I've explained in passing the difference between the annual inflation rate and the annualized monthly rate, but based on some feedback it sounds like it might be a good idea to do it a little more clearly. It won't take long.
It all starts with an index of price levels, calculated by whichever agency is measuring inflation. For example, here is the PCE core index over the past year:
It's easy to calculate the change in price level over the past year or the past month:
- Annual (year-over-year) = 121.804 ÷ 116.100 = 1.049 = 4.9%
- Monthly (month-over-month) = 121.804 ÷ 121.386 = 1.0034 = 0.34%
- Annualized month-over-month = 1.003412 = 1.042 = 4.2%
The annual inflation rate, which is the one you commonly see in headlines, is just the rise in the price level between this month and the same month last year. It's generally the best measure, but it has one drawback: it doesn't show you what's happening right now. If the price level jumps in a single month, the annual rate is going to stay high until that month finally drops out of the data—even if the price level has been stable ever since.
The monthly inflation rate is just the opposite. It shows you precisely how much prices rose over the past 30 days, but the downside is that this is pretty volatile. Generally speaking, a single month's change isn't something you should pay much attention to. However, the trend over time of the monthly rate can be useful.
Finally, the annualized monthly rate is just a bit of arithmetic. It shows what the annual inflation rate would be if prices rose at this month's rate for twelve months straight. It's a handy transformation because we're all used to thinking in annual terms.
With that said, here is this morning's chart again:
As you can see, the monthly rate does jump up and down a fair bit, so no single month provides reliable data. However, the fact that the trendline is generally down means that inflationary pressure appears to be easing.
Aah. Weird break at May 21 because it is annualized and we have data up to May 22.
"... However, the fact that the trendline is generally down means that inflationary pressure appears to be easing."
This is pure wishful thinking. There is no good reason to look at the trend since April 2021 instead of February 2021 or November 2020. And the fact that this makes a big difference indicates the data is probably too noisy to conclude much of anything.
On the contrary, what I see in that chart is an overall upward trend with a coinciding temporary shock to the system that resulted in an even larger than expected bit of inflation. If this were a 3-month rolling average of the annualized monthly rate, that would be pretty apparent - it would smooth out the extremes a bit.
What I see is inflation that was in the 2.something-% area prior to 2021 and then going to somewhere a little above 4%, with no real sign yet of it slowing down enough to stabilize or turn back downwards.
Well, then why isn't inflation rising???? Because it looks stopped to me.
But looking at "what's happening right now" seems awfully misleading because there is always a lot of noise in these statistics. Just look at that month-to-month plot of the huge swings up and down. If I had been plotting local gas prices month-to-month there would have been a big uptick this spring, then some downward drops, and now it is back up to the highest it has been. If I chose the moment to look when the prices had fallen I would have mistakenly thought gas prices were going down. In any case, the public and voters don't care about the PCE Core Inflation Rate--they are looking at how much gas is at the pump, what groceries cost, and what their checkbook balance is at the end of the month. Because so many people live paycheck-to-paycheck they know exactly what inflation means to their own finances. They see it every paycheck.
Because so many people live paycheck-to-paycheck...
That is the tragedy, or the black comedy, of life for too many folks.
Having increasingly rejected religion, the traditional opiod for easing the pain, they have turned to the real stuff.
“the annualized monthly rate is just a bit of arithmetic” which happens to imply an underlying mechanism with a time constant of a year or more, whether there is any rational basis to assert this, or not; and to multiply noise by 12 to boot.
I have always held the position that both the annual and annualized are important parts of the whole picture and it is poor journalism to only give half the picture.
The sad thing is that the fed rate hike is going to get the credit for the decrease that happens in annual inflation which will occur in July/August, when all that happened is that some of the highest inflation months have fallen off the stats. There will then be a ton of stories reinforcing that FED rates have a lot influence over inflation rates. My position is that they are basically a tiny bit of noise on top of semi-random global events.