Tomorrow the BLS will announce the headline inflation rate for May and there will be much gnashing of teeth and hot takes everywhere. I've already warned you that the official number will be unrealistic because it's based on a comparison to last May, which was artificially low. But what is realistic? Behold my handy inflation calculator:
As an estimate of the real-world inflation rate, I've eliminated the dip in 2020 and replaced it with a straight line. Using this, the CPI index in May 2020 was 259.391.
So when the BLS announces its growth number, ignore it for a moment and instead take a look at the raw index. Then divide it by 259.391 to get a more realistic growth rate.
Please note that this is not an attempt to "deskew" the inflation number or to provide a "true" inflation number. The BLS number will be correct. However, it will also be misleading if you want to know the true state of the economy. My version is less misleading, even though it's technically wrong.
But fear not, inflation hawks! Even using my calculator, the inflation rate for May is likely to be fairly high. It will be at least 3% and probably a bit higher. That's plenty high enough to run around demanding that we throttle the economy now now now in time to affect the 2022 election rein in spiraling inflation.
Let the clutching of pearls BEGIN!!!
"My version is less misleading, even though it's technically wrong."
This, I think, is one of those syntactic gems that makes your blog so enjoyable, Kevin.
Yes! Plus 1 for Kevin's statement AND for Aldous's comment.
Insert my usual question about how much of the federal debt is financed at what rate and how that would change with rates on t-bills etc here.
Was the flatline/dip in early 2019 given similar treatment?
There is a "true" inflation rate, which is the month/month number. This is as close as you can get to the instantaneous rate of change. You can multiply by 12 to get the yearly rate:
https://fred.stlouisfed.org/graph/?g=EC61
Last month the yearly rate was 9.2%, which is pretty high. You don't improve your ability to predict the future by taking an average over arbitrarily longer periods in the past, which is what Kevin seems to think he is doing. To predict future inflation, you have to consider actual factors affecting prices, which is difficult at present. But the idea that any particular single number, averaged over whatever arbitrary period, is predicting an "overheated" economy is absurd, for reasons too complicated for a blog post or comment. There are certain things which are leading to higher prices, but these factors are not likely to persist. If more big stimulus bills were likely to pass things might be different.
I'm not going to give the time of day to anything under 4%. We need more inflation to catch up to all those sub-2% years.
Pingback: The Inflation Rate in May Was Either 5% or 3.5% – Kevin Drum