Is the Phillips Curve Still a Curve? Evidence from the Regions

Abstract

The way in which wages respond to very low rates of unemployment remains a key source of uncertainty in Australia, partly due to the lack of historical evidence to draw upon. To help fill this gap, we study data on unemployment rates and wages growth across local labour markets over the past 20 years. The considerable variation in economic conditions across local labour markets allows us to infer the strength of the relationship between unemployment and wages growth (i.e. the wage Phillips curve) at very low unemployment rates that are rarely seen at the national level. We find strong evidence that the wage Phillips curve is indeed a curve, rather than a straight line. When the unemployment rate exceeds 7½ per cent, the Phillips curve is flat and wages growth is unresponsive to changes in unemployment. Wages growth then becomes increasingly responsive to changes in the unemployment rate as the unemployment rate falls to lower and lower levels, most notably below 4 per cent. These findings have implications for monetary policy, particularly at the current juncture given the Reserve Bank of Australia's central forecast for the unemployment rate to fall to multi-decade lows in the next few years.

Non-linearity in the Phillips curve: evidence from Nigeria

“…The authors found that wage rates rise steeply when unemployment is low, and the changes in wages become flat at high unemployment rates. The same conclusion was observed by Bishop and Greenland (2021) in the case of Australia. Similarly, Cristini and Ferri (2021) examined the different types and degrees of nonlinearity in the price Phillips curve in the USA over the period 1961Q1-2019Q4, deploying both pairwise and threshold techniques.…”

A new perspective on the asymmetric Phillips curve: Fresh evidence from ECOWAS member countries

“…This finding validates Phillip's (1958) theoretical argument that inflation increases steeply when unemployment is low, and the change in inflation becomes flat at a high unemployment rate. This outcome is congruent with the findings of Kumar and Orrenius (2015), Nalewaik (2016), Albuquerque and Baumann (2017), Bryne and Zekaite (2017), Bishop andGreenland (2021), andOnatunji et al (2023), who discovered similar findings in their investigations. Additionally, the results show that positive (rising) unemployment has a significant negative effect on inflation, while negative (declining) unemployment has an insignificant negative impact on inflation in Cape Verde, Gambia, and Guinea-Bissau.…”

“…The results reveal that wage rates increase significantly when unemployment is low, and the rate of change in wages becomes con- Trade-off stant at high unemployment levels. Bishop and Greenland (2021) found similar results when examining the Australian economy. Furthermore, Cristini and Ferri (2021) studied the various forms and levels of nonlinearity in the price Phillips curve in the United States over the period 1961Q1-2019Q4 by employing both pairwise and threshold techniques.…”

The Non-Accelerating Infection Rate of Unemployment as an alternative to the Non-Accelerating Inflation Rate of Unemployment for the period of the COVID-19 pandemic: The example of the industries of the Czech Republic

“…Babecka, et al (2022) confirmed the cyclical dynamics of inflation according to the PC for the Czech economy as well. Bishop and Greenland (2021) find strong evidence that the wage PC is indeed a curve rather than a straight line. Hronova, et al (2021) draw attention to the fact that the indicators of the inflation rate and the unemployment rate are still necessary to evaluate the trend of the difference in their values over time.…”