Are markets efficient?

[Music] many economists View financial markets as efficient with prices incorporating all available information about future values behavioral economists say that model simply doesn't reflect how markets actually work so are markets efficient welcome to the big question the monthly video series from Chicago Booth review I'm Hal whitesman and I'm joined by an expert panel Eugene farmer is Robert Arma cormi distinguished service professor of Finance at Chicago Booth well known for his empirical analysis of asset prices and for developing the efficient market hypothesis he was the joint recipient of the 2013 Nobel Prize in economic science and

Richard theor is the Charles R Walgreen distinguished service professor of Behavioral Science and economics at Chicago Booth he's director of Booth's Center for decision research and co-author of the bestseller nudge his more recent book is misbehaving the making of behavioral economics panel welcome to the big question Gene farmer let me start with you uh you came up with this efficient market hypothesis so tell us briefly what is it well it's a very simple statement that prices reflect all available information testing that hypothesis turns out to be more difficult but it's a simple hypothesis in principle

okay Richard theor do you agree that market prices reflect all available information well uh like Jean says uh it's easier to say than to test and I like to distinguish two aspects of it uh one is whether you can beat the market that's the one most people are most interested in and the other is whether prices are correct so if prices reflect all information then they should land on the right price and we we can separate those two questions uh because they're different okay but the B basic premise about the containing information is something you

don't agree with then it sounds like it's almost impossible to test that hypothesis um ex except through the the two questions that I've asked can you beat the market and are prices right okay is that right Gan farmer does if you say that prices reflect all available information it necessarily means that a price is right at any particular point in time that's the statement of the hypothesis but it's a model it it's not completely true no models are completely true they're approximations to the world the question is for what purposes are they a good approximation

um as far as I'm concerned they're a good approximation from almost every uh purpose I don't know any investors who shouldn't behave as if markets are efficient for example uh and there are all kinds of tests with resp respect to the response of prices to specific kinds of information in which the hypothesis that prices adjust quickly to information looks very good there are others where it looks less good uh so it's a model it's not entirely always true but it's a good working model for most practical uses okay is that right is it a good

working model well uh again I think for the first part can you beat the market I think Jean and I are in uh virtually complete agreement uh which is that's a good working hypothesis for any investor uh does that mean you should assume any individual investor should assume that markets are efficient behave as if as would be the way you put it behave as if uh well yeah certainly there's evidence going back to the thesis of Mike Jensen who was I guess maybe one of Jean's first students was around at the same time who did

a thesis on whether mutual funds on average beat their benchmarks and have you account for their fees they don't that was in the 60s it's been updated a zillion times and it you know you can quibble about exactly how to do it but that's uh approximately true so uh just investing based on fees is not a dumb thing to do uh um regardless of the nuances that Jee and I might get into okay but you talked earlier about the in some cases you know the model Works in other cases it works less well and in

in your book um Richard thy you talk about the 1987 crash Black Monday and um that was you give that as an example of how prices are not right the efficient markets don't really work you say if prices are too variable they're in some sense wrong it's hard to argue the price at the close of trading on Thursday October 15th and the prize at the close of trading the following Monday more than 25% lower can both be rational measures of intrinsic value given the absence of news but isn't the idea that efficient markets are supposed

to be unpredictable yes but unpredictable doesn't mean rational so you know I have a two-year-old granddaughter who runs around like crazy and I defy anyone to use a rational model to predict what she's going to do next so she will be unpredictable but uh her behavior isn't well captured by uh a a model of uh maximizing anything other than whatever she calls fun so is the market do Market behaving the same way as your granddaughter well uh sometimes I don't think anyone thinks that the value of the world economy fell 25% that day nothing happened

so if markets were efficient there would be a certain bounded level of volatility so well in the absence so it's not a day when uh World War II was declared but it was a time when people were talking about perhaps an oncoming recession which turned out not to not to happen so on hindsight this was a a big mistake but it needn't have been so in hindsight every price is wrong yeah so but that's 2020 hindsight and uh but what I would say is merely the big fluctuations that entire week two of the biggest up

