Raghavendra Rau On Insights From Behavioral Finance & Whether It’s Possible To Beat The Market

What Is Behavioral Finance?

Aiden Singh: Behavioral finance is a fairly new research paradigm that really began to gain traction following the work of Daniel Kahneman & Amos Tversky. And it’s one of the primary areas you conduct research in at Cambridge Business School. 

Can you share with our readers a little about what behavioural finance is, how it differs from more conventional finance research, and why it’s of interest to you personally?

Raghavendra Rau: Well, my research covers a bunch of diverse areas, all the way from investment banks to hedge funds to mutual funds. 

But the key concept that connects everything I do together is the fact that they all deal with human beings.

Basically we have two types of research in finance. 

One type of research is this area of asset pricing, option pricing, and so on, where you have a bunch of numbers which are predicting another bunch of numbers. 

I'm not crazy about that kind of research simply because I like thinking about what people are doing. It's more interesting to me to observe that the CEO does this or the analyst does that. That's more intuitive.

And that, I think, is one of the concepts which distinguishes classical finance from behavioral finance. 

Classical finance actually has a set of only six ideas. And the problem with those ideas is they all have to fit together. Otherwise, if one of the ideas doesn't work, none of the ideas work.

And at the core, the simplest idea is that ‘there is no free lunch’ - no one gives you something for nothing. And if there's a free lunch available, which essentially means I can get money for nothing, then everybody would try to take advantage of it and the opportunity would disappear right away. That's a core concept in classical finance. 

Now, classical finance people do not say that human beings don't have biases; they say people do have biases. So I have biases; you have biases. 

What classical finance people say is that, in the aggregate, those biases cancel out simply because if you consistently make mistakes - for example, paying too high a price because Elon Musk tweeted about a stock - somebody else would be happy to take your money by selling you that stock at a very high price. So they make money and you lose money, and you're driven out of the market.

The only people who survive in the market are the rational investors who are driven by mathematical concepts like net present value, portfolio theory, and so on. 

So the classical finance people say we don't really need to worry about individual human biases because they don't matter in the markets. They just cancel out.

Behavioral finance takes the perspective that, because these biases are correlated, we may all behave in the same way at the same time. And the effect on market prices is large enough that the smart investors cannot easily take a stance against it. So high prices can stay high. 

There's a popular saying in finance: the market can stay crazy longer than you can stay solvent. Yeah, you might possibly be super rational, trading against the market, but if the market continues to be crazy, you run out of money eventually.

So what that means is individual human behavior does have an impact on prices - and that's the whole essence of behavioral finance. Behavioral finance basically says individual biases are correlated, and they do impact pricing.

Aiden Singh: The implications of this are pretty dramatic: these are two very different approaches to thinking about the market that lead to two very different conclusions about how markets behave.

If you make the assumption that there's no free lunches and in the aggregate biases cancel out, as traditional finance does, then you arrive at the idea of market efficiency.

And if you take the behavioral finance perspective which says that market participants’ biases line up with each other's, you come to a perspective that something like speculative bubbles are possible and maybe markets aren't always so efficient.

So these are two very different perspectives on finance which lead to two very different perspectives on how markets work in the aggregate.

Raghavendra Rau: Absolutely. And a key part of both of them is that neither means that market is predictable.

So in classical finance theory, because the market is so efficient at pricing in all the relevant information, nobody can predict where the market is going to go simply because any new information will be incorporated into the prices. 

And in behavioral finance too, the market is totally unpredictable because you don't know how long the biases are going to last. 

So at the end of the day, both approaches say you can't make excess money in the stock market. But they come to that conclusion from very different perspectives.

- - - - - - -

Methodology

Aiden Singh: A lot of finance research is looking at numbers and data, running regressions, building models, that sort of thing.

But in your research, you like to look at individual behavior. When you're doing the type of research that you're doing, are you looking at individual stories of individual people? Are you trying to build narratives that explain certain things you're seeing? How does the methodology differ from that of traditional finance research?

Raghavendra Rau: Many of my research papers are actually built around newspaper articles.

