Bryan Caplan  

Keynesian Bets Bleg

PRINT
The Pyramid of Macroeconomic I... Adam Smith on the Public Choic...
The strongest evidence for Keynesianism, in my view, is introspection.  During the last five years, however, I've often heard Keynesians claim that their views have been vindicated by events.  To which I'm sorely tempted to respond:
It's easy to claim "vindication by events" when your predictions are not only vague, but consistent with the main competing theories.
How could Keynesians convince me that events are really on their side?  The usual: Bets.  Find someone on "the other side" who disagrees with you, hammer out precise terms, and publicly shake on it.

I am not aware of any such Keynesian bets.  But I could easily have overlooked some.  Have I?  The closest thing to Keynesian bets of which I am aware, strangely, are the inflation bets that David Henderson and I made with Bob Murphy.  But I doubt that leading Keynesians want to accept EconLog bloggers as their intellectual representatives. 

Hopefully someone will point out some pertinent bets that I've overlooked.  In any case, though, the quantity of Keynesian wagering is not fixed.  My challenge for Keynesians: If you really believe in your own predictive acumen, could you please propose some bets?


Comments and Sharing





COMMENTS (14 to date)
Bostonian writes:

Economists make forecasts, collected in for example the Survey of Professional Forecasters. You could look at whether some schools of economists forecast better on average than others. It would take some judgement to attribute economists to various schools of thought.

Money managers are constantly making bets that depend in part on the economic outlook. What views of economics do the most successful ones have?

Roger McKinney writes:

Bostonian:

"What views of economics do the most successful ones have?"

Check out Mark Skousen's "Structure of Production." He shows that many follow the main points of Austrian econ without knowing it.

Vladimir writes:

Bets for trifling amounts are better than ordinary arguing, but why stop there? Any meaningful non-trivial prediction about future macroeconomic trends should be straightforward to translate into a market-beting, EMH-violating investment strategy.

Therefore, I will be convinced only when I see macroeconomists betting their savings on the predictions of their theories, not just trifling sums.

Rick Hull writes:

Bostonian and Consistent have missed the point: a bet is substantially different from a forecast. Both opposing sides must agree to the terms, guaranteeing the embedded forecast is significant, i.e. controversial. As well, the terms must be so specific, a well as victory conditions, such that each side sees a fair chance of winning.

[Giving up on the remainder of this comment. This commenting form is not playing nicely with my android phone browser. ]

Matthew Martin writes:

A potential problem with asking people to bet on their predictions is that you are asking them to make an unconditional bet on a conditional forecast. That is, keynesians do believe we could have hyperinflation in the near future if the government starts a massive stimulus program that pushes us back to full employment, or if there is a massive negative supply shock that reduces output potential (think asteroid strikes north america, hyperinflation ensues).

Now, you can potentially bet on conditional forecasts, but since both Keynesians and anti-keynesians are making conditional forecasts, they aren't guaranteed to be complementary--it will be possible for events to unfold in a manner such that neither party wins.

And if one party is making a conditional forecast but the other party is making an unconditional forecast, then it simply isn't a fair bet.

Consistent if nothing else writes:

[Comment removed for supplying false email address and for multiple policy violations. Email the webmaster@econlib.org to request restoring your comment privileges. A valid email address is required to post comments on EconLog and EconTalk.--Econlib Ed.]

Bob Murphy writes:

But I doubt that leading Keynesians want to accept EconLog bloggers as their intellectual representatives.

Bryan and I must read different leading Keynesians.

Doug writes:

"Money managers are constantly making bets that depend in part on the economic outlook. What views of economics do the most successful ones have?"

Most money managers make their alpha on stock specific predictions. Very few managers consistently make money from macro-based views.

In this realm George Soros is definitely the most successful macro-themed manager of all time (if not the most successful manager, period). If you read anything by him, his economic school of thought is at best highly unorthodox, at worst vague and incoherent.

The problem with picking successful managers is that you'll achieve much higher risk-adjusted returns from getting your short-term bets right, than getting your long-term bets. You can achieve much higher returns if you can successfully predict tomorrow's stock market move everyday, rather than next year's every year.

So a lot of great macro managers don't really make great forecasters. For example if you read Soros' predictions quite a lot turn out to be way off the mark. But what makes him a great trader is that he very quickly flips his opinion and gets the timing right. So when he's wrong he doesn't really lose that much money, and when he's right he makes a killing.

Vito writes:

A bet with Krugman would be the most widely reported and get the most attention from the most people... most likely. You, or someone, should publicly challenge him.

Himanshu Sanguri writes:

Dear Bryan, I am also sorely tempted to respond on your request for bet :-) by a comment made by Ruchir Sharma in his work "Breakout Nations". It goes like, in past, experts used to do a large number of predictions, out of them, any one succeeds, and experts pat their own backs. These days, experts have taken the new route of predicting things, whose outcome takes a longer time, and when we shall reach at result time, most us either would not exist, or will have forgotten the predictions. I have been critically analyzing, how to map Keynes doctrines to present day globalized economies, where employers, employees and consumers are floating across the globe. Today, demand and supply has no uniformity in terms of demography, economy and governments. Anyways, I will wait for the bets placed.

Rick Hull writes:

To help get things started, I propose a side bet:

No one will make a bet, representing the Keynesian side, that Bryan will accept the other side of, in the comments of this blog post, within 1 week. Stakes: $1

i.e. for me to lose this bet, a commenter on this blog post, claiming to represent the Keynesian viewpoint, would have to propose a bet for which Bryan would accept the other side. This would have to happen within 1 week.

Anyone care to take my bet? I am happy to negotiate alternate terms. Note that it would be relatively trivial to win this bet if you've already got something in mind to propose for Bryan.

Arthur_500 writes:

Betting on the future is done daily by those who believe in horoscopes. My favorite is a sign found on the Internet (where everything is real). The handwritten sign is in front of a hotel sign for a physic convention and it reads, "Conference cancelled due to unforeseen circumstances."

Keynesians claim that the economy would be gangbusters today if the gubment had 'invested' more heavily in TARP, etc. But when the data shows the programs that did exist didn't do diddly squat they claim it is due to mismanagement or some other excuse.

It must have been unforeseen circumstances. Want to bet?

Jacob A. Geller writes:

I'm not a Keynesian, but I'm willing to bet that 1) the budget deficit shrinks to 2% of GDP by January 1, 2015, 2) unemployment fails to fall below 6.5% by the same date, and 3) inflation fails to reach 3% by the same date.

All three conditions must hold in order for me to win the bet. Loser buys lunch for the winner.

I'm open to negotiation, of course, and maybe I'm also willing to bet on a fourth condition having to do with real supply shocks or monetary policy or something. I'll consider adding any condition(s) that improve the bet's ability to rule out non-Keynesian explanations for the coexistence of a shrinking fiscal deficit, high unemployment, and low inflation over the next ~18 months.

In case you're wondering why a non-Keynesian would bet on Keynesianism, it's the same reason Noah Smith bet against high inflation: worst-case scenario, I have to buy lunch for someone I admire and respect.

Jacob A. Geller writes:

PS -- This is a little difficult, but a sophisticated alternative bet might involve a sort of "reaction function," or vectors of deficits, unemployment, and inflation rates. E.g.

1) "if deficit = 2% of GDP, then conditions #2 and #3 are...",
2) "if deficit = 2.5% of GDP, then conditions..."
3) "if deficit = 3% of GDP, then..."

Comments for this entry have been closed
Return to top