Household Economic Survey (Income) 2019/20

Information on New Zealand households’ income, housing costs, and material well-being is based on data collected as part of the Household Economic Survey 2019/20. This survey was in the field from 01 July 2019 till 24 March 2020 (instead of the scheduled 30 June 2020) when it was called off due to COVID19. ###Changes made to HES questionnaire in 2019/20 The following are some of the key changes made in the way we collected data in HES 2019/20 compared with previous HES years: Questions on gender and sexual identity questions were asked the first time in HES and were included in the demographics module. The gender question asked people to describe their gender – whether male, female, or another gender, for example, non-binary. Combined with a question about sex at birth, this information helped to ensure that the transgender population is better reflected in the data. This two-step approach is recognised as international best practice. A self-complete questionnaire was introduced for the first time. Some demographics questions such as on gender and sexual identity were asked in the self-complete module of the questionnaire instead of being interviewer administered. This was done in view of the sensitivity around the gender and sexual identity questions. Questions on disability were added for all members of the household 2 years and over. The Household questionnaire was shortened and some questions were moved to other modules in the Income Questionnaire. Some questions were removed from the current Jobs and previous jobs module as this information is now available from admin data. Some modules are no longer required and have been removed as topics are covered in admin data: Earnings Compensation, IRD Income, Work & Income Income, Hobbies ((now captured in the Other Regular Income module). Reference and collection 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 income in the previous 12 months. For example, an interview in November 2019 would collect the household’s income and wellbeing in the 12 months from November 2018. This means that households interviewed for the HES1920 survey in 2019 will include some 2018 income, while those interviewed in 2020 will include some 2019 income. Data input Assignment of cases is centralised in Salesforce, a system that allows for real-time observation of response rates. The team of interviewers use BLAISE to conduct household surveys. BLAISE is a computer assisted interviewing (CAI) software that guides the interviewer through the correct sequence of questions. They are displayed one at a time and have automatic routing built in to make sure that respondents are only asked questions that are relevant to them. Once submitted, the data is stored and a response is logged in our centralised tracking software, Salesforce. This allows for a real-time overview of progress in reaching the target response rates for demographic groups. The data is then fed through various editing stages, before being loaded into the processing database, EPIC. ###Admin data is used to replace survey data Despite best efforts to obtain accurate income data from respondents, survey data on income will always be subject to some uncertainty. This is due to respondents not being able to remember or not disclosing all sources of income over the year to the interviewer. Respondents may also provide ‘rough estimates’ of amounts or amounts that are exclusive of taxes paid. In some cases, family members may not know the income of all other family members. We know that salary and wages can be overstated when compared with admin data, usually due to respondents forgetting changes in income over the year. Benefit income is often understated, due to failure to recall small periods of benefit receipt through the year. The IDI is a large research database that holds microdata about people and households. The IDI contains full tax data related to individuals, including data provided by employers for each employee (the employee monthly schedule), self-employment income, and some investment income. Data from the Ministry of Social Development includes benefits paid, including working for families’ tax credits, and accommodation supplement. For the 2018/19 HES, we asked respondents their income but have used the admin data when calculating our statistics. For the 2019/20 HES and ongoing, respondents have not been asked to provide their income amounts for income variables we are able to obtain from admin data. Some income sources are not currently covered by the data in the IDI. These include investment income, some sources of irregular income, and non-taxable income. These income variables continue to be collected directly from respondents. While most salary and wage income is provided on an individual’s pay day and flows through into the IDI on a quarterly basis, other income (for example, self-employment income) relies on individuals providing their tax returns, which can be delayed before being included in the IDI. Due to this timeliness issue, we use self-employment income provided to us by the respondent. Linking admin data target The use of admin data in HES requires linking individuals in HES to the IDI spine, a dataset to which all datasets in the IDI are linked to. This link to the IDI uses address, address history, name, and date of birth. A high link rate is needed to ensure the best quality data. The link rate of the overall HES sample in September 2020 to the IDI is 95 percent (with a false positive rate of 1.7 percent). The link rate for children is lower than for adults because date of birth is not collected for children (although age is). This is not expected to affect the estimation of child poverty measures as we rely on the HES data to tell us about the presence of children in households. All income (including benefits) is allocated to the adults in the household. Linking to the IDI enables