dayss in history occurred that week and three of the biggest down days and nothing was happening other than the fact that people were talking about how markets are going up and down like crazy all over the world so that's one indirect way that we can measure market efficiency this was uh essentially the approach that was pioneered by uh Bob Schiller who uh Jean shared the Nobel Prize with and his argument was prices fluctuate too much um to to be explained by a rational process right and J Farm is that right that when there's a certain

level at which prices just fluctuate too much too much there's a test for that and the test says that uh when we look at longer periods of time the variance of the price changes should not grow like the length of the time period if there is all of this temporary variation in prices that's not rational and in fact that test does not indicate that there's temporary variation in prices so you've got to kind of come up with a different test uh shiller's model was based on the proposition that there's no variation to time and expected

returns but but we know there is a ton of variation in expected return so that that kind of branch of testing people lost interest in because you really can't come to any conclusions a straight test of whether there's temporary variation in prices says no there isn't you can't identify it um and another test which would be is there too much variation in expected returns to be attributed to rational Behavior well now you have to Define what you mean by by that and that's that's terbly difficult but is you argument then that that investors are constantly

changing their their view of the expected future value of of the sh my view is that risk aversion moves dramatically through time um in particular it's very high during bad periods and it's lower during during during good periods and that affects the pricing of assets and then the expected returns you expect and so every time there's a stock market crash people come to you or they they rubbish efficient the idea of efficient market markets and and point to the these Vol this massive volatility and say doesn't that prove that efficient markets are wrong and doesn't

prove that at all what it proves is what is that risk changes a lot very quickly well and it information changes a lot Through Time dick theor let me turn to you uh bubbles Gene farmer famously does not allow the word use of the word bubbles but I'm going to use it do do bubbles exist how do we Define bubbles well I I think it's hard to say I'm going to present two examples one of which is uh I think more convincing than the other the the first one is I have a graph that I

think you'll show the the viewers uh it was produced by um uh Jean's son-in-law John Cochran uh and it's a a graph of house prices over a very long period of time in the US and what it shows is is that for a long period of time house prices were roughly 20 times rental prices and then starting around 2000 they go up depending on which measure you use they go up a lot or they go up really a lot uh and then they go back down uh after the financial crisis now um the this I'm

not I can't use this graph to convince Gan that markets are inefficient true um because because the graph stops too soon uh well yeah but uh you know uh we're not going to live long enough to no no no I mean if you continue the graph it goes back to the peak well but so what's what's what's the bubble the down the up the St going down well uh okay so once again we're in agreement which is that uh studying data like this you it's impossible to know for sure whether something's a bubble there there

are hints in that graph that prices seem to have diverged from a level that had existed for a very long time they went up and then they went down was this because of irrational exuberance uh Alan Greenspan's uh phrase um what we do know is that in the markets like Vegas and Scottdale and South Florida where prices were going up the most expectations of future price appreciation were also the highest and that could be rational but is I'm skeptical of that and of of course in hindsight it was wrong but let me present another example

which is amusing at least um this is have I told you about this one the Cuba fund no okay this is good so there's a uh closed and mutual fund uh that happens to have the ticker symbol [Music] Cuba now um close and funds uh have been studied for many years there are a special kind of mutual fund where the shares trade in markets and the price of the shares can deviate from the value of the assets that they own so uh this particular fund uh although it has a ticker symbol Cuba of course cannot

invest in Cuba uh a that would would be illegal uh and B there are no Securities so it it's Holdings of Cuba are zero and for many years it traded at a discount of about 10 to 15% of net asset value meaning that you could buy a $100 worth of their assets for $85 to $90 which is a good bargain then um if we look at the chart all of a sudden one day the price skyrockets and it sells for a 70% premium and you can probably guess what happened that was the day that President

