typical three-year gap of this series. The difference is due to the postponement of data collection for the year ended June 2022 Household Economic Survey in response to COVID-19 restrictions. Therefore, the change in the four years may be greater than for the typical three-year gap. Apply caution when comparing trends.
Increases in wages and benefits
- New Zealand Superannuation rate (gross) increasing for single living-alone from $538.24 to $578.67 between 01 April 2019 and 01 April 2023; single sharing from $495.10 to $532.43 between 01 April 2019 and 01 April 2023; and ‘both partners qualifying’ from $408.66 each to $439.79 between 01 April 2019 and 01 April 2023.
- minimum wage had two increases during year ended 30 June 2023, the first being from 1 April 2022 from $20.00 to $21.20, and the second from 1 April 2023 from $21.20 to $22.70. In our data the 2022 increase will be evident across the entirety of our data, while the 2023 increase will only have a partial impact.
- increases in the starting out and training wage from $16.00 to $16.96 from 1 April 2022 and then to $18.16 effective 1 April 2023.
Stats NZ data (CPI, LCI, and unemployment figures)
- the Consumer Price Index (CPI) annual inflation rates for June 2019 to June 2023 were:
o June 2019, 1.7 percent
o June 2020, 1.5 percent
o June 2021, 3.3 percent
o June 2022, 7.3 percent
o June 2023, 6.0 percent.
- The Labour Cost Index (LCI) annual changes between June 2019 and June 2023 were:
o June 2019, 2.1 percent, unadjusted 3.6 percent
o June 2020, 2.1 percent, unadjusted 3.1 percent
o June 2021, 2.1 percent, unadjusted 3.4 percent
o June 2022, 3.4 percent, unadjusted 5.1 percent
o June 2023, 4.3 percent, unadjusted 5.9 percent.
- The unemployment figures between June 2019 and June 2023 were:
o June 2019, 4.1 percent
o June 2020, 4.1 percent
o June 2021, 4.0 percent
o June 2022, 3.3 percent
o June 2023, 3.6 percent.
Official Cash Rate
- The Reserve Bank’s Monetary Policy Committee have been increasing the Official Cash Rate throughout the reference period. The remit of the Monetary Policy Committee is to keep inflation between 1% - 3%. The timeline for increases is as follows:
o 6th October 2021, increased from 0.25 to 0.5 percent (this was the first OCR increase from 0.25 since March 2020)
o 24th November 2021 increased to 0.75 percent
o 23rd February 2022 increased to 1.0 percent
o 13th April 2022 increased to 1.5 percent
o 25th May 2022 increased to 2.0 percent
o 13th July 2022 increased to 2.5 percent
o 17th August 2022, increased to 3.0 percent
o 5th October 2022, increased to 3.5 percent
o 23rd November 2022, increased to 4.25 percent
o 22nd February 2023, increased to 4.75 percent
o 5th April 2023, increased to 5.25 percent
o 24th of May 2023, increased to 5.5 percent.
Government initiatives and subsidies
- introduction of the cost-of-living package in the 2022 budget which included a cost-of-living payment of $350 that was paid to individuals aged 18 or over, earning $70,000 or less, who were New Zealand tax residents living in New Zealand and weren’t eligible for the Winter Energy Payment.
- introduction of transport-related taxes
o A 3.5-cent per litre nationwide increase in petrol tax was introduced on 30 September 2018.
- The transport relief package (effective from March 2022 until 30 June 2023) entailed the following measures which co-occurred (partly or completely) during the July 2022 – June 2023 survey cycle:
o From 1 April 2022 to 30 June 2023, public transport fares were reduced by 50 percent.
Response rate for HES expenditure 2022/23
For the overall Household Economic Survey, the target achieved sample rate is 70% which is 20,000 responding households out of the 28,500 households that are initially selected. However, for the 2022/2023 HES, the achieved target sample size was reduced from 20,000 down to 15,000 households.
The final achieved sample in the overall HES 2022/2023 survey included 14,100 households (overall achieved sample rate of 67 percent).
The selected sample size for HES expenditure which covered detailed household expenditure, including the 1-week diary, was maintained at 5,500 households with a target of 3,500 households.
The achieved sample rate for HES expenditure 2022/2023 was 61.5 percent (3,384 households) compared with a target of 63.6 percent (3,500 households). This was the third highest achieved sample rate for the current time series since 2007.
The achieved sample rates for HES expenditure 2007-2023 were:
o Year ended 30 June 2022/23 – 61.5 percent (3,384 households)
o Year ended 30 June 2018/19 – 71.7 percent (3,944 households)
o Year ended 30 June 2015/16 – 63.3 percent (3,499 households)
o Year ended 30 June 2012/13 – 54.6 percent (3,003 households)
o Year ended 30 June 2009/10 – 56.8 percent (3,126 households)
o Year ended 30 June 2006/07 – 52.8 percent (2,902 households).
The response rate (the measure of how many eligible households responded) was 68.7 percent in HES 2022/23.
Achieved sample rate compared with the response rate
The achieved sample rate is calculated as the number of eligible households that responded divided by the total number of dwellings sampled. Essentially, it tells you what percentage of the sample responded to the survey. Expressing the achieved sample as a rate controls for population growth.
| Eligible responding sample | ||
| ASR | = | ____________________ |
| Ineligible sample + eligible responding sample + eligible nonresponding sample |
The response rate is calculated as the number of eligible households that responded to the survey as a proportion of the estimated number of total eligible households in the sample.
| Eligible responding population | ||
| Response rate | = | ___________________ |
| Eligible responding population + eligible nonresponding population |
The achieved sample rate differs from the response rate because it includes the ineligible dwellings in the denominator. This difference means that the response rate is particularly sensitive to the classification of household eligibility. As a result, the achieved sample rate is more stable over time than the response rate. The impact of any bias arising from non-responses is minimised by non-response adjustment and the calibration to population benchmarks.
Imputation for HES expenditure 2022/23
Imputation replaces missing values with actual values from similar respondents.
Two imputation methods are used in HES – nearest neighbour donor imputation and median imputation (the latter for expenditure only).
The nearest neighbour donor imputation method replaces missing values by data values from another record called a donor. A donor is selected by finding a respondent with matching characteristics to the recipient. Median imputation uses the median of the acceptable values to replace a missing value.
We introduced donor imputation into HES in 2009/10, and now use it in all HES releases. We also applied imputation to every previous HES cycle and revised the data accordingly.
The donor imputation is applied to a household where the household does not supply all the required income or expenditure information but supplies sufficient information to be retained in the sample.
For individuals aged 15 years and over, we imputed income from investments and self-employment. We use income from wages & salaries and benefits from admin data, however we impute for eligible individuals (aged 15+) who did not link to the IDI. In addition, we impute age for respondents who do not provide an age.
We also impute local and regional council rates for respondents who have not provided enough information for us to calculate their rates. A form of manual imputation is used to impute interest rates. We also use donor imputation for all eligible individuals (aged 15+) who did not complete a diary but are in-scope for the diary, i.e., not away overseas.
A diary imputation methodology was developed for HES 2015/16, in which less than 3.5% of diaries required imputation, and imputation was only done for households for which a diary was already collected. For HES 2018/19 the imputation approach was changed to include all eligible individuals age 15+. In HES 2022/23, around 20% of the diaries were imputed.
Data quality
In late January and mid-February 2023, Aotearoa New Zealand was affected by torrential rain and tropical cyclone Gabrielle. The adverse weather and resulting flooding caused significant damage and disrupted our usual collection activities across the North Island. Face-to-face interviews were completely suspended in these two regions. These regions collectively accounted for 5% of Aotearoa New Zealand’s population. As the timeline for resuming interviews remained uncertain then, statistical investigations were conducted to assess the potential impact of excluding data from these regions on the national expenditure.
Based on the analysis undertaken, omitting data from these regions did not exhibit any substantial difference in total expenditure for Aotearoa New Zealand. Consequently, missing data from these regions had minimal effect on national-level estimates with our current estimation procedure. Interviews in these two regions resumed towards the end of March 2023.
Weighting process
Regular checks were implemented on the interim weights generated in our data base system as of December 2023. These weights were derived using HES 2022/23 benchmarks and estimation groups defined by the existing criteria. By implementing these checks, we aim to guarantee the integrity of the weighting process and provide a robust basis for the analysis of survey results.
It is crucial to conduct a thorough examination of the weights to ensure the accuracy and reliability of the results. We have checked the weights at the person-level and household-level. We can confirm that the sum of the selection weights at the person level is approximately 5 million and the sum of the selection weights at the household level is around 1.9 million. These are what we expect to get. The benchmarks have been met by ensuring that the sum of the final weights equals the benchmark total. There is no individual with negative weight.
Sample errors
We calculate sampling errors using the jackknife method. It is based on the variation between estimates of different subsamples taken from the whole sample. Sampling errors by income source and housing-cost type are provided for each published table. Customers should take care when interpreting income or housing-costs estimates with sampling errors greater than 20 percent – they are statistically less reliable than estimates with sampling errors less than or equal to 20 percent.
en-NZ
