Recall, EJ Antoni and Peter St. Onge argued the US economy has been in recession since 2022.
Source: Antoni and St. Onge (2024) Figure 3.
Dr. Antoni has never declared the end to the recession which allegedly began in 2022, nor has he retracted his assertion. I wondered whether we had ever exited the recession, using their definition of GDP. In my previous exercise (paper here), I could not replicate their results.
However, I now believe I have reverse-engineered the deflator they used.
First, instead of using the GDP deflator published by the BEA, I use a modified CPI. This modified Antoni-St.Onge CPI is given by:
Where CPIex-shelter is the BLS CPI ex-shelter, PCase-Shiller is the Case-Shiller S&P national house price index, and i30yr mortgage is the 30 year mortgage. The ex-shelter CPI is rescaled to 2017=1, and the term in the square bracket is rescaled to 2017=1 as well. This makes anything deflated using this index couched in terms of 2017$.
I can’t figure out where the 0.4 weight on house prices comes from; for the CPI, the weight on shelter is about 0.3. For the PCE deflator it’s even less. However, a 0.3 weight does not replicate Antoni-St.Onge Figure 3.
Deflating the entire GDP using this alternative Consumer Price Index yields this picture (where I include nominal and BEA real GDP for comparison). Why one would want do deflate the entire GDP by a consumer price index is beyond me.
Figure 1: Nominal GDP (blue), real GDP as defined by BEA (tan), and adjusted real GDP using a replicated deflator of Antoni-St.Onge (2024), as calculated by Chinn (green), all in percent deviation from 2019Q1. NBER defined peak-to-trough recession dates shaded green. Source: BEA, BLS, Freddie Mac, Case-Shiller S&P, all via FRED; NBER; and author’s calculations.
Using latest vintage of nominal GDP, it looks like the recession (defined as a contiguous sequence of positive GDP growth), ended maybe in mid-2024.
However, it would be best if Antoni and St.Onge updated their own calculations to verify the recession of 2022 is over.
