David R. Henderson  

Bet on Inflation

PRINT
Scott Sumner Fights Dirty... For Economics Teachers...

Two days ago, I bet a former student of mine, Thomas Strenge, on the inflation rate for the next year. He's betting that the CPI-U between December 2010 and December 2011 (reported January 14, 2011 and January whatever, 2012) will be greater than or equal to 7.0 percent. I'm betting not. Amount of bet: $100.

There's been a lot of discussion about betting on this blog by Bryan and on marginalrevolution.com by Tyler Cowen. Why do I do it? Mainly because it's fun. But secondarily as a teaching tool. When my students make careless statements in class about future events or magnitudes, statements whose veracity I strongly doubt, I bet them small amounts ($20 or less) as a way of getting them to pay more careful attention. I don't know if it works, by the way. I'm not necessarily putting Thomas in this category because I think that 7% is at least a plausible bet. But I still think there's about a 65% probability that I'll win.


Comments and Sharing


CATEGORIES: Monetary Policy



COMMENTS (27 to date)
Joey Donuts writes:

@David

Don't get over confident.
CPI-U for Dec. 2010.

0.5 percent. Annual rate of close to 6%.

Gabriel rossman writes:

> But I still think there's about a 65% probability that I'll win.

Actually it would be a decent bet even if you thought the odds were slightly against you since if you lose you'll be paying out in inflated currency -- kind of a "heads I win, tails you lose" type situation. A real bet about inflation would be stated in real dollars, or better yet, canned goods.

MikeP writes:

But I still think there's about a 65% probability that I'll win.

Are you kidding?

Do you know how many numbers there are greater than 7 and how few there are smaller than 7?

Tom P writes:

Wow. Market inflation expectations are only 2.3%, and you win if inflation is under 7%? Nice work, David. You should be confident of the outcome.

(Market expectations of inflation: http://www.bloomberg.com/apps/quote?ticker=USSWIT5:IND)

Justin Rietz writes:

The Cleveland Fed has a nice web page that tracks inflation expectations: http://www.clevelandfed.org/research/data/inflation_expectations/index.cfm

Based on the data, I'd have to side with David.

John Hall writes:

65% are you joking? It's 95% at a minimum. Try to get him to bump it up to like $10,000.

RD writes:

Yeah, I think you've got this one in the bag. He surely could have bet someone giving a much lower number.

David R. Henderson writes:

MikeP writes:
Are you kidding?
Do you know how many numbers there are greater than 7 and how few there are smaller than 7?

There are many more, but that's relevant only if each has an equal probability. MikeP, do you think the probability of inflation being one million percent equals the probability that it will be 2%?

David R. Henderson writes:

Tom P and John Hall,
Good point. I should have said "at least a 65% probability." I was so confident of my bet that I didn't even check the data that Tom P rightly refers to.

In June 2008, I made a bet with a friend about pending inflation. I had read alarming statements by economists about what was going on below mainstream radar at the time. I bet that CPI would increase by 9% during the coming year.

Wow, was I wrong. Prices fell during that year!

But if you want to empathize with me you could agree that something BIG did happen. During that year my investments lost 1/3. But it was not inflation. Very educational.

Thomas Strenge writes:

Haha, we'll see David. This is certainly a bet I would prefer not to win as high inflation is not in anyone's interest. My bet is based on the following observations:
First of all, fiat currency is based on nothing more than confidence. Hence, inflation has the potential to become a self-fulfilling prophecy as more and more people loose confidence in the magic of Ben Bernanke.
MV=PQ, right?
Ben is forcing up M. Barack is discouraging Q from growing. So, I see the conditions right for an increase in V to drive up P.
Second of all, I see the warning signs. Gold is up. $100 oil is on the horizon (and that without significant US or European economic growth). Grocery prices are up as are commodity prices.
Third, yes prices have been down, but that's because producers have made themselves more efficient. There is a limit to that. For example, I currently work in the tire industry. Every producer is raising prices from Michelin down to the smallest Chinese producer.
Lastly, has the Fed ever accurately forecast the onset of inflation? ;-)

Philo writes:

@ T. Strenge:

"[H]as the Fed ever accurately forecast the onset of inflation?" You are probably right not to trust the Fed. But you should trust the market as a whole, and it is saying 7% is 'way too high.

MikeP writes:

There are many more, but that's relevant only if each has an equal probability. MikeP, do you think the probability of inflation being one million percent equals the probability that it will be 2%?

Sorry it wasn't clear, David, but I was joking.

I too think this is a safe bet. I'd think a good guess is that the expected inflation rate is normally distributed around 3.5% with a standard deviation of 3.5%, giving an 84% chance that it is less than 7%.

And don't forget that you win in the event of catastrophic deflation too -- and then get paid in more expensive dollars!

Troy Camplin writes:

I'd get David to include food and energy, as they are rather conveniently left out of inflation calculations.

Six months ago I paid $0.99 for a gallon of milk. A few days ago I paid $2.34. That's 135% increase.

