Scott Sumner  

Quantum finance

PRINT
I'm a Liberal... Facts...

The New York Review of Books has a very interesting interview of George Soros. At one point he is asked about the recent revolution in the Ukraine. Here is part of his response, and then a follow-up question:

[Soros] Contrary to all rational expectations, a group of citizens armed with not much more than sticks and shields made of cardboard boxes and metal garbage can lids overwhelmed a police force firing live ammunition. There were many casualties, but the citizens prevailed. It was a veritable miracle.

Schmitz: How could such a thing happen? How do you explain it?

Soros: It fits right into my human uncertainty principle, but it also reveals a remarkable similarity between human affairs and quantum physics of which I was previously unaware. According to Max Planck, among others, subatomic phenomena have a dual character: they can manifest themselves as particles or waves. Something similar applies to human beings: they are partly freestanding individuals or particles and partly components of larger entities that behave like waves. The impact they make on reality depends on which alternative dominates their behavior. There are potential tipping points from one alternative to the other but it is uncertain when they will occur and the uncertainty can be resolved only in retrospect.

On February 20 a tipping point was reached when the people on Maidan Square were so determined to defend Ukraine that they forgot about their individual mortality. What gave their suicidal stand historic significance is that it succeeded.


I was intrigued by the quantum mechanics analogy. At the same time I wasn't entirely convinced. I'm certainly no expert on quantum superposition, but I believe the claim is that something exists in two states simultaneously. The idea is so paradoxical that some physicists believe the universe splits in two when an observation is made. Rather than being random, we can say that both outcomes happen deterministically, but in different universes. (Please correct me if this is wrong.)

I don't find it all that paradoxical that people sometimes behave like individuals and at other times behave like members of a mob. But I do find the Efficient Markets Hypothesis to be somewhat paradoxical. This hypothesis says that individual traders cannot know more than the market, and hence deviations of assets prices from trend are essentially unforecastable. Consider the following example:

A very knowledgeable trader walks onto the floor of a market where gold is actively traded. To make things simple, let's assume that there is no risk premium, and that the futures price equal the expected future spot price. Suppose October 2014 gold futures are trading at $1300, but this very smart trader believes that gold will be trading at $1360 in October. Is this information valuable?

In one sense the answer is no. The market forecast is freely available, and is much more reliable. But in another sense the answer is yes. If the trader decides to act on his knowledge, the market will become even more efficient. The market price will immediately represent a more accurate forecast of the expected future spot price.

The average opinion of traders is dumb; the average of their opinions is brilliant. (Have I used 'average' the same way both times?)

Alternatively, let's suppose markets are efficient. What makes them efficient? Lots of traders who dig up information on the value of gold. And why do they do that? Because they believe markets are inefficient.

One of my favorite examples is the joke about the two university of Chicago professors walking along a sidewalk. One says "Look, a $100 bill." The other replies, " That's not possible, someone would have picked it up already." When I tell my students the joke, I ask them whether it is a pro-EMH joke or an anti-EMH joke. But here's the problem, I'm not quite sure myself. Consider the following two interpretations:

A. The EMH represents Truth with a capital T. In that case the joke seems to be mocking those who believe in the EMH.

B. The EMH is a highly useful theory. For philosophical pragmatists it doesn't get any better than that. In that case the joke seems to be supportive of the EMH.

I prefer B. I don't know about you, but I often see money lying on the sidewalk. But it's always coins or small bills. Not once in my life have I seen a $100 bill. And now the EMH tells me why I shouldn't waste my time walking around fruitlessly hoping I'll find a $100 bill down there. I should look up at the birds and the trees and the clouds. I should enjoy life.

But then who will pick up the $100 bills? Two groups of people. Those who foolishly think they will be able to find them if they look hard enough. (Oddly, some of those foolish people will find them.) And those who know it's a very long shot but enjoy treasure hunting. Or enjoy looking at sidewalks. Some of them will also be successful. Especially those who know where to look---like very early in the morning outside an expensive nightclub. Or in an area where there are lots of drug deals. There are enough people like that to keep sidewalks cleared of $100s.

I enjoy collecting old prints. And because I enjoy doing so, I help to keep the old print market efficient. But not sidewalks, and not stock markets. Someone else will have to do those jobs.

PS. I recommend the Soros interview---most of it is much better than the passage I quoted.


Comments and Sharing


CATEGORIES: Finance



COMMENTS (19 to date)
Andrew_FL writes:

You've neglected the third group-those who would not as a general rule go looking for $100 bills but happen to notice them by chance. I think you'll find most hundred dollar bills dropped on the ground are picked up by those folks than people who were intentionally looking for them.

