Household inflation expectations (H1)

Purpose of collection

The primary purpose of collecting the data shown on the H1 table is to capture inflation expectations of the New Zealand general public to support the Reserve Bank with inflation forecasting and policy development.

Survey methodology

Periodicity

The data is collected quarterly, and goes into field after the consumer’s price index (CPI) inflation data have been released by Stats NZ.

Survey population, sample design and sampling method

Data is sourced from the RBNZ’s Tara-ā-Whare Household Expectations Survey. This data is collected by the research provider, Research NZ, on behalf of the Reserve Bank.

The target population is the New Zealand usually resident population aged 18+.

The survey population is the New Zealand usually resident population aged 18+ who are part of the Dynata panel. Dynata is the largest online survey panel provider in Australasia and is used as the sample frame for this survey.

New Zealand residents who aren’t on the Dynata panel are out of scope, and are not part of the survey population.

A large number of Dynata panellists who meet the survey population definition are randomly selected to participate in the online survey. They are invited to participate via an invitation email which includes a unique survey link (which means an individual can only complete the survey once). The survey is set up with quotas on gender, age, ethnicity and region, and a blended sample is used to over-sample Māori and Pacific to ensure a representative response (reflecting New Zealand’s make up). As specific quotas fill up, there is a move to target another wave of invitations to groups that have not yet responded as quickly. The total achieved sample is approximately n=1,000 respondents every quarter.

The total achieved sample is approximately n=1,000 respondents every quarter.

The survey data is weighted using RIM weighting (Random Iterative Method) on age, gender, region and ethnicity from Stats NZ Census data to ensure the data is representative of the New Zealand population.

The Reserve Bank of New Zealand (RBNZ) receives anonymised unit record data from the research provider, which are then aggregated to produce summary statistics.

Timeliness

Data is released around the end of the second month of the reference quarter.

Access by the public

Statistics release calendar

The Statistics Release Calendar provides a long-term plan of scheduled releases. It is updated and released on the first working day of the month.

View the statistics release calendar

Integrity

Provision of information about revisions and notice of major changes in methodology

While seeking to provide a high degree of consistency and long-term comparability of a time series is important, it is also essential that the series is evaluated and changes carefully applied over time to ensure it is still high quality and fit for purpose.

Outlined below are the key changes that have occurred to the data collection since it began in 1995. It is important to keep these changes in mind when interpreting the time series.

The first changes to the survey occurred from December 2008 where the sample shifted from participants aged 15+ to a sample of those aged 18+ (i.e. only adults). In addition to this, the total achieved sample size also reduced from n=1,000 to n=750 respondents.

From June 2018 the survey changed from telephone to online due to the significant cost and diminishing number of people using landlines, and the sample size was increased back to n=1,000 respondents to enable greater analysis options.

From March 2022, changes were made to the questionnaire to provide more precise estimates, increase the value of the survey and better align with international best practice. Please see the series description tab for further detail. The survey has stayed online with a total achieved sample of around n=1,000 respondents.

The RBNZ Household Expectations Survey has been redeveloped over 2021/2022 to improve data quality and coverage, and better align with international best practice for capturing household inflation estimates.

We have renamed the improved collection Tara-ā-Whare - Household Expectations Survey. The word “Tara” is derived from Pakitara, or the walls, and “Whare” means a house. Tara-ā-Whare is also used to describe going to door to door, to ask questions. This survey is run by an external survey provider on behalf of The Reserve Bank of New Zealand. Respondents are asked questions relating to their expectations of CPI inflation, house prices, and the broader financial system.

The survey history is as follows:

• The data from 1995 through to September 2008 is sourced from ACNielsen's Marketscope telephone survey of n=1,000 people aged 15+.
• Between December 2008 and July 2018 the survey is sourced from UMR Research’s nationwide omnibus telephone survey of n=750 people aged 18+.
• Between July 2018 and March 2022 the survey is sourced from UMR Research’s nationwide omnibus online survey of n=1,000 people aged 18+.
• From March 2022 the survey is sourced from Research NZ’s nationwide online standalone survey of n=1,000 people aged 18+.

The data has always been weighted to be representative of the New Zealand population.

Survey content

From Q1 2022 we have refreshed the questionnaire wording and are now using a more precise estimation method for capturing expectations. Outlined below are the main changes that occurred. It is important to keep these changes in mind when interpreting the time series.

The most impactful wording change is to medium-term inflation expectations. In this question we now ask for expectations of an annual inflation rate in a specified time period (of 12 months) 5 years from now, whereas previously we asked for expected inflation in 5 years’ time.

Respondents are now asked for their estimate of a specific inflation rate, whereas they were previously asked to choose from a range of pre-defined categories. As a result, we are now able to compute more precise summary statistics and perform a wider variety of analysis techniques.

To see survey questions and series calculations for before Q1 2022 please refer to the pdf attached to the table called “Survey questions and series calculations prior to Q1 2022”.

Survey questions and series calculations from Q1 2022

View the survey template

The questions asked and the series calculations are outlined below.

For more information on weighting, outliers, and interpreting key series please refer to the methodological note attached to the tables.

Perception of current inflation

1. Overall, in the last 12 months, do you think there has been inflation or deflation?

That is, the price of goods and services (i.e. the price of the ‘basket’) increased or decreased.

