Bryan Caplan  

The Mankiw-Krugman Non-Bet

PRINT
Don't Miss the Next Stossel... Buffett on Mark to Market...
Here's news I can use: Krugman talked smack about Mankiw, Mankiw politely proposed a bet, and Krugman ignored his offer.  Which reminds me - I'm still waiting for Andy Kessler to accept my bet on the bailout.

I have a dream that one day, people who refuse to bet on their statements will be viewed with greater contempt than those who bet and lose.  Who's with me?


Comments and Sharing





COMMENTS (19 to date)
Richard writes:

Bryan, I'll bet you that that would never happen. Of course, from my perspective that bet is heads I win (money), tails I win (public regard).

caveat bettor writes:

Why would Krugman ever want real risk-adjusted returns next to his name? What's the point of staying in the ivory tower?

Zac writes:

I'm with you, for the most part. If you bet, and lose, and change your mind, that's respectable enough. If you bet, lose, refuse to change your mind about the issue and act like a sore loser, you might as well have not bet at all.

Also, most of these sorts of bets are "gentleman's bets" where some token amount is offered is essentially meaningless. When someone says they'll bet $1000 or more on a claim, that makes it interesting. Men with $100K+ incomes making $100 bets means nothing.

Charlie writes:

"How can you fail to acknowledge that there’s huge slack capacity in the economy right now? And yes, we can expect fast growth if and when that capacity comes back into use."

Krugman never defends the administration's growth pronouncements, he is making a more academic argument about unit roots. This is obvious since he says "if and when" that capacity comes back.

Suppose Krugman decided to attempt to win this bet with Mankiw. PK would probably spend a lot of time going through various growth forecasts, he'd spend a while thinking about different policy changes that might effect these forecasts here and abroad, then he'd think about the robustness of his results.

Then he would write to Mankiw with a lowball proposal, most likely Mankiw would reject anything that isn't the exact administration view, bc it requires so little effort, but suppose he plays along, he'd have to form a similar forecast. Through a process similar to PK, he would come up with a forecast. It could be the same, but suppose it is different. Through a haggling of back and forth they would come to a bet.

Both of their time is very valuable, so to make this process worth it for both people the bet would have to be very large. The closer the two's beliefs, the larger the bet has to be. Since both are probably risk averse, it's likely there is no acceptable bet. For both people to believe they are earning an acceptable profit, the bet probably has to be a large part of their portfolio, and of course that is money that could be invested in other ways and spread accross risks.

So in the end, a lot of time and energy would get wasted on a bet that would probably never get made. And since both people probably have substantial variability in the paths the future may follow, the outcome of the bet would give us very little info about who's intellectual paradigm is more accurate.

Zhaofeng Xue writes:

Of course I am with you. Sometimes you don't even have to suggest a bet. Recently I had a number of email exchanges with someone who disagreed with me in a confident way, sort of like "Though I am a layman, I am sure ... ." But when I suggested to put his name down on my post, he, prudently, said "No."

jb writes:

Totally agree, Bryan - if Krugman won't take the bet, or at least explain why he thinks the bet is unfair, odds are he isn't confident enough in his position to risk the possibility of admitting a mistake.

And if he's not confident in his position, why should I be?

Mankiw's the clear winner here. And if the bet were the other way - if Krugman had proposed it, and Mankiw ignored it, even though I kinda detest Krugman, I would have had to accept that he at least had some conviction behind his beliefs.

Robin Hanson writes:

I'm with you Bryan!

Neal W. writes:

I have an unrelated question.

Why is it that people in general think they know more about economic theory than they really do? Given that economics is not a course that one typical gets in public school (and never pursue it on their own), you'd think that people would be more skeptical of voting over economic issues than they are.

I thinks it's because economics is deceptive in that it seems simpler than it really is. Everyone would find it weird if presidential candidates had as part of their platform, how to calibrate the mass spectrometer, or took position stands on controversial topics in quantum mechanics, and then people voted based on those issues so that the candidate could implement policy based on that.

Yet, people have no problem voting to impliment economic policy for which they have no understanding.

