Stats NZ has a new website.

For new releases go to

As we transition to our new site, you'll still find some Stats NZ information here on this archive site.

  • Share this page to Facebook
  • Share this page to Twitter
  • Share this page to Google+
Information about data

The data used in this paper comes from the Household Economic Survey (Expenditure) 2012/13 and 2015/16, and the Census of Population and Dwellings.

Household Economic Survey (HES)

The Household Economic Survey (HES) collects information every year on household income and housing costs, with an additional collection of household expenditure every three years.

HES is continuously in the field, which smooths out any seasonal variations. This method of collection ensures its ability to create an accurate picture of annual household income. The survey measures expenditure every third year. While we can measure some indicators of energy hardship in HES annually, the expenditure data is only available three-yearly. Relevant questions from the Economic Standard of Living Index (ELSI) have been available since 2012/13 so we have a short time series. HES has a relatively small sample size (with an achieved sample size of around 3,000–4,000 households), which limits cross-sectional analysis.

HES data based on objective and subjective measures

Most of the information for this paper comes from HES, and uses objective information on household energy costs as well as subjective material well-being measures relating to energy hardship. These self-reported measures are asked of one person aged 18 or over in the household. We have assigned these responses to the household, however, we know there can be some individual variability, particularly in the experience of coldness. See Appendix 1 for more information about ELSI data.

HES measures expenditure on household energy from the latest power/utility bill, therefore energy costs are averaged over different seasons. Since energy bills can vary considerably between winter and summer, energy expenditure as a proportion of income is likely to be higher in winter. High energy bills are likely to create greater stress for the household, although there are ways that households can manage their fuel bills if this is an issue. These include:

  • a level pay system where the household pays the same amount per month (or fortnight) – this system prevents winter peaks in expenditure
  • prepayment metering, where households purchase a set amount of electricity, as their budgets allow.

Census of Population and Dwellings

The Census of Population and Dwellings occurs every five years (although there was a seven-year gap after 2006 due to the Canterbury earthquakes). Census information on energy hardship is based on the question “Mark as many spaces as you need to show which of the following are ever used to heat this dwelling”. The response “don’t ever use any form of heating in this dwelling” is currently the only census question that can provide any information about energy hardship. As it is, it has limitations because we don’t know whether the household did not use heating because they could not afford it or because they had an energy efficient home. Additional questions on housing quality in the 2018 Census would create a more accurate picture of energy hardship than is currently available due to the limitations of ‘no heating used in this dwelling’.

Our approach

In Background to energy hardship we will discuss the range of indicators used internationally to measure energy hardship. In New Zealand we do not have data to be able to calculate all of the same measures used internationally, however, there are a number of other ways that energy hardship can be estimated.

We will explore nine indicators from Household Economic Survey (HES) and one from the Census of Population and Dwellings. This is similar to an approach used in Australia. Australian researchers applied five measures to the Household, Income and Labour Dynamics in Australia (HILDA) Survey, and concluded that although there were wide differences in the size of these populations according to the measure used, there was “a clear association between socio-economic disadvantage and the experience of fuel poverty, regardless of whether disadvantage is measured using an income poverty definition or the multidimensional indicator of social exclusion” (Azpirante, Johnson, and Sullivan, 2015). Overall, they found that households who reported that they were ‘unable to heat the home’ experienced the greatest social exclusion of all groups.

The indicators we use from HES include cost-to-income/expenditure measures (objective measures):

  • households that spent twice the median proportion of income on domestic energy (before and after housing costs)
  • households that paid 10 percent or more of their income on domestic energy (before and after housing costs)
  • households where domestic energy costs are in the highest quartile as a proportion of all expenditure.

And some subjective measures:

  • the inability to pay utility bills on time (electricity, gas, water, or rates bills)
  • the percentage of dwellings that were hard to heat or keep warm
  • the percentage of dwellings that were damp or mouldy
  • the number of people who put up with feeling cold a lot.

We combined indicators in two ways. First, we created a number of energy hardship indicator measures, and looked at how many energy hardship indicators households experienced. Then we created a composite measure, where we combined all the energy hardship indicators and gave each household a score. Households with a combined score in the top quintile were considered to be experiencing energy hardship.

We also looked at households in the New Zealand census who said they used no heating in their dwelling, to assess the validity and utility of this measure in providing information on energy hardship.

  • Share this page to Facebook
  • Share this page to Twitter
  • Share this page to Google+
  • Share this page to Facebook
  • Share this page to Twitter
  • Share this page to Google+