Kevin Hassett Continues to Mislead on the Sunday News Shows
Well, I mean, recounting what I'm hearing and what a lot of people are hearing.
But let me ask you about inflation, because we had the Bureau of Economic Analysis came out this week -- this is the Fed’s go-to gauge for inflation -- saying that inflation increased to 3.8 percent in April.
Are you -- are you concerned about that? I mean, we’re seeing, you know -- and, obviously, a lot’s going to depend on what happens in the Middle East. But are you concerned about this nagging inflation and it might actually force the Fed to increase interest rates?
Hassett replied:
HASSETT: No. You know, first of all, you know, you’re exactly right, the energy prices are high, and they’re high around the world. Gas prices are high, and that’s extremely frustrating, and it’s something that we’re working on doing lots of different things to minimize the disruption. And hopefully, again, the problem of the Gulf will be over soon, and then things will go back to normal.
Absolutely true that the top line number, which includes energy, is a high number. But if you look at core, the history of this is, the energy shocks don’t usually peak through the core. That’s why, at the Fed, you know how they always talk about, well, core inflation this, core inflation that.
Well, actually, as a consumer, you care about overall inflation. But the reason they talk about core is that’s the central tendency of what’s going to happen over the next year or two.
And if you look, at the Cleveland Fed, you probably go here all the time, Jon, but if you don’t, the Cleveland Fed has an nowcaster for CPI, what they think the next month is going to be. And they’ve got core CPI at 0.23, which I think is about right, because that’s what our experience is, is that these energy stocks, painful while they happen, do not feed through to core.
Ignoring Hassett's little dig of "you probably go here all the time, Jon", I was struck by the smallness of the number 0.23. I thought that I did hear the this was projected for next month but it seemed that I had always heard monthly numbers given as annualized numbers so as to make them comparable with the annual numbers that are normally discussed. Hence, I went to the Inflation Nowcasting page of the Cleveland Fed at https://www.clevelandfed.org/indicators-and-data/inflation-nowcasting to see if the 0.23 number was correct and, if so, why it was so small. Following is the first table on that page:
Inflation, month-over-month percent change Month CPI Core CPI PCE Core PCE Updated ---------- ------ ---------- ------ ---------- ------- June 2026 0.12 0.23 0.20 0.27 06/05 May 2026 0.46 0.23 0.40 0.27 06/05 ---------- ------ ---------- ------ ---------- ------- Note: If the cell is blank, it implies that the actual data corresponding to the month for that inflation measure have already been released.As can be seen, the table does list 0.23 as the value for Core CPI for June. However, immediately following that table is the following table:
Inflation, year-over-year percent change Month CPI Core CPI PCE Core PCE Updated ---------- ------ ---------- ------ ---------- ------- June 2026 4.05 2.83 3.90 3.34 06/05 May 2026 4.18 2.82 3.99 3.33 06/05 ---------- ------ ---------- ------ ---------- ------- Note: If the cell is blank, it implies that the actual data corresponding to the month for that inflation measure have already been released.As a year-over-year number, this gives Core CPI for June to be 2.83 percent. This is about a percent less than the 3.8 percent total inflation that Karl had just cited for April but above the Fed's stated target of 2 percent inflation. Hassett could have easily cited this number. To paraphrase Hassett, he probably reads beyond the first table all the time! In any case, the 2.83 number appears to simply be the annualized value of the 0.23 perent month-over-month value. Starting with 0.23 percent, you can convert it to a ratio by dividing it by 100 and adding 1. This gives a ratio of 1.0023. To get the annualized value, you then raise 1.0023 to the power of 12 to get the ratio 1.027952. You then convert this back to a percent by subtracting 1 and multiplying by 100. This gives 2.7952 percent.
This 2.7952 percent number is very close to the 2.83 percent given in the second table. The difference is likely due to round-off error. In fact, you can likely get a more accurate value for the month-over-month change by working backwards. You can convert 2.83 percent to the ratio 1.0283 by dividing by 100 and adding 1. You can then convert this to a month-over-month ratio by raising it the power of (1/12). This gives the value 1.002328. You can then convert this to the month-by-month percent of 0.2328 percent by subtracting 1 and multiplying by 100. This is likely much closer to the precise month-over-month percent than the 0.23 percent figure.
In any case, why would Hassett give the month-over-month figure and force the casual listener to calculate the annualized figure? The answer seems obvious. Hassett's goal was to give the casual listener as small a number as possible and mislead them to believe that core CPI was incredibly low.
Comments
Post a Comment