Taimyoboi writes:

"...people who refuse to bet on their statements will be viewed with greater contempt ...."

I'm not sure how Krugman could be viewed with any greater contempt nowadays. The absolute value of his signal-to-noise ratio has to be less than epsilon by now.

beamish writes:

No. We should only hold people in contempt for their wickedness.

James A. Donald writes:

Bets are a poor indicator of the facts, because large bets are an accurate indicator of people's propensity to weasel, lie, emit a fog of irrelevant complications, and just plain not pay when they lose the bet.

We had this system on the extropians list, and it was a complete disaster, with each large bet leading to an inky fog of complexification to disguise the fact the loser was welshing. No one ever paid, because the honest people only made bets when they had insider knowledge or specialized knowledge, such that they knew the bet was sure thing - so honest people never lost a bet, and losers never paid.

Arare Litus writes:

I'm with you.

Graeme Bird writes:

Present company excepted of course. But this idea of betting is getting to be a real menace since it falls into the wrong hands. I see this as an actively harmful practice. No doubt the people here would have won their bets and the bets relate to something that would prove their point.

But too often we are seeing an endless round of proposal of bets from other people as a substitute for having a reasoned argument. Its just another distraction to the academy, which ought to be weeding out bad doctrine and sorting out the good stuff. Which in the end will bring you to the conclusion that George Reisman is right about pretty much everything and is the ultimate arbiter in your profession.

Its pretty important to get things right. The replacement of REASON by side-bets is part of a very big problem. Your country has become BarryLand. Cook County and surrounding districts. And you are being scuttled and if you don't move fast you will be destroyed and you will see friends die with blood in their mouths.

The idea is not to make bets but to attack the idiotic notion of the Keyensian multiplier head on, get serious about closing down bureaucracies by the bushel, and try and stop this mass-looting and coming collapse.

I never thought I'd ever see anything this sad. Our great ally destroying itself at break-neck pace.

Kurbla writes:

I agree with James, Arare and Graeme - this betting thing is at best, unnecessary complication and distraction, the generator of negative emotions with carry over into unrelated fields of professional life. At worst, from time to time, you'll see police and courts dealing with resulting problems.

John Fast writes:

Charlie wrote:
Suppose Krugman decided to attempt to win this bet with Mankiw. PK would probably spend a lot of time going through various growth forecasts, he'd spend a while thinking about different policy changes that might effect these forecasts here and abroad, then he'd think about the robustness of his results.

As an economist I believe that Krugman would do that even without a bet. In real sciences, any scientist who didn't gather evidence, run calculations, and then make a prediction based on his model would be scorned and mocked, and rightly so.

Economics will be a real science -- comparable to, say, physics after Galileo -- "if and when" that starts to happen in it as well.


Neal W. wrote:
Why is it that people in general think they know more about economic theory than they really do? Given that economics is not a course that one typical gets in public school (and never pursue it on their own)

I believe you just answered your own question.

It's exactly as you said, combined with or even caused by the fact that there is little or no consequence for people in general being right or wrong about economic theory. (Which Bryan pointed out in his book.)

Joshua Lyle writes:

Anti-anti-betting:
The problem is that we've seen some empirical studies suggesting that betting can actually work quite well as a way of arriving at truth. It is a way of weeding out bad doctrine, sorting out good stuff, and at best is not just a complication and distraction but a useful tool in aggregating and improving the quality of information. Criticisms about welshing, negative emotions, and the like are spot on, but they don't attack the heart of the matter and suggest that we need to improve rather than abandon betting systems, and certainly also apply to standard discourse, which doesn't lack for face-saving and emotional recrimination. If you can demonstrate a better way, please share your results because I'm keen to add to my toolset.

Kurbla writes:

Lyle, if you post the link to data, I'll comment it.

Joshua Lyle writes:

Kurbla,
I'm currently unable to do much research, but I'd point first to the Iowa Electronic Market.

mjh writes:

I'm with you. But you realize that this means that Paul Ehrlich is thus more respectable than Paul Krugman. Hmmm...

Comments for this entry have been closed
Return to top