3 min read
How worried should we be about US inflation?
EFG Chief Economist Stefan Gerlach looks at the behaviour of US inflation as measured by the deflator for personal consumption expenditures (PCE) and whether recent rates are really a cause for concern.
The latest data showed the US CPI rose in May, by 0.6% month-on-month and 5.0% year-on-year, the highest annual inflation rate since 2008. The rising inflation has attracted attention but should policy makers be concerned?
One view that attracts considerable support among some financial market commentators is that inflation will stay persistently and uncomfortably high, reflecting a fundamental change in the economic environment that the Fed has yet to understand. Under this view, the Fed may suddenly change tack and sharply tighten monetary policy, potentially leading to a negative financial market shock. The other view, which appears to be held by many Fed policy makers, is that recent increases in inflation are very much what one would have expected given the decline in prices during the start of the covid pandemic last year.
Recent inflation data
To understand the recent behaviour of PCE inflation, it is useful to consider the base effect. To do so, it is easiest to use inflation computed using compounded growth rates. The recent pattern has seen the new data pushing up 12-month inflation and the old data that drops out pushing it down. One reason inflation has risen so sharply recently is that the price level fell as Covid struck in March (‑0.27%) and April (-0.53%) 2020, reducing 12-month inflation at that time. But this decline in the PCE index pushed up inflation in March and April 2021 by the same amount as these data points dropped out of the calculations.
The outlook for inflation
While recent inflation data has been strong, the Fed is insisting that its outlook for inflation has not changed. Apparently, the recent data do not appear to be at odds with the Fed’s view of the likely evolution of inflation in 2021.
To explore this issue, we estimate a simple forecasting model for PCE inflation on data for 2000-2020. While the actual inflation rate in May was 3.84%, the forecasted inflation rate was 3.13%. The likelihood that actual inflation and the forecasts will be identical is of course nil. To judge whether this difference is so large as to suggest that the behaviour of inflation has changed, we use a ‘fan chart’ that shows the range of possible inflation rates that are compatible with the model.
US PCE inflation and model forecasts
If actual inflation falls outside the forecasted range, we conclude that the inflation process, as captured by the forecasting model, has changed. As can be seen, it never does. While inflation is a little high, it is not so high as to call the forecasting model into question. Looking ahead, the forecasting model suggests that inflation will decline over the summer months.
There is always potential for the inflation process to shift, but the analysis here indicates that recent inflation data provide little evidence that a shift has occurred. Indeed, it is not surprising that Fed officials have reiterated their view that inflation is only high temporarily and will soon revert to more normal levels. Nevertheless, the risk of such a shift taking place has risen recently and we remain vigilant to its possibility, although that is only likely to become apparent - if it occurs at all - as the second half of the year progresses.
Download the full edition of our Infocus publication here.