Forecasting wage inflation
Forecasting wage inflation is challenging in today’s economy but remains essential for understanding labour market pressures.
Forecasting wage inflation is challenging in today’s economy but remains essential for understanding labour market pressures. Like many organisations that monitor the economy closely, BERL produces projections of New Zealand wage inflation, i.e., the expected annual growth in wages and salaries in the absence of unanticipated economic shocks.
This article explains why we do this and outlines, in non technical terms, the main methods used to forecast wages. We also discuss some key risks that could cause outcomes to differ from projections.
What is wage inflation?
Before defining wage inflation, it is useful to clarify what is meant by “the wage rate” in a macroeconomic sense. Statistics New Zealand (StatsNZ) publishes several wage measures in its quarterly labour market releases, such as those for the December 2025 quarter.
The main measures are:
- Labour Cost Index (adjusted LCI): Tracks pay rates for a fixed set of jobs, holding labour quality and quantity constant
- Unadjusted LCI: Includes promotions and other compositional effects by not adjusting for job quality
- Average ordinary time hourly earnings: A Quarterly Employment Survey (QES) measure of earnings per paid hour
- Average weekly earnings of full‑time equivalent (FTE) employees: A QES measure of gross weekly earnings per FTE job.
Annual percentage increases in these measures all indicate wage growth. The[l1.1] adjusted LCI is the measure usually used to assess wage inflation and is the preferred indicator used by the Reserve Bank of New Zealand (RBNZ), as it best reflects underlying inflationary pressure from wages.
By contrast, the unadjusted LCI and the QES measures capture changes in what workers actually earn, including promotions, hours worked, and shifts in employment composition.
How to forecast wage inflation
StatsNZ publishes adjusted LCI growth separately for the public and private sectors, and we routinely forecast these components for different applications, including for our Quarterly Economic Brief.
There are two broad approaches to forecasting wage inflation:
Using the history of the series, typically through time‑series models such as ARIMA, where future outcomes are projected based on past behaviour and recent forecasting errors
Using related variables, drawing on economic theory and observed relationships between wages and other indicators.
In practice, these approaches are often combined. In the post-Covid environment, we place greater weight on models that incorporate related variables. Consistent with the classic Phillips Curve idea, 2024 RBNZ research concludes that labour market conditions, especially the unemployment rate, do play an important role, although the relationship is not stable over time.
Further RBNZ work shows that measures of labour market slack, such as the gap between the current unemployment rate and its equilibrium level, can help explain wage pressures. The difficulty is that the equilibrium unemployment rate (often referred to as the NAIRU) is unobservable and changes over time, and so these indicators must be interpreted cautiously and supplemented with professional judgement.
Why are wage inflation forecasts useful?
Wage inflation is most meaningful when considered alongside other variables. Households assess wage growth against changes in living costs; firms weigh wage costs against productivity and revenue; and government agencies must account for wage growth when planning budgets.
Wage inflation forecasts are also central to monetary policy. For the RBNZ, wages link labour market conditions to medium-term inflation outcomes. Persistent wage growth that is inconsistent with productivity trends risks embedding inflationary pressure, even as headline inflation falls. Wage forecasts therefore help to assess whether labour market tightness is easing and whether current monetary policy is likely to return inflation to the target band.
Clearly, credible wage inflation forecasts support decision-making across the economy.
Risks and uncertainties
Wage inflation projections are subject to considerable uncertainty. Net migration remains volatile following large post-Covid swings, and its impact on wages depends on the characteristics of both arriving and departing cohorts.
This year’s general election also adds another layer of policy uncertainty. While the RBNZ is operationally independent, changes in the weight placed on inflation control in the Bank’s remit would affect labour market conditions and the medium-term path of wage growth.
Economic shocks, whether domestic or global, can also quickly render any forecast obsolete. One emerging trend that warrants monitoring is the adoption of agentic AI. As Acemoglu has argued, wage effects will depend on whether AI systems augment workers as advisers or substitute for them as autonomous agents, with very different implications for labour demand and wage growth.
Taken together, these uncertainties reinforce the importance of systematic analysis, which combines statistical models, economic theory, and professional judgement when assessing wage dynamics.