Obama announced his intention to relax relations with Cuba so a bunch of Securities you could buy for $90 on one day it costs you $170 the next day now that I call a bubble uh and it unlike the first case where Jean and I could argue uh forever as to whether those home prices were rational or irrational I'm pretty sure Jean doesn't think that it would be smart to pay $170 for $100 worth of cruise ship line and Mexican companies and all through this period there was no change in the value of their assets so

it's not like the market was anticipating some boom in the Caribbean this is just a mistake okay jene farmer this is an anomaly but it's it's also a bubble in your terminology both well it's a one day bubble no no it goes up and then it takes a year to come back to come back well it drops most of that 170 the next well I mean a few months later it's still anyway it's an anecdote not it's there's a difference between anecdotes and evidence right okay so uh as you know I have lots of these

anecdotes like uh my paper with Owen on palman 3om okay that one right um and so when I mean this this was an example where a a part of a company was worth more than the whole company that can happen but yeah but because the rest of the company can be unprofitable yeah but the rest of the company was actually the only profitable part in this case so I just say for more details on I just say for more details on that case people can read your book misbehaving correct but the point here is is

you think I don't deny I don't deny that there exist anecdotes where there are problems I don't deny that it's it's just for example for bubbles I want a systematic way of identifying them in my view it's a simple proposition you have to be able to predict that there is some ending to it and all the tests that people have done trying to do that they can't do it uh so statistically people have not come up with ways of identifying bubbles um I think that there's a lot of identification of bubbles uh based on 2020

hindsight and it's very easy to do in that uh that situation for example irrational exuberance which uh well that was shill it takes credit for that but if you actually date the time the market goes up more afterward and never goes back to that point so irrational exuberance never went away if that's what it was so so th this is where we are I mean and why do I bring up amusing anecdotes which I agree this this is uh um it's a speck um and when Owen Lamont and I presented our paper in the Finance

Workshop uh presenting another one of these anecdotes Jean and I got into a discussion of icebergs and Jean's point was that like this is the iceberg yeah that I can go find these cute little anecdotes this is the whole problem uh and you know okay so little stuff can go wrong and I my argument is and there's no way to prove which one of us is right is look these are the few cases where we can test whether the price and intrinsic value are the same and you know it shouldn't be that a small unprofitable

part of a company is worth more than the entire company where the the rest is profitable it shouldn't be true that shares of the Cuba fund are selling at a 70% premium now you know I go to these and say look if the market can't get this right wa there there are other examples of cases where you can't test it for example per Mutual markets are a good example you can test whether the yards are good predictors of eventual outcomes and they turn out to be very good well they're very very good although there's something

called the favorite long shot bias so uh if you go to the racetrack you shouldn't bet because they take 177% but uh if you do you want to bet on favorites because like a 100 to One Long Shot will win One race out of 400 so the the prices are correlated um and the deviations aren't enough to beat the 17 % spread but uh but there are some anomali just I just want to press you there for a second how do you define the bubbles well I would say bubbles are when prices exceed uh a

rational valuation of the Securities being traded right but what what's the test of that well the only tests that are clean are these anecdotes uh like closed in funds where we know the value of the assets and we know the price and we can see that they're different for something like the real estate market uh we only have a suspicion and we can't prove it although um I I have some ideas I'm working on with what one of our golf buddies to um figure out how to predict when a bubble's going to end okay sounds

like a good reason to play more golf but to go back to the iceberg for a second uh I mean if if financial markets are inefficient and you're saying there there's more that we haven't seen that's the point of the iceberg example where where are the biggest inefficiencies well again it depends on uh which definition we're using so where are you most likely to be able to beat the market smaller uh smaller firms um less developed countries but although even there the the advantage that active managers have um is relatively small okay but there are

there I would those to me seem like I that one you'd have to test whether that actually works we have tested that that one doesn't work but things that are more systematically tested Ed that our indications of some degree of markeet efficiency are for example the accountants have long established that the adjustment of announcements to earnings is very quick but not complete it takes a few more days before this's complete adjustment not enough to make any profits on but so what it's still a slower adjustment so that's an indication that the Market's not completely uh