Please select one option:

• Inflation (the price of the ‘basket’ has increased overall)
• Deflation (the price of the ‘basket’ has decreased overall)

2. As a percentage, what do you think has been the rate of inflation/deflation in the last 12 months? Use whole percentages and/or parts of percentages to tell us. If you’re not sure, a good guess will do. (numeric answer given)

The median current inflation estimate is calculated by taking the middle value of the sorted numerical responses given in question 2.

The mean current inflation estimate is calculated by taking the weighted average of the responses given in question 2.

Expected inflation 1-year

3. In the next 12 months (from month, year of survey date to month, year of survey date + 12 months), do you think there will be inflation or deflation?

Please select one option:

• Inflation (the price of the ‘basket’ has increased overall)
• Deflation (the price of the ‘basket’ has decreased overall)

4. As a percentage, what do you think will be the annual rate of inflation/deflation in the next 12 months? Use whole percentages and/or parts of percentages to tell us. If you’re not sure, a good guess will do. (numeric answer given)

The median expected inflation rate 1-year estimate is calculated by taking the middle value of the sorted numerical responses given in question 4.

The mean expected inflation 1-year estimate is calculated by taking the weighted average of the responses given in question 4.

The net percent expecting higher inflation – derived estimate is calculated by using the weighted responses to question 1 and question 3. The calculation is as follows:

(weighted no. expecting 1 year ahead inflation to be higher than their perception of current inflation - weighted no. expecting 1 year ahead inflation to be lower than their perception of current inflation) / (weighted no. of responses where there was an estimate for both current and 1 year ahead inflation) x 100

The net percent expecting inflation estimate is calculated by using the weighted answers to question 1. The calculation is as follows:

(weighted no. expecting inflation 1 year ahead - weighted no. expecting deflation 1 year ahead) / (total no. of weighted repsonses) x 100

Expected inflation rate 2-years

5. And what do you expect two years out from now (i.e from month, year of survey date + 12 months to month, year of survey date + 24 months)? Do you think there will be annual inflation or deflation?

Please select one option:

• Inflation (the price of the ‘basket’ has increased overall)
• Deflation (the price of the ‘basket’ has decreased overall)

6. As a percentage, what do you think will be the rate of annual inflation/deflation in the 12 months from (month, year of survey date + 12 months to month, year of survey date + 24 months)? Use whole percentages and/or parts of percentages to tell us. If you’re not sure, a good guess will do. (numeric answer given)

The median expected inflation rate 2-years estimate is calculated by taking the middle value of the sorted numerical responses given in question 6.

The mean expected inflation rate 2-years estimate is calculated by taking the weighted average of the responses given in question 6.

Expected inflation rate 5-years

7. Thinking further out into the future now. Do you think there will be inflation or deflation five years out from now (i.e. from month, year of survey date + 48 months to month, year of survey date + 60 months)?

Please select one option:

• Inflation (the price of the ‘basket’ has increased overall)
• Deflation (the price of the ‘basket’ has decreased overall)

8. As a percentage, what do you think will be the annual rate of inflation/deflation in the 12 months from (month, year of survey date + 48 months) to January (month, year of survey date + 60 months)? Use whole percentages and/or parts of percentages to tell us. If you’re not sure, a good guess will do. (numeric answer given)

The median expected inflation rate 5-years estimate is calculated by taking the middle value of the sorted numerical responses given in question 8.

The mean expected inflation rate 5-years estimate is calculated by taking the weighted average of the responses given in question 8.

Net percent expecting higher house prices

9. In the next 12 months, do you think the average price of houses in your region will:

Please select one option:

• Increase overall
• Stay about the same
• Decrease overall

The net percent expecting higher house prices estimate is calculated by using the weighted responses to question 9. The calculation is as follows:

(weighted no. expecting house prices to increase overall - weighted no. expecting house prices to decrease overall) / (total no. weighted of responses) x 100

Expected house price inflation 1-year

10. As a percentage, how much do you expect average house prices will increase/decrease in the next 12 months, in your region. Use whole percentages and/or parts of percentages to tell us.

If you’re not sure, a good guess will do. (numeric answer given)

The median expected house price inflation 1-year ahead is calculated by taking the middle value of the sorted numerical responses given in question 10.

The mean expected house price inflation 1-year ahead is calculated by taking the weighted average of the responses given in question 10.

Expected house price inflation 5-years

11. Thinking now about house prices in 5 years’ time. Do you think the average price of houses in your region will be …

Please select one option:

• Higher than they are now
• About the same
• Lower than they are now

12. As a percentage, how much do you expect average house prices will increase/decrease in 5 years time, in the part of the country you live in.

If you’re not sure, a good guess will do. (numeric answer given)

The median expected house price inflation 5-year ahead is calculated by taking the middle value of the sorted numerical responses given in question 12.

The mean expected house price inflation 5-year ahead is calculated by taking the weighted average of the responses given in question 12.

Symbols and conventions for summary table

General notes

• Individual figures may not sum to the totals due to rounding
• Percentage changes are calculated on unrounded numbers
• You are free to copy, distribute and adapt these statistics subject to the conditions listed on our copyright page.