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.

The resurgence of the New Zealand Phillips curve

This Note assesses how the Phillips curve, which describes the relationship between inflation and economic activity has evolved over time.

Willy Alanya-Beltran, Maui Brennan, and Punnoose Jacob

Key findings

  • This Note assesses how the Phillips curve, which describes the relationship between inflation and economic activity, has evolved over time. A steeper Phillips curve can help explain why inflation increased so sharply in the COVID-19 period and would also imply that inflation should decrease quickly as capacity pressures ease.
  • In general, the slope of the Phillips curve (across a range of inflation and activity indicators) drifted upwards during the pandemic and beyond, from the relatively lower levels observed between 2008 and 2019. This indicates a stronger relationship between activity and inflation since 2020.
  • The most noticeable steepening of the Phillips curve in the post-2020 period has been observed when job-to-job labour flows and a QSBO labour shortage measure are used as activity measures. Both indicators have been shown in previous work to correlate well with inflation. In contrast, we find that the changes in the slope of Phillips curve featuring the unemployment rate have been relatively mild.

Why we did this research

The research presented in this Note is aligned with our monetary policy research agenda themes — inflation and the labour market.

Central banks manage aggregate demand by moving policy rates, and when the aggregate supply curve is flat, firms move prices less for a given change in demand. This implies that, other things constant, monetary policy has a lower impact on price inflation when the supply curve is flat.  This Note assesses the evolution of the Phillips curve in New Zealand during the inflation targeting era using time-varying parameter models. In particular, we assess if the elasticities of various measures of inflation to various measures of economic activity have changed over time.

By mid-2021 the public health restrictions which accompanied the arrival of COVID-19, and related impediments to global supply chains, together with expansionary fiscal and monetary policy, rekindled inflationary pressures that had been subdued for a long time. Constrained global supply was worsened by the Russian invasion of Ukraine in early 2022. Consequently, inflation rose above central bank targets in many developed economies, including New Zealand, and remained high as economic activity improved.

We find that the time-varying parameter Phillips curves estimated for most inflation-activity measure combinations in New Zealand, have considerably steepened during the COVID-19 pandemic, and beyond. If the steepening of the estimated Phillips curve is due to the steepening of the aggregate supply curve, it implies that small changes in economic activity triggered by fiscal and monetary policy can contribute to bigger changes in inflation, as experienced recently. On the flip side, if the Phillips curve remains steep, current and future macroeconomic policy settings can bring about larger changes in inflation with small changes in activity.

What data have we used?

Table 1 lists the 7 inflation and 7 activity measures used in the estimations. The quarterly dataset is compiled using standard time-series from Statistics New Zealand, the Ministry of Business, Innovation and Employment, and the Inland Revenue Department, as well as estimates of core inflation and economic activity produced by the Reserve Bank.

Table 1: Inflation and activity measures

Inflation measure Activity/Slack measure
CPI (GST-adjusted) – 1990Q1 Unemployment rate – 1990Q1
RBNZ sectoral factor model for core inflation – 1996Q2 RBNZ unemployment rate gap – 1990Q1
RBNZ factor model for core inflation – 1994Q3 RBNZ output gap – 1994Q3
Tradables CPI (GST-adjusted) – 1990Q1 Employment to working age population ratio – 1990Q1
Non-tradables CPI (GST-adjusted) – 1990Q1 Vacancy-unemployment ratio – 1992Q4
Labour cost index (LCI) wage (adjusted for productivity changes) – 1996Q3 QSBO labour as a limiting factor indicator – 1998Q3
Quarterly employment survey (QES) wage – 1992Q4  Job-to-job transition rate – 2002Q1