Phillips Curve Redux?

Back in my undergraduate days (at the end of the 1960s, if you're asking), the Phillips curve was a significant part of the macroeconomics courses. The original paper by William Phillips, published in 1958,  posited an inverse relationship between the unemployment rate and wage gains. By the time I was at university, other economists had broadened this out to suggest an inverse relationship between unemployment and inflation itself. You can read the history of the concept here. 

I was never a great fan of the Phillips curve, not least because the original data plot (embedded in the linked article) did not exactly show what you would call a robust relationship. Later analyses have also tended to show that the concept is too simplistic a representation of complex economic relationships to be of much use as either a forecasting tool or a guide for policymakers.

It's surprising, then, to find that here in 2022, monetary policy in the major economies seems to be driven by Phillips's ideas. The Federal Reserve is openly admitting that the rapid monetary tightening of the past several months is aimed at calming labour markets (i.e. raising unemployment) in order to get inflation down to the 2 percent target. The Bank of Canada and the Bank of England are using similar logic in setting their own policies. 

The problem with this should be apparent. The recent inflation spike is largely the result of global supply chain pressures, initiated by the COVID pandemic and exacerbated by the Russian assault on Ukraine. There is little or no resemblance between these circumstances and those which the Phillips curve attempted to describe. Hence there is little good reason to think that slowing the economy and raising unemployment will do much to reduce inflation. To put it another way, it is very difficult indeed to know just how high unemployment might have to go in order to curb inflation that has almost nothing to do with tightness in labour markets.

All of which is offered by way of introduction to the September jobs market data for the US and Canada, released this morning. In both countries, the data were a little stronger than expected, which immediately prompted markets to increase their bets on further aggressive rate hikes.  It may be that neither the markets nor central bankers really believe in the Phillips curve, but right now both are acting as though they do. 

In the US, the BLS data show that employment rose by 263,000 in September, slightly above market expectations. This pushed the unemployment rate a tick lower, to 3.5 percent. The monthly gain is well below the average posted so far this year, which stands at 420,000, but there is no doubt that the jobs market is still tight, despite the lack of GDP growth so far this year. Still, there is not much evidence that this tightness is pushing wages higher: average hourly earnings rose 5.0 percent in the year to September, more than three percentage points below the latest headline CPI print. No wage-price spiral there.

As for Canada, StatsCan reported today that employment rose by 21,000 in September. The unemployment rate dropped by 0.2 percentage points to 5.2 percent, driven not only by the rise in employment but also by a slightly surprising fall of 20,000 in the number of people looking for work. The rise in employment largely reflected a rebound in employment in the educational sector, which underwent a large and unexplained fall in August, so the headline numbers may overstate the underlying strength in the jobs market. That said, the market is still tight, with high vacancy levels, yet there is still little sign of wage pressures: hourly wages rose 5.2 percent in the year to September, well below the 7 percent gain posted by CPI. 

Even before the data appeared, both Fed and Bank of Canada spokespersons had stressed their belief that further rate hikes would be needed to get inflation back to the target range. The Fed may well be looking at yet another 75 basis point hike next time out, while the Bank of Canada may throttle back to 50 basis points, unless it is spooked by the weakness in the exchange rate. One wonders what William Phillips, the very definition of what Keynes long ago referred to as "some defunct economist", would think of all this.