We hear a lot about how US expansionary policies are causing inflation to accelerate. Is this inflation much faster than what is happening elsewhere (e.g. Eurozone, UK)? Consider the application of diffs-in-diffs, using 3-month annualized consumer inflation rates (calculated as log-differences) for the US, Eurozone, and UK.
figure 1: Three-month annualised inflation rates for US CPI (black), Eurozone HICP (turquoise) and UK CPI (red). The authors seasonally adjusted the euro area HICP using geometric X-12 before calculating inflation rates. Recession dates as defined by NBER are shaded from peak to trough in gray. Source: BLS, Eurostat, ONS Calculations by FRED, NBER and authors.
Inflation in the US has been higher than in the euro zone and the UK. But is the situation so significant, both quantitatively and statistically? To answer this question, I looked at changes in the difference between inflation in the US and the Eurozone (US and UK), from pre-pandemic to post-pandemic. This is the diff method in diff.
Define the annualized monthly/monthly inflation difference between the US and country i:
Take this variable and run the following regression:
Where coronavirus diseaseTon is a dummy variable that takes the value 1 from 2020M02.
The alpha coefficient is the pre-pandemic inflation differential between the US and country i; the beta coefficient is the change in the post-pandemic inflation differential.
Using the HAC robust standard error, I found that the estimated beta coefficient is 0.0028 (HAC robust standard error 0.018) for the US-Euro area and 0.0098 (standard error 0.010) for the US-UK. In both cases, the beta coefficients do not approach the usual level of statistical significance.
So while it is true that inflation in the US is 1 percentage point higher than in the UK and 0.3 percentage points higher in the euro area, the difference is not statistically significant.