About the Chicago story, it is not clear to me whether the second professor is supposed to be serious in his assertion. Perhaps he is just joking with his colleague. In that case we read a report of the telling of a joke, and not a joke intended to make the reader laugh.

But supposing it is intended to make the reader laugh, then I would think it would work as you said Scott, with B but not with A.

I have made an informal study of humor. Humor, unless I am mistaken, is about coming to grips with unpleasant truths. When we laugh we are facing some pain or disappointment in our situation, and simultaneously getting beyond that pain or disappointment.

A theorem falls out of my study of humor: You can not crack a joke without risking offending someone.

With the Chicago joke, a person who believes interpretation A (EMH is unshakably true) either will not get the joke or, if he senses he is being mocked, will probably be offended.

About the models of quantum physics, I suppose the duality derives from the poverty of our models to date. That is, with our experience we have made some pretty good models about particles and how they behave, and other pretty good models about waves and how they behave. Our models satisfy and charm us. Then we discover a new thing which does not fall entirely into either model. So we accuse the thing of duality.

This could be a joke. It makes me laugh.

BC writes:

In today's version of the $100 bill joke, the second professor would not be telling the first not to bother picking up the bill. He would instead say, "Be careful! If you pick that up, you will be taking a lot of hidden risk."

happyjuggler0 writes:

About two years ago I saw what looked like was a $100 bill folded in quarters on the sidewalk on a residential street in LA. In broad daylight, during the afternoon. First time I saw anything larger than a small bill, which itself was quite rare to see lying around.

I was aware of the EMH joke, which was my first thought.

My second thought was that it might not indeed be real because it was folded; I thought someone might be playing a practical joke. I looked around but I could see no one looking at me; indeed no one was anywhere near me.

So I picked it up, unfolded it and looked at it. It was indeed real! I looked around again, and no one was patting their pockets or anything; indeed as I mentioned no one was really around at all. So I pocketed it.

By the way, I wasn't looking for money on the ground; it just stood out as I approached it. Andrew_FL wins.

P.S. I don't believe in the EMH; I believe in something I call AIMH, the asymmetric insight market hypothesis. Someone has to have the first correct insight to events/forcings with implications that are not obvious at first sight. I see no reason why some more insightful players can't be right before the "mass institutional money" figure something out with enough consistency to make very real gains, or alpha if you will.

ChrisA writes:

@happyjuggler0
Your explanation is similar to mine. There are certainly arbitrage opportunities available in any market, but they are not predictably available. So the people that come across the arbitrage opportunity take it, which makes the market efficient. But, on average, the cost of actually looking for arbitrage opportunities is above the value of the arbitrage opportunities that you will find.

This also explains some of the success of investors, when they are able to make money seemingly for extended periods of time. Its not just that they are lucky in the sense that they are making random bets that work, they really have found an arbitrage that works in one particular area. But when they move to another area of investment, they have the same issues as everyone else. So even real "beta" is luck in some way. This is perhaps like the John Paulson case where he was able to find a really good arbitrage by betting on housing loan defaults in 2008. This was a real market inefficiency which he took advantage of. But as his future success has shown, he could not repeat this. He was lucky not in that the bet he made had casino like or lottery type odds, with a negative expected value that happen to payoff against the odds, but that he found the bet and had the capital and connections to profit from it.

Another way of looking at this is that yes, there really can be $100 bills left on the sidewalk so you should pick them up if you see them, but it is not worth going into the "looking for $100 bills left on sidewalk" business.

Steve Sailer writes:

Soros sounds like he's mocking the interviewer.

Scott Freelander writes:

Scott,

The paradox that matters involves a supposed lack of arbitrage opportunities versus the need for arbitrate opportunities to keep smarter investors in the market. If smart investors leave the market, then there are more arbitrage opportunities. As Friedman pointed out decades ago that this puts a limit on the efficiency of markets.

Jack PQ writes:

I usually interpret the Chicago professors' story in terms of the Grossman-Stiglitz paper (costs and benefits of information). Basically, the odds of finding a bill are inversely proportional to the value of the bill (or coin). You often find pennies, nickels, dimes, sometimes a dollar bill. But a $20? a $100? Almost never.

Re: Soros, have you read Rudiger Dornbusch's review of one of Soros's books? The review is in a Dornbusch collection of essays. It's spot-on, somewhat harsh, but not cruel.

Chase writes:

Very interesting post!