I must say, though, that if the Fed increases the money supply by 7% or more, then expecting 7% inflation isn't irrational. Whether it all occurs in the next year is another issue.

ed writes:

"I'd get David to include food and energy, as they are rather conveniently left out of inflation calculations."

They are not left out of CPI-U. This is a common misconception. Please stop spreading this false belief!

They are only left out of "core inflation," which is a tool for macroeconomic forecasting only and is not used for anything else (like indexing of benefits for example).

David R. Henderson writes:

ed is right about the CPI-U. It includes food and energy. For some reason, the media tend to report so-called "core inflation" but that is not the standard measure of inflation.

Vince Cate writes:

I think we have at least a 65% chance of hyperinflation starting in the next year. By this I mean more than 5% inflation in the month of December 2011 alone.

The conditions for hyperinflation are debt over 80% of GNP and deficit spending over 40% of total government spending. With these conditions you usually only get a couple years till everyone stops buying government bonds. Then the government has to print for the full deficit and also all the bonds coming due. Since the government has been looking risky for awhile, people have moved into short term bonds. So huge numbers of bonds come due. The US is already monetizing about $100 billion per month with a deficit of around $100 billion per month. As fewer and fewer people "roll over" bonds the Fed has to buy more. Again, many short term bonds will come due very soon. As inflation and inflation expectations go up and fewer people will buy bonds and spend their money faster, so the velocity of money will pick up. There is a feedback loop of less bond purchases and higher velocity of money causing higher price inflation and higher price inflation causing less bond purchases and higher velocity of money. This feedback loop is hyperinflation.

joe cushing writes:

Not everyone will be hurt by inflation. Only owners of fixed rate debt and paper cuurency will be hurt. If we had 1000% inflation this year, my income would increase around 10 times in nominal terms but my upside down mortgage balance would would stay the same. I could pay the thing off in 3 months. Of course, it would be hard to borrow money again. Lenders would require higher rates for years after the inflation stopped. I supose that hurts the ehole economy.

allen writes:

I'd be interested to hear the rationale for the prediction and why David is certain enough to put his money where his mouth is.

Adv writes:

"If we had 1000% inflation this year, my income would increase around 10 times in nominal terms"

Do you get paid in something other than dollars?

What's to say that your empoloyer (or customers if you run a business or are otherwise self-empolyed) will increase your pay in commensurately with inflation of 1000%?

If you're salaried, you'll likely be making the same dollars per month and your mortgage will continue to be the same dollars per month.

Floccina writes:

Betting, isn't that illegal?

Troy Camplin writes:

Thank you, ed, for educating me. There is much I am still learning about economics.

However, your statement that I should "Please stop spreading this false belief!" implies I am doing so out of something other than ignorance, which is very ungenerous of you. One should assume less than vicious motives until proven otherwise.

P. Binder writes:

If I were Mr. Stenge, I would feel more confident if there were an honest judge like John Williams. The government will find a way to massage the numbers into utter nonsense and to deny the reality staring us in the face.

8 writes:

Rising energy and commodity prices accompanied growing money supply and growing work forces around the world, as the Boomers came of age.

This time, the Boomers are retiring. The young work forces are in Indonesia, Vietnam, the Middle East; the rapidly developing economies in China, Brazil and India. Which also happens to be where the inflation is hitting hardest. In the West, higher commodity and energy prices, along with high unemployment, won't lead to wage hikes, it will lead to consumer retrenchment and more deflation. (Assuming no collapse in confidence that triggers hyperinflation.)

Vince Cate writes:
(Assuming no collapse in confidence that triggers hyperinflation.)

But why would you assume that when conditions are set for hyperinflation?

Thomas Strenge writes:

Reflecting on some of the comments, I was struck by some occupational differences. I believe the underappreciated truism that businessmen make poor economists. I find it confirmed by some of the nonsense very successful people such as Warren Buffett and Bill Gates send out into the world as a profound insight. That said, it appears to me that many economists are hamstrung by their reliance on statistics which is a) always backward looking and b) only useful if there is trend continuity in the future. Businessmen, on the other hand, have their greatest success when they guess correctly about a future discontinuity.

As for deflation, I don't see any significant problem with a gentle gain in the value of money, especially since the steady devaluation going on since 1913. Indeed, a case could be made that the productivity revolution of the 1990s should have produced some deflation. Therefore, the Fed's gentle inflation during that time not only robbed us of the approximately 2% annual inflation witnessed, but maybe an additional 2% along the way in foregone gains from deflation.

Silas Barta writes:

I should have mentioned this earlier, but ...

The odds that individuals give each other in bets like these is almost certainly different from the odds on the outcome implicitly given in the derivatives markets. So anyone making a bet like this with you can probably get a dutch book by making the opposite bet with the big boys, even if it's a stupid position for them to hold in their bet with you.

I wonder if Julian Simon did that...

Comments for this entry have been closed
Return to top