How accurate are inflation expectations? The literature says two things:
- Any measure based on consumer expectations is pretty worthless. This is no surprise since consumers don't really know much of anything.
- Market-based measures, such as the breakeven rate, are reasonably accurate, generally within a percentage point or two of reality.
So let's take a look at the 5-year breakeven rate, which forecasts the inflation rate five years in the future:
As you can see, inflation over the past 20 years has been running at roughly 2% and the breakeven rate has also been at about 2%. This means that the breakeven forecast is pretty good when inflation is steady.
It's also no problem that the forecast didn't predict the big temporary spikes in 2009 and 2021. There's no way it could do that.
But there is a problem here: the predicted plummet for 2014. What happened here is that actual inflation in 2009 went negative (red line), so the bond markets mindlessly predicted that five years later in 2014 it would also be negative (blue line). Needless to say, this makes no sense and it didn't happen.
Conclusion: market forecasts of inflation may be better than consumer forecasts, but the "market" basically just figures that inflation in five years will be about the same as it is today. As long as inflation is steady, this works OK. But then again, as long as inflation is steady, who needs forecasts?
CPI inflation is a point of debate. If you revise down 1970's inflation then real wages boomed in the 70's which is shown by the surge in real retail sales. The decline in real wages then in the 80's led to slower growth of real retail sales. I have always said, government accountability with inflation is bad. It's a heavily weighted and hedonically adjusted to fit a narrative. In the 70's they made a error and it was costly(and yes, inflation was still too high, but incomes were more than keeping up).
Real wages declined in the 70's and 80's no matter which index you use.
Nope, not when you adjust inflation downward. Real wages boomed. It's why consumption was so high. Look at 1979. Inflation going from 13.3 to 8% caused positive income growth.
Kevin:
You are showing your ignorance.
Example, All AGW models miss big changes. Does that make them worthless? No, of course not. You look for the best models you can find.
There IS no better inflation forecasting model than the difference between TIPS and Treasuries. (Yes, I am sure some people have tweaked it so their model is slightly better.)
So you just discovered that no model is going to be accurate jumping off the deep end a couple of times a decade.
You could have made a fortune betting that inflation was going to be higher that forecast during the 5 year period ending 2014. But you didn't. Very few people did. And those that did probably always predict low inflation.
What is your guess for inflation over the next five years? Higher or lower than the difference between TIPS and Treasuries?
It's not totally clear from the chart, but it looks like the red line is the one-year increase in CPI (i.e. June 2021 CPI divided by June 2020 CPI), while the blue line is the 5-year breakeven rate (i.e. the rate of inflation over 5 years that would provide the same present value for a 5-year Treasury inflation-protected security (TIPS) vs a nominal, non-inflation-adjusted Treasury security.
These aren't really comparable. The red line should be the fifth root of June 2021 CPI divided by June 2016 CPI, to get the average annual rate of inflation over those five years.
If anything, I think that adjustment makes the market forecast look worse. The red line would be much less jumpy if it were a trailing 5-year rate rather than a single year rate, and that means that the differences are more pronounced, particularly the completely-off-base expectation that inflation would stay negative *for five years* starting at the end of 2008.
The market doesn't really know more about this than anyone else.
The market is not good at predicting inflation.
But what is better at predicting inflation?
I really don't know of anything that comes close.
Do I need to point out that in particular, this observation is all that is needed to dismiss those drumbeat chants of INFLATION! IS! JUST! AROUND! THE! CORNER! from the usual suspects?
You've been handed a rule ... or at least a rule-of-thumb. Furthermore, you give every indication agreeing that it is correct; I suggest that you apply this bit of academia to the real world.
I don't pay attention to the 5 year inflation numbers. I pay attention to the 10 year numbers which are around 2.3%.
If I had to guess, and I am paid not to guess, I would bet it will be below 2%. But what the hell do i know?
I dunno. I'm wondering how to make sense of your reply to my comment on the applicability of that observation you've been pooh-poohing to the inflationistas.
The underlying reality gets ignored. Inflation is a monetary phenomenon that involves matching the money supply to the economic activity. Baby boomers started entering the workforce in about 1965 and the supply of workers quickly exceeded the economy's need for workers. So we had excess workers that people called hippies and marveled over community living. Ten years later the Boomers had been assimilated into the economy and began spending money that the economy didn't provide: so there was inflation as the excess demand drove up prices. Now Boomers are starting to retire or die and this changes the economy as well. All the second homes, collections of assets, and other Boomer stuff will start flooding the market. Fun times ahead.
No Friedman for me, thank you very much. Monetarism as government policy was discarded long ago as not an accurate description of reality.
Wow. We agree on something.
Evidently neither consumers, bond buyers or economists know what's going to happen, especially in the future.
There are two kinds of people.
People who don't know what the average inflation rate will be for the next 5 years
And
People who don't know that they don't know.
I dunno. I'm wondering how to make sense of your reply to my comment on the applicability of that observation you've been pooh-poohing to the inflationistas.
Most of the people in the upper half of the income distribution have all the junk they need. I can’t imagine anyone is thinking “I must buy more junk now because the price of junk is rising.”
Maybe some fo, but I don’t. That’s dumb.
Filter further on being over fifty and I might agree with you. But there's a lot of relatively young people close up to that boundary who are still paying downed a well-incurred debt load. Buying property, college degrees, etc. will do that to you.
Ehhhh, yes and no.
The problem with the 2009 "prediction" isn't that the market mindlessly thinks the "same" but the truth isn't any comfort on the power of the market either... during the Lehman crisis there was both (a) a flight to quality that drove nominal treasury yields down and (b) some mundane plumbing/technical/liquidity issues in the TIPS market such that yields actually went up more than "fundamental analysis" would have suggested.
Net effect (where breakeven is difference between the two yields) went negative. Not great!
'Any measure based on consumer expectations is pretty worthless. This is no surprise since consumers don't really know much of anything.'
No, this is true because unlike the stock market- which is almost purely based on perception of the perception of the perception of other people- inflation is a measure of the Real World. In other words, expecting high inflation does not make it so.
Maybe I'm missing something here, but it looks like you're missing a significant point, Kevin.
On the one hand, you argue correctly that bond markets tend to project today's inflation rate out more or less indefinitely. In general, that's true, If the inflation rate today is running at x%, US bond markets tend to assume that in 5 or 40 years it will also be running at x%. At least that's how I'm interpreting your statement that "the bond markets mindlessly predicted that five years later in 2014 it would also be negative (blue line)."
But then you fail to note or comment on the fact that with today's actual rate reported to be 3-8+% (I'm not going to engage the controversy over what figures to use), the bond market is NOT projecting those amounts into the future. Quite to the contrary, long-term interest rates have hardly changed at all.
One of the usual answers is that the Fed is manipulating that through QE, but unless the Fed is actually INCREASING its QE purchases (which to the best of my knowledge it is not), that explanation fails. The rate of inflation has increased, the QE purchases have stayed the same (or actually diminished) and bond market interest rates are NOT rising.
In short, the usual "mindless" process is NOT happening right now. I'll leave to others the task of interpreting that fact, but I think it's well worth noting.
Bond yields are sorta useless. Once the Delta slop falls and as seen in India and the UK, it will likely be sudden and rapid. Yields will likely spike.