David R. Henderson writes:

Scott,
I’ve never looked for a $100 bill, for the obvious reasons, but, with my Kirznerian entrepreneurial alertness, I did find one once.

ThomasH writes:

There is nothing about efficient markets that says that a participant may not have non-public information (or a non-public way of processing public information) that may not allow her to beat the market.

ThomasH writes:

Not entirely on topic, but I think that the citizens of the UK ought to have small shrines to Soros in their houses for having kept them out of the (precursor to) the Euro. Yet I'm not aware of him being that popular there.

Lorenzo from Oz writes:

The Soros interview has the great virtue of pointing the finger at the ECB, which remarkably little commentary on the Eurozone crisis does.

Ian writes:

I have a bachelor's degree in Physics, so I'm by no means an expert, but I do know about the wave-particle duality.

It's not magic, and there's nothing in the theory that has anything to do with alternate universes. The wave-particle duality is counter-intuitive because it looks like quantum wave-particles interfere with themselves like waves, but they're discretely detected like particles. This phenomenon is very well established and described mathematically. If you can get your head around the double slit experiment, it'll get you pretty far down the road of understanding the phenomenon.

The alternate universe explanation is fun, but it's just an explanation to try to get around the counter-intuition.

Scott Sumner writes:

Ian, I'm obviously no expert here, but there are extremely bright physicists who believe the multiple universe explanation most certainly is necessary.

Everyone, lots of good comments, I agree with most. Later today I'll do a post explaining why I think the EMH is not just strange, but also truer than supply and demand.

Regarding Soros, I view him as being extremely bright, well intentioned, right on most non-economics issues (drugs, human rights, militarism, etc) and right on some economics issues like the ECB, and wrong on others, like fiscal policy.

Joel Aaron Freeman writes:

For anyone who skipped the interview, I am not a fan of Soros, but the interview is quite good.

Mike Rulle writes:

Need to stay short or will write all day. EMH deals with a premise of expected value of zero outperformance and a random walk around zero. Hence luck can obviously find $100 bill, but as others said, it's not a reasonable business. According to Vanguard over any 15 year period 32% of active managers outperform index. But expected value is negative among all active managers (surprisingly interesting caveats by Vanguard and Sharpe on this point). And like hot streaks in basketball and baseball they are not predictive---i.e., one cannot figure out who will outperform next year. (have to admit, 32% was higher than I thought it would be, even if not predictive).

Traditional risk free rbitrage has basically gone away.

Equity markets have not followed a pure random walk since 1987---close--- but it veers enough. Example: since 1987 crash, rising volatility has coincided with declining prices at extremes. There is evidence of conditional probability. Target constant vol a long only S&P contract and the return/drawdown ratio rises significantly at same average vol as the underlying S&P without target volatility. If enough traders do this, it will go away. Equity market exhibits serial correlation in volatility.

EMH folks dealt with markets as they are, not how they became what they are. There does seem to be a paradox in EMH. If everyone followed the passive index, what would its price be? I guess markets would just trade the index itself and not the stocks in the index. Seems weird. But EMH people do not believe you need active people to create proper index value.

Soros. He calls his fund the Quantum fund. I could not get page 50 of one of his books it was so bloated. What I thought he meant by quantum in trading is that the act of trading itself changes values, hence the act of trading alters the act of analyzing markets---which struck me as an absurd premise---because the sum of markets is the sum of trading. But he was analogizing the split screen experiment mentioned by Ian, where the act of measuring changes the outcome in some random way.

At the extreme Soros makes a point---like when he cornered the pound. But this kind of trading happens all the time----where two groups of big money try to get the other to drop out first and give it up to the other guy. Examples---Ameranth and nat gas, The "Whale", Citadel and the 10 year treasury in 2006 or so, 1980 Silver move by the two brothers whose name I forget---etc.

Sticks against rifles? Really?

Chris Kerr writes:

It seems to me that the EMH in Economics is the equivalent of the Null Hypothesis in the natural sciences. You start off by assuming an efficient market. Sometimes you come across a situation where the evidence is not consistent with an efficient market. If you can work out a way of modelling and exploiting the inconsistency then eventually the content of the EMH is updated to make it consistent i.e. the market becomes efficient.

Similarly quantum mechanics started off as inconsistencies between experimental data and the null hypothesis. After various people came up with theories that could explain the observations, the "null hypothesis" changed - e.g. for the discovery of the Higgs Boson the null hypothesis was all of quantum mechanics up to now but without the Higgs Boson.

https://xkcd.com/892/

Comments for this entry have been closed
Return to top