efficient the whole process the whole momentum phenomenon gives me problems it could be risk but if it's risk it changes much too quickly for me to capture it in any asset pricing model so that one gives me the biggest problems of of all so the point is not that markets are efficient it's you know they're not that's just just the model the question is how inefficient are they I I tend to give more weight to systematic things like uh failure to adjust completely to uh earnings announcements uh or momentum uh than to anecdotes which uh

to me seem less they're fine but they're just little things popping up they're they're curiosity items rather than evidence but dick Taylor highlighted one one area where you do agree which is about the value premium that low price price stocks T do better that one's that one's unresolvable so but but you both have different explanations why that's the case so so can you give your explanation and tell me what my explanation is that uh value stocks are just riskier than grow stocks um initially the people who thought that wasn't true thought that there was an

Arbitrage opportunity in in value versus growth that if you went long value and short growth you'd get a portfolio that had a very low variance and a high return turned out that wasn't true if you went long one and short the other you got a a net that had a very high variance so it looked smelled and tastes like a risk factor but you can't really establish that unless you can tell me why this source of variance carries a different price per unit than other sources of variance because that's what you're into is as soon

as you deviate from the basic Capital asset pricing model you're really seeing different sources of variances carry different prices per unit of of variant and the fmer French three Factor model was kind of the first one to to put that into our operation and 20 years have passed and people have been trying for 20 years to identify whether that's due to uh some taste factor or something people are trying to hedge against and you know although I have a vested interest in saying good somebody's identified what it Hedges against I don't I don't really find

it convincing the arguments on on either side so I think that's just an open issue at this point that's why I said that that issue is basically unresolvable at least as far as the test so yeah I I pretty much agree with that um but you know Gan and Ken have gone now to a five Factor model uh where we're still working on that uh maybe four maybe okay but uh you could add momentum and go six yeah there you go um and you know in my view there was one rational model of stock prices

and that was the capital asset pricing model um and I think in a world of rational investors the capm would be true no that's false but that's what I think right there's no multi-period model that ever leads to the capam well in any case it's certainly not true um that's true and and we have these other factors uh like size and value and now profitability uh and investment uh now I've looked hard to find the way in which value stocks are risk rier than growth stocks and have been unable to find them I agree with

Jean that betting on that spread is a very risky activity any hedge fund that did that would have gone out of business in the late 1990s um but that that doesn't mean that the explanation for the abnormal returns is is due to risk no nobody can prove that so what is your explanation I think that uh value firms look scary and U they get a premium for that there's another story that has they don't have to look scary they just people don't like them uh and economists don't argue about taste so value stocks tend to

be low performing companies who have few investment opportunities and aren't very profitable and maybe people just don't like those so that that story to me has more appeal than a Miss pricing story because Miss pricing at least in the standard economic framework should eventually correct itself um and it shouldn't keep repeating whereas taste can go on forever well but so I would disagree with that um which part W so I don't think you can call it taste so I don't know I'm not saying I can call it that that appeals to me more suppose you

say you like $20 bills and you're willing to take 420s for 100 now that's taste now I'm going to make a lot of money that's an Arbitrage yeah well there no Arbitrage here so but the question is what if if the people who dislike value stocks and that's just taste and it's wrong that's not wrong remember now we're economists you're behavioralist you that's even worse so you you don't you don't comment on people's tastes I yeah I do when they say that they like 420 is better than 100 that's an arbitr that's different well um

suppose I tell you I like apples better than oranges uh then that's taste okay so that's stocks and gr stock we're both we're both I'm not arguing for it I'm just saying it's responsibility well but look we're we're both affiliated with Asset Management firms and both of our firms invest at least partially in small value stocks now uh we're hoping to earn high returns and do so well sometimes yeah sometimes not all the time or it wouldn't work if we're buying those stocks because people don't like them we're only going to make money if they

