Browser issue

It looks like the browser you're using doesn’t work well with our website. For a better experience, please update to the latest version of Chrome, Edge, Firefox or Safari.

Evaluating measures of price-setting behaviour: New evidence for New Zealand

We use the term ‘price-setting behaviour’ to describe how firms dynamically adjust their prices based on past and expected future inflation. This paper examines which indicators of price-setting behaviour best help explain and predict non-tradables inflation in New Zealand, showing that no single measure tells the whole story.

Justine Lee

What this paper is about

Inflation is ultimately shaped by the prices firms choose to set. Those decisions reflect demand and supply conditions but are also influenced by how households and businesses interpret past inflation, what they expect inflation to do next, and how firms’ own costs and pricing plans are evolving.

This paper compares a broad set of modelled and survey-based indicators of price-setting behaviour to assess how well they explain and forecast non-tradables inflation in New Zealand. It considers traditional modelled measures based on past inflation, households’ and businesses’ inflation expectations, and newer indicators that are closer to firms’ day-to-day pricing decisions, such as expected own costs and selling prices. The results highlight that different indicators are useful in different circumstances, especially when inflation dynamics are changing quickly.


Key findings

  • Models that include measures of price-setting behaviour predict non-tradables inflation better than models that exclude them.
  • Compared to survey-based measures of price-setting behaviour, modelled measures generally perform well.
  • Survey measures also contain useful signals. Households’ perceptions of current inflation and firms’ own cost and selling price expectations can perform strongly in some settings.
  • The usefulness of different measures changes over time. Price-setting behaviour indicators were especially informative during periods of inflation volatility, including the post-COVID-19 inflation episode.
  • Combining indicators can improve forecasting accuracy, suggesting that different measures capture complementary information regarding inflation dynamics.
  • The results support using a suite of price-setting behaviour indicators rather than relying on a single ‘best’ measure when assessing inflation pressures.
  • For monetary policy, the findings reinforce the value of monitoring how inflation expectations and firms’ pricing intentions evolve, particularly when inflation is moving rapidly.


Why we did this research

Understanding inflation requires more than just tracking demand, supply, wages, or import prices. Policymakers also need to understand how firms respond to past inflation and how expectations about future inflation feed into pricing decisions. This is particularly important for non-tradables inflation, which is more closely tied to domestic pricing behaviour than tradables inflation.

We already use price-setting behaviour in our core forecasting framework, but there are many possible ways to measure it. This paper asks which measures are most useful, whether survey-based measures add value alongside modelled indicators, and whether their usefulness changes across different inflation environments. The post-COVID-19 inflation surge makes these questions especially relevant, as it provides a recent episode in which firms and households had to respond to unusually large and rapid changes in inflation.


What data have we used?

The analysis uses quarterly New Zealand data from 1995 Q1 to 2025 Q2.

It compares 23 possible measures of price-setting behaviour, including:

  • modelled measures based on past CPI inflation
  • household and business inflation expectations from our surveys and the ANZ Business Outlook
  • firms’ own cost and selling price expectations from the ANZ Business Outlook and NZIER’s Quarterly Survey of Business Opinion.