March 01, 2019

BERL Data Newsletter - March 2019

The BERL data newsletter is designed to provide commentary on data issues and news, including methodological changes, new measures, reviews, delays, and other technical issues and news.  

This newsletter covers:

  • Changes to the Household Economic Survey 
  • Measuring child poverty: concepts and definitions

Changes to the Household Economic Survey

Statistics New Zealand announced on 20 February 2019 that it will be expanding the Household Economic Survey (HES) for the 2018/2019 survey.  This is being done in order to ensure Statistics New Zealand can obtain good quality measures of child poverty, as required by the Child Poverty Reduction Act (2018).  

Currently HES is unable to provide adequate data due to its small sample size (3,500 to 5,000 responding households) and has a sample bias which means that information on low-income and high-deprivation households cannot be extracted from the survey.  Therefore, Statistics New Zealand plans to undertake the following change with the 2018/2019 HES:

  • Expand the survey sample to 28,500 households, which with an expected 70 percent response rate will yield 20,000 responding households.
  • Ensure full representation of subgroups (low-income household and regional areas) through stratification of the sample.  This will ensure that the data from the HES can be split by region and by different household types.  To do this Statistics New Zealand will apply the following stratification objectives:
    • 12 regional areas (with West Coast, Tasman, Nelson and Marlborough merged, along with Hawke’s Bay and Gisborne)
    • Urban / rural split with 1.35 urban sampling areas for each rural sampling area
    • New Zealand Deprivation Index 2013 to ensure that there is a good sample for each decile
    • Income of households from the 2013 Census to ensure proportional representation of different income levels.
  • Add a booster sample to increase the number of Maori in the sample based on electoral roll.

So what does this mean for users?    

Overall the changes to the HES will improve the quality of the data supplied to the users, as well as increase the amount of information and data that Statistics New Zealand can make available.  This is mainly due to the increase in sample from 5,000 households to 28,500 households across the country.

In particular the changes will mean that users:

  • Should be able to get data for each of the 12 regional areas, rather than just the five broad regions previously available.
  • May get data breakdowns for particular subgroups (such as low-income households, or Maori) as the new sample will use the NZ Deprivation Index 2013 and 2013 Census household income to ensure it proportionally represents these groupings.  The previous sample was only stratified to split the sample into Urban / Region and the five broad geographical regions.
  • Have increased confidence in the quality of the data outputted from the survey as the sample will be more representative of the broad range of New Zealand households.   

The full paper on the changes to the Household Economic Survey can be accessed here.

Measuring child poverty: concepts and definitions

The introduction of the Child Poverty Reduction Act (2018) requires the government to set three-year and ten-year targets on four primary measures, and that the Government Statistician will report on 10 measures of child poverty each year.  Statistics New Zealand will use its Household Economic Survey (HES) to gather the data it requires to report on the 10 measures of child poverty.

Under the Act the 10 measures Statistics New Zealand needs to report on are:

  1. Low income: Less than 50 percent median equivalised disposable household income before housing costs for the financial year.
  2. Low income: Less than 50 percent median equivalised disposable household income after housing costs for the financial year.
  3. Material hardship.
  4. Poverty persistence.
  5. Low income: Less than 60 percent median equivalised disposable household income before housing costs for the financial year.
  6. Low income: Less than 60 percent median equivalised disposable household income after housing costs for the financial year.
  7. Low income: Less than 40 percent median equivalised disposable household income before housing costs for the financial year.
  8. Low income: Less than 40 percent median equivalised disposable household income after housing costs for the financial year.
  9. Severe material hardship.
  10. Low income and hardship: Less than 60 percent median equivalised disposable household income after housing costs for the financial year and material hardship.

On 21 February 2019, Statistics New Zealand released a paper on its concepts and definitions for these 10 measures.  This paper spells out the definitions Statistics New Zealand will use in reporting on these 10 measures.  While most of the definitions are standard Statistics New Zealand definitions, such as the definition of a household, there are some key definitions that users should be aware of.  These are as follows:

  • A child is defined as being aged under 18, though for the equivalence scales (the process used to equivalise disposable household income) individuals aged 14 and over are defined as adults.
  • A child is a member of a household if they spend four or more nights a week there, or if they spend equal time between two separate households, they will be included in the survey if they are present at the time of the survey.
  • Disposable household income is the sum of all disposable income for household members aged 15 years and over.
  • The equivalence scale (the process used to equivalise disposable household income) assigns a value of 1.0 to the first adult, then 0.5 to each additional adult in the household (anyone 14 years or older), and then 0.3 to each child in the household (anyone under 14 years).
  • Measure one will also be produced using the OECD square root equivalence scale (which is calculated by taking the square root of the number of people in the household) so that it can be reported to the OECD.
  • Housing costs consists of expenditure on mortgage payments, rent payments, property rate payments, and payments associated with building-related insurance.
  • A household will be defined as being in material hardship if it scores six or more points from the 17 items in MSD’s DEP-17 index.  
    • MSD’s DEP-17 index is a list of 17 questions which ask questions around enforced lack of essentials, economising behaviour, restrictions on spending, and financial stress and vulnerability.  An example of this is a household member will be asked if they have a meal with meat, fish or chicken (or vegetarian equivalent) at least every second day? If the answer is no due to cost then the household would get a point, otherwise the household would get zero points.  A second example would be a question on the number of times in the last 12 months an electricity, gas, rates or water bill could not be paid on time.  An answer of two of more times would score a point, otherwise no point would be scored.    
  • A household will be defined as being in severe material hardship if it scores nine or more points from the 17 items in MSD’s DEP-17 index.

The full paper on the concepts and definition of measuring child poverty can be accessed here.