change their mind some or some people change their mind and so that's why the taste argument uh I don't think there's any I mean I think you're you're more behavioral than me now I'm an economist e economics is behavioral there's no doubt about it isn't that your point that all economics should be Behavior but there's a difference his point is why we have stopped just then no no no wait there is a difference because yours is irrational Behavior mine is just Behavior oh no I I hate the word rational oh good uh so I the

distinction I make is whether behavior is predictable for a rational model okay and I'm willing to include behavior that is not predicted by rational oh okay no I I I would agree with that and and look I think you know at the end of my book I call for what I call evidence-based economics and I think that's what Gan does and has always done the there's no way the point I was making about the five Factor model or the three Factor model is there's no way you can derive that from some axiomatic first principles no

you can't th this it falls in the context of Merton's model but you have not identified the relevant State variables that would give to it and I think it's actually more complicated than that that no one of these is really associated with a state variable they're all just linear combinations of multiple State variables combined in different ways right which makes the problem very difficult to but but the the way in which you and I are the same is we're both interested in understanding the world right and you know I have some puran interests in things

like the Cuba fund uh uh but um at you know at the main level I think we we would both like to know whether what caused housing prices to go up so fast and then back down and uh and then back up again and then back well yeah certainly part of way up not not exactly in the same places but um and if if those prices were wrong in some sense then it would be good to know absolutely and so if total agreement on that if I were the chair of the FED um or in

charge of fetty and Franny if I saw places like Vegas and Scottdale were in the 1990s I would be raising lending requirements you could borrow at well there were the you know liar loans uh but you you could borrow at very low uh interest rates and very low down payments into what looks like uh pretty pricey market so uh so you're thinking policy makers should use bubbles as a as a way to step in and interv but very gently it's not that I think that um policy makers know what's going to happen but uh if

they see what looks disturbing they can lean against the wind a little bit and that's as far as I would go something we certainly both agree about is no not on that one no no I'm going to say something I think we both agree about is that we stock markets good or bad are the best thing we got going so nobody's devised a way of allocating resources that's better than markets in general than markets in general we're in total agreement about that we're in disagreement about whether policy makers ever get it right though uh well

it sounds like you're asking the policy makers to step into the market you said was was so almost surely will do more harm than good yeah well I I'll uh the the argument that whether the policy makers ever get it right I think that's a little strange unbalanced whether they cause more harm than no uh but you know if they listen to us then no no then that surely cause more harm than Jee farmer you said earlier you were behavioral uh Economist has your thinking been shaped by the sort of Behavioral Science that Vick thalor

has pioneered uh no but was the three Factor model a response to some of the work that absolutely not it was a response of the data basically it's what we call an empirical asset pricing model it really it has a it has this vague connection to Merton's model but it was really empirically inspired those were factors that were screaming at us from the data basically and we put them in there and got a lot of credit for it but it really was kind of an obvious thing to do um and it's had a 20-year run

so we can't complain about that that's as long as the run that the cap capm had so can't can't really complain um but to your mind has behavioral uh science what impact has it had on economics oh I think they're sh every economics department is into it to some extent or another right yeah um it's still kind of what I'd call curiosity items rather than in other words 20 years ago I made this criticism of Behavioral finance that it was really just a branch of efficient markets because all they were were complaining about us so

I was like probably the most important behavioral Finance person because without me there was no behavioral Finance you're the reference point yeah I'm the got to criticize you know but uh I and I I still think that's that's true there is no behavioral asset pricing model that can be tested front to back well I think there's no asset pricing model who that's a really nealis point of view though well but you you have said Dior theoretical one you've said that you referred to the I mean the efficient market hypothesis Remains the kind of the standard

to to which your work is directed yeah that's true of all right all economic models so you know uh expected utility theory is the right way to make decisions under uncertainty people don't um you know in in in my managerial decision-making class I I give them rules at the end of class and one is ignore Sun costs assume everyone else doesn't and that's kind of my philosophy of life I I believe the rational model and I think a lot of people screw it up and that we can build richer models and models with a better