So I read a newspaper article in the morning, and I say, ‘That's a weird behavior, what explains that behavior?’. Then I go out and collect data on that.

Aiden Singh: So your approach is data driven, just as classical finance research is. It's just a different approach to handling data.

Raghavendra Rau: Right. And the methodology I use is pretty standard finance methodology. But the way I think about things is, can I couch this in more psychological terms? That’s what I'm trying to focus on.

- - - - - - -

Market Under-reaction Vs. Market Overreaction

Aiden Singh: Let’s discuss some of the conclusions behavioral finance has drawn about market behavior. 

The behavioral finance literature posits two contradictory models of irrational investor behavior: one in which investors under-react to new information about a stock and one in which investors overreact to new information about a stock. 

You’ve conducted research in this area. Can you explain the logic behind the two models to our readers and which model your research suggests is more accurate? 

Raghavendra Rau: So the under-reaction model is as follows: market participants have their preconceived ideas about a stock. And so, when new information about that stock arrives, they will wait for more confirmatory information to arrive before actually making a move. And, as a result, they underreact to the new information: they don't buy the shares right away because they want to wait for this new information to be confirmed.

The overreaction model says something different. Suppose you’ve got a company that has been doing super well in the past. And suppose new information about this company comes out. This model suggests that investors will react very quickly because they will expect the company to continue doing well. So investors will quickly buy shares and there will be an overreaction.

So the difference between the two models is in its assumption of whether investors wait for confirmation when new information comes out or whether they extrapolate from the past.

And the overreaction model suggests investors will extrapolate too far based on past behavior. While the underreaction model says investors will wait for additional confirmation before they react. 

The problem is that both of them anticipate exactly the same outcome: prices will overshoot and then come down later on. 

But which causes what? 

So in one of my papers, I tried to take a look at this. What I basically say is: if you're under-reacting or if you're overreacting, what makes people think that the reaction is complete by the time the next piece of information arrives? 

It could be that a new piece of information arrived, but you're still under-reacting to the previous piece of information.

So does this new information confirm your previous view or does it contradict what you previously thought? If it contradicts what you previously thought, you're kind of saying, ‘Oh, I was right in waiting because this new piece of information is contradicting me’. 

But if the new piece of information comes out and confirms what you previously thought, you feel that you were right in waiting because now you have better information and you’re more confident in what you’re going to do. 

So in this particular case, what we're saying is that if you take additional pieces of information either in the same direction or in opposite directions, it's all under-reaction. 

And that creates a phenomenon called momentum. That is, stocks that have done well in the past continue to do well in the future. And stocks that have done badly in the past tend to continue doing badly in the future. 

Aiden Singh: And the observation that stock prices can be influenced by momentum is not something that would be predicted by classical finance. 

Raghavendra Rau: Right. Absolutely. According to classical finance, you should not see this kind of behavior at all on an aggregate basis. But momentum has been shown to be everywhere in financial markets: stocks, bonds, foreign exchange.

You always see this under-reaction pattern. 

- - - - - - -

Trading On Momentum?

Aiden Singh: You mentioned before that both the traditional methodological approach and the behavioral approach arrive at differing perspectives on how markets work, but neither concludes that you can exploit market inefficiencies. 

But here, you're saying that we consistently see momentum play a role in shaping market prices. Why is that not exploitable?

Raghavendra Rau: Well, here's the question: when does the momentum run out?

Let me ask you a question. For example, would you buy Bitcoin right now? 

Aiden Singh: No. I have never bought it and wouldn't touch it. No. 

Raghavendra Rau: It's currently priced at $100,000. 

So some people would say, ‘Let me buy it right now because it's got further to go - it’s got ‘momentum’.

And other people will say, ‘$100,000 for something which has no basis in reality -  why would I buy that?’ 

So it’s difficult to exploit momentum because you don't know when it's going to turn. That's the key problem here. 

Aiden Singh: So it's an issue of the turning point. It's not enough to just know that momentum can drive behavior. You would also need to know when that momentum is going to run out.

Raghavendra Rau: Absolutely. And we don't have a way to know. We have no way of predicting that.

- - - - - - -

The Effects of Corporate Name Changes

Aiden Singh: You and your colleague, Professor Mike Cooper, are among the world’s leading scholars on the effects of corporate name changes. And here too you’ve found an interesting result which traditional finance would not anticipate.

In particular, you’ve looked at the effect of corporate name changes on firm valuation during and after the dot-com bubble. What did you find?

Raghavendra Rau: The literature until then pretty much said that if companies change their names, it doesn't really affect the fundamental cash flow of the business, so the stock price should not change.

But one evening my wife and I were hosting a party at our house. And Mike and I were discussing, as finance people do, the day's Wall Street Journal.

Aiden Singh: Sounds like my kind of party.

Both: Laugh.

Raghavendra Rau: And the Wall Street Journal had this article which was basically looking at companies that had changed their names from ‘something’ to ‘something.com’. And the companies that had done this had a huge pop in their stock price. 

And, Mike and I said, ‘Wait, how is that possible? The cash flows haven't changed. All these guys have done is change their name. How has the stock price gone up?’ 

So we fired up the computer, looked into it, and within 15 minutes, we found 30 companies which had changed their name from ‘something’ to ‘something.com’.

And about that time, my wife got mad because we were abandoning the party. So she made us go back to the party. But luckily, one of our PhD students was there. So I put him in front of the computer and he became a co-author of the paper that came out of this.

We ultimately found about 200 companies that had engaged in this kind of weird name change thing. 

So the thinking of those companies was: well the internet is amazing and we have to be involved with it. So they tried to find anything which might get them more closely associated with the internet. 

And some of these companies had absolutely nothing to do with the internet. 

We also found that after the dotcom bubble was over, the same companies that added dotcom to their names got rid of the dotcom. And then they had another boost to their stock price. 

We saw the same effect in mutual funds. So when growth is what’s in, we see mutual funds changing their name to have ‘growth’ in their names. When value was what’s in, they changed their names to have ‘value’ in their name. But the underlying holdings stayed the same. So, the actual underlying mutual fund is the same, but you see a pop in their stock price when the name change occurs. 

And we've seen examples recently like Long Island Iced Tea, which changed its name to Long Island Blockchain in 2017.

Kodak tried to introduce a cryptocurrency. Now it's coming into AI. So these are the hot buzzwords. And every time a buzzword trends, you will see companies pretending to be part of that buzz.

Aiden Singh: As you mentioned, you’ve also looked into name changes by mutual funds. How do these name changes affect flows in and out of those funds? 

Raghavendra Rau: We found that flows into mutual funds increased after name changes to include trendy buzzwords, even though the funds didn’t change their underlying holdings.

- - - - - - -

Exploiting These Findings To Beat The Market?

Aiden Singh: Let me propose a narrative to you and you tell me what you think is wrong with it, if anything. We've talked about momentum affecting price behavior. We've talked about these crazy things, like name changes, affecting the way the market values a stock.

Couldn't a savvy investor use valuation techniques, try to assess the actual underlying value of the company, and then say, ‘Hey, momentum or this weird market reaction to a name change has mispriced the stock either too high or too low. And we'll trade accordingly - we’ll exploit that difference between what valuation techniques tell me that stock is actually worth and what it's priced at today.’ Why is that not possible? 

Raghavendra Rau: Well, the problem here is you don't really know what the benchmark should be. So if you look at the portfolio theory, it tells you the capital asset pricing model is the way to go. The problem with the capital asset pricing model is, empirically, the data doesn't fit the capital asset pricing model.

So Fama & French proposed what they call the 3-factor model. So the three factors are 1) market risk, 2) small minus big (firm size), and 3) book-to-market value.

But there's no theory here. You're just pulling out a bunch of factors and trying to make it fit the data. I think they're going up to 6 factors now.

Why those 6 factors? Well, because it fits the data.

And then there's a recent paper also by Pat Akey and Adriana Robertson which actually shows that the fundamental factors themselves have been changing over time. 

Basically, that means you don't know where the benchmark is. You don't know what the risk is. And at the end of the day, how do you know that you're actually beating the market? 

This is called the joint-hypothesis problem. It's a giant problem. 

- - - - - - -

Where Does A Cambridge University Behavioral Finance Professor Invest His Money?

Raghavendra Rau: That's why I don't mess with any of this. 

The easiest way, even for a finance professor, is to invest in low-fee index funds. 

There's no stress. And the market has done enormously well in the last 10 years. So from 2011 till today, I would probably say I've tripled my money with no effort at all. 

Aiden Singh: If you're a professor at Cambridge Business School putting your own money in index funds, I assume that's what you would recommend to a friend or a family member?

Raghavendra Rau: Yes. Absolutely.

Aiden Singh: So I take it you're not a believer in the strong-form efficient market theory at the very least?

Raghavendra Rau: I'm probably not. Even in the semi-strong form or weak form. 

But, here's the problem: I don't believe that we can explain all these things using rational market theory, but they're still not predictable. You still can't make money. 

So with my own investments, I have everything in low fee index funds.

- - - - - - -

What Do We Make Of Warren Buffett Then?

Aiden Singh: So Warren Buffett is an exception? Is it luck? 

There is the idea that if you have monkeys throwing darts at a dartboard, most monkeys are going to miss. But if you have enough monkeys doing it, one of them is going to hit several times in a row just by chance.

Is that how we should see Warren Buffett’s track record in the stock market? There are so many investors playing the markets, one of them was bound to have several hits in a row just by sheer chance?

Raghavendra Rau: The key part is that most of Warren Buffett's outperformance happened in the first 15 years. After that, his performance has been pretty much equivalent to the market - hasn't really done better or worse. 

So the outperformance is only for 15 years. And 15 years is a relatively short time. 

If you take 1,000,000,000 people and have them toss coins, statistically, it’s entirely possible that one of them will toss 20 heads in a row.

Aiden Singh: So when we see prominent finance people who have had success in the market over 10 years, 15 years, we should be skeptical of whether that can continue indefinitely? 

Raghavendra Rau: If you take, for example, all the major hedge fund managers like Steve Cohen or Bill Ackman, all these guys are making lots of money every year - literally billions of dollars year after year. 

The key part is most of that is driven by the fact that they already have a billion dollars. The first billion is always the hardest. And if you have a billion dollars of your own money and you're investing it even in low risk treasury bills, you're earning $50,000,000 a year right there. So if you have billions to invest, it's easy to continue to earn a billion dollars.

So the trick is that getting to the first billion is difficult. And once you're there, you're reinvesting. 

- - - - - - -

Majority Shareholders & ‘Tunnelling’

Aiden Singh: Let’s turn our attention now to your work on transactions between publicly-listed firms and their controlling shareholders, and so-called ‘tunnelling’. 

Can you define for our audience, what is ‘tunnelling’?

Raghavendra Rau: Let's say you control 51% of the company. So you can do whatever you want with the company without the other shareholders being able to say anything about it. 

Let’s say this company is manufacturing furniture. So one thing you can do, for example,  is set up your own company to supply wood to this company. And, of course, you price your wood really high. 

The furniture company would be able to do better by buying wood on the open market. But because you control this company, you can ‘tunnel’ out money from the remaining shareholders of the company.

That is essentially tunneling. It's somebody who has the majority ownership of the shares of company A and another company B which perhaps is a supplier to company A at a rate which is above market rates. 

Or company B could be a customer of company A, and the controlling shareholder of company A tunnels out money by having company A sell goods to company B at a really low price.

We looked at data on this from Hong Kong and China. 

But the key part is, at the end of the day, if you do this carefully enough it's very difficult to figure out. Was this really tunneling? Or was it just that this guy was giving company A some additional services that it would not have received elsewhere? Maybe the controlling shareholder of company A says that the wood company B is supplying it at above market rates is specially treated. Or the quality of the wood is different. Or it's coming from a sustainable source.

Aiden Singh: Using data from China, you’ve carried out research into what types of firms are most likely to be subject to ‘tunnelling’. What did that research find?

Raghavendra Rau: So we looked at the direct effect on the share price, and we found it's, obviously, negatively related. 

We also looked at how the company holds its assets and assessed how that affects the tunnelling related negative effect on share price.

Let's say there are two companies, both controlled by a majority shareholder. Company 1 has all its current assets in the form of cash. And company 2 has all its current assets in the form of hard to sell inventory. In this case, Company 1 can be tunneled very easily but company 2 cannot. So if I'm a minority shareholder, I'll pay a higher price for shares in Company 2 than those in Company 1 because the shareholders can see that the money in Company 1 is more easily able to be stolen.

So basically, investors anticipate that if the money is easy to steal, it will be stolen. And so they will pay a lower price for the shares. 

Aiden Singh: So your research finds that market participants are aware of the potential for tunneling?

Raghavendra Rau: Yes, and that stock prices reflect that.

- - - - - - -

Sell-Side Analyst Coverage

Aiden Singh: So we’ve discussed your research on investor behavior. Another key player in the world of publicly-traded firms and stock markets is sell-side analysts.

What does your research indicate about whether sell-side analyst coverage adds value to a publicly-traded firm? And what effect does a loss of analyst coverage have?

Raghavendra Rau: So there are two streams of thought about what an analyst does. 

Approach one number basically says the analyst carefully analyzes the company and gives you additional information which you previously didn't have, so now you can value the company better. And this additional coverage and information means investors will be willing to pay a higher price per share than if there was no coverage. So in this line of thought, the sell-side analysts are creating value.

The other stream of thought is that they are not actually creating value. What's happening is that they're bringing the stock they’re covering to your attention. In other words, you pay more attention to the stock because the analyst is talking about it whereas otherwise it would be out of sight, out of mind. 

Previous research papers had looked at what happens when you lose analyst coverage; so what happens if a stock goes down  from 5 analysts covering it to 4 analysts covering it.

What we looked at in our paper was what happens if you were previously covered by an analyst and then nobody's talking about you. And we found that the company finds it way more difficult to raise money. The cost of capital goes up dramatically - both the cost of equity and the cost of debt. And we found that the company is significantly more likely to go bankrupt. So, essentially, being in the forefront of attention matters.

And so a lot of companies actually pay for independent coverage, which, of course, is always positive coverage. But the key question is, what is that independent coverage doing? Because everybody knows it's positive: if I'm paying you to write a report on me, you're not going to write a report which makes me look bad. So what good is that report doing? 

The key thing here is that the report is bringing you to the attention of people. That's the value of that report.

Aiden Singh: You’ve also looked into whether sell-side analysts change their behavior when moving to a new investment bank. What did you find?

Raghavendra Rau: The key part here was looking at what makes an analyst want to cover a particular firm. 

What we saw was when an analyst moves from one bank to another, the new bank has different clients from the old bank. So if you’re working with the old bank, you have some incentives by the company to basically say, ‘Hey cover these guys better; cover these guys more favorably because they are our clients’.

What we looked at was, you move from bank one to bank two, do you cover a different set of stocks? The answer is yes.

You cover the stocks who are clients of your new company. And analysts suddenly become much more favorable towards the new clients and much less favorable towards the guys they’re leaving behind.

So moving from one firm to another does change analyst behavior.

Aiden Singh: And does bringing in a new sell-side analyst affect the flow of deals to the hiring investment bank?

Raghavendra Rau: Absolutely, that's why they do it. 

No company will hire a bank if it thinks the bank’s analysts are biased against it. What's the point? This bank’s analyst is trash talking me all the time, so I refuse to hire this bank. Or they say, I'll hire you, but you need to fire that guy first. 

Aiden Singh: So your research lends itself to viewing sell-side analysts, not as some objective arbiter of whether this is a good company or a bad company, but as a sort of vehicle for bringing in deals to a bank. 

- - - - - - -

Everything Is Aligned Against You Beating The Market?

Aiden Singh: So according to your behavioral approach, if you're an investor trying to beat the market everything is aligned against you.

There's momentum in price behavior, but it’s not exploitable because you can’t predict when the momentum will reverse course.

Markets react to corporate name changes in ridiculous ways. 