all in-scope and eligible individuals aged 15+ from responding households to have admin income data, i.e., wages and salaries, and benefits, whether they responded to the survey or not. This increases the number of usable responses in the dataset. Records which were not linked to the IDI had income imputed to reduce the potential bias from these records. External influences Changes in income and housing costs may be influenced by one-off real-world events. Events that could have influenced the HES 2019/20 data are: As a result of the country-wide lockdown effected from mid March 2020 due to COVID-19, field interviewing for HES was stopped in the 3rd week of March 2020. This caused our achieved sample size to be smaller than what we aim for every year. Of the initial 28,500 households selected in the HES 2019/20 sample, we achieved complete responses from 16,151 households. increase in the adult minimum wage from $16.50 to $17.70 effective 1 April 2019 and further to $18.90 effective 1 April 2020. increase in the starting out and training wage from $13.20 to $14.16 effective 1 April 2019 and to $15.12 effective 1 April 2020. New Zealand Superannuation rate (gross) increasing for single living-alone from $475.42 to $490.73 on 01 April 2020; single sharing from $437.14 to $451.29 on 01 April 2020; and ‘both partners qualifying’ from $360.42 each to $372.27 on 01 April 2020. Official Cash Rate (OCR) dropped from 1.50 in July 2019 to 1.00 in August 2019 and stayed at that level through till March 2020; low OCR kept mortgage rates down. Response rate for HES 2019/20 We had set a target achieved sample rate of 70% and were aiming to achieve at least 20,000 responding households in our sample. 28,500 households were initially selected for HES1920. However, it is important to note that interviewing was suspended on 25 March 2020 due to the COVID-19 lockdown. 21,395 households were allocated to the 9 months in which we were out in the field interviewing before the COVID-19 lockdown started. In this 9-month period we achieved complete responses from 16,151 (overall achieved sample rate of 75.44 percent). The number of achieved responses was therefore lower than what we aim for every year. The response rate for HES1920 was 81.59 percent. Response is also monitored at regional level and by NZDEP2013 status. 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 Achieved Sample Rate = ____________________ Ineligible + eligible responding + eligible non-responding 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 Response rate = ___________________ Eligible responding + eligible non-responding 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. Imputation for HES 2019/20 Imputation in HES replaces missing values with actual values from similar respondents. For HES 2019/20, we imputed missing values for the following variables: Age Gender Sex Ethnicity Highest qualification Disability status Number of individuals before and after imputation Number with one or more imputed field Percentage with one or more imputed field Households 5,113 31.7% People 6134 14.4% Households include imputation both at the person level and household level Sampling errors Sampling error is a measure of the variability that occurs by chance because a sample rather than an entire population is surveyed. We can calculate the level of uncertainty around a survey estimate by exploring how that estimate would change if we were to draw many survey samples for the same time period instead of just one. This allows us to define a range around the estimate (known as a “confidence interval”) and to state how likely it is that the real value that the survey is trying to measure lies within that range. Confidence intervals are typically set up so that we can be 95% sure that the true value lies within the range – in which case this range is referred to as a “95% confidence interval”. We calculate sampling errors using the jackknife method. It is based on the variation between estimates of different subsamples taken from the whole sample. The tables below summarise the sampling errors between 2014/15 and 2019/20 by income source and housing-cost type. The tables also indicate the variability of the estimates between the six surveys. 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. Sampling errors for average annual household income, by income source (for households receiving that source of income) Year ended 30 June, 2015-20 Income source Level sampling error (%) 2014/15 2015/16 2016/17 2017/18 2018/19 2019/20 Wages and salaries 3.4 4.0 3.9 3.5 1.5 1.7 Self-employment 13.9 13.3 12.8 11.3 8.7 7.3 Investments 14.4 18.6 25.0 27.2 11.4 7.0 Private superannuation 15.4 22.9 21.2 20.4 New Zealand Superannuation and war pensions 1.8 2.1 2.5 2.0 1.2 1.3 Other government benefits 6.3 7.1 6.3 6.1 2.3 2.8 Other regular sources 28.2 20.2 22.6 14.7 6.9 6.0 Total Gross income 3.2 4.4 3.9 3.7 1.6 1.3 2012/13 and 2013/14 income figures are based on re-based figures. Sampling errors for average weekly household expenditure, by housing cost type (for households with that type of expenditure) Year ended 30 June, 2015 -20 Expenditure item Level sampling error (%) 2014/15 2015/16(1) 2016/17 2017/18 2018/19 2019/20 Property and ground rent 3.7 4.6 4.0 4.5 1.9 1.8 Other payments connected with renting 16.9 19.4 17.4 16.0 8.2 11.8 Total rent payments 3.8 5.0 4.4 4.8 2.0 2.2 Mortgage principal repayments 7.0 8.8 6.4 5.3 2.6 3.2 Mortgage interest payments 4.7 8.9 6.9 7.7 3.6 3.3 Application and service fees for mortgages 52.9 42.4 41.6 21.5 16.8 21.3 Total mortgage payments 4.4 7.6 5.4 5.8 2.3 2.6 Property rates 2.9 2.7 3.4 3.0 1.5 1.8 Building related insurance 3.8 5.0 3.8 3.4 2.1 1.8 Other housing costs 26.0 31.2 36.0 29.4 10.8 21.1 Total housing costs 2.9 4.4 3.6 3.9 1.7 1.8 1. Diary expenditure excluded from sample error calculations to improve comparability between HES (Expenditure) and HES (Income) years. en-NZ