predictive power if we include the way people actually behave as opposed to the way fictional creatures that are the so-called eons econs that are uh you know as smart as Kevin Murphy and uh and have no self-control problems um but I don't know anybody like that one of the criticisms that's made sometimes of you is that what you're pointing out essentially are anomalies like the Cuba uh fund and there is no overarching Theory which other people can then um try to reject so do you need a theory will there be a theory no there will

well there won't be a new overarching Theory we've got one it just happens to be wrong like all theories but yeah so and it's so it's not going to be like the comper revolution where having the Earth in the middle was clearly wrong and having the sun in the middle was right um it's not going to be like that it's going to be more like engineering so physics in its pure form with lots of assumptions doesn't build build good Bridges you need engineering and uh that's that's what the behavioral approach to economics is and I

don't think it's really all that different than what Gan does is this really an academic debate you said at the beginning that you essentially agree about investing strategy what regular investors should do so is this a debate that really affects the the typical investor retail investor Jan farmer um I don't think I mean I think when kman was asked after he got the Nobel priz how should investors behave he basically said they should buy Index Fund that's that seems to be the bottle but that he they come from it from a different perspective because since

they think everybody's irrational the only way to make them rational is to have them tell them what to do that that's possibly uh rational whereas I think the rational thing to do because prices reflect available information pretty much is to is to is to be a passive uh investor um but my my complaint about lots of the stuff that falls under behavioral finance and this is not a complaint of him I always say that he is very um he knows the the the uh the psychology aspect of stuff and he's already always oriented uh towards

that but he has lots of there are lots of acolytes of Behavioral Finance who are pure data dredges all they're doing is out there looking for anomalies they have no connection to anything in in Psychology if you look at uh a behavioral Finance NV thing it will be populated with those kinds of of papers that are pure data dredging looking for uh anomalies and that's I think I don't know I think I would I'd cut them off the program if I well you know I I'll agree to do that if we can cut the theory

dredging but what about this this uh this idea that this may be an academic debate doesn't really affect individual investors or indeed as you said earlier both of you involved with with money management firms that seem profitable so someone would say well you're coming at it from completely different perspectives we're able to make money basically with the same strategy as you pointed out what is the big disagreement here well the strategies aren't exactly the same and uh and David Booth is a better marketer than anybody at Fuller and Taylor but um no I I think

I if there's a non academic point about this it's whether things like let's say the rise of technology stocks in Jean's honor I won't refer to it as a bubble uh whether that was a misallocation of resources and on hindsight it was onight it was and if the rise and fall of technology stocks was a bubble um really internet stocks you mean not yeah essentially internet stocks uh although even companies like Cisco were uh but technology stocks are still good fraction of the market so so I mean if prices can be off you know Fisher

black said he defined an efficient market as prices within a factor of two fish is said lots of crazy things though so so uh that's my definition of market efficiency and you know I have a Chicago Nobel Prize winner I'm resting on and I if that's right then um you know we had during those days a lot of our MBA students were quitting after their first year to go out and make their billion and most of them didn't so um and you know the same is true for the housing market and uh so I think

these are important questions that are are not just academic disputes and uh they'll be very important in trying to understand the way the global economy works now I'm not saying we can recognize them when they're happening although I'm working on that um but I do think that we can have a pretty good hunch and uh that solving that uh a bubble detection committee uh would be highly useful uh if it were reliable and we're not there yet and just saying it's impossible I think I'm not saying it's impossible just saying so if if then we

can agree on it's hard to tell except for my cute anecdotes like Cuba in in general it would be very useful to know to what extent all economic outcomes are due to uh rational or irrational interplace um we don't really know that I no this is this is a point that's more that's not that's not just the that would that would improve everybody's uh lives understanding is more understanding is better than less understanding okay well on that note uh our time is up this has been a fascinating discussion and maybe we can do it again

when you have come out with your bubble uh research look for look forward to that uh my thanks to our panel Eugene farmer and Richard theor for more research analysis and commentary visit us online at review. chicago.edu and join us again next time for another big question [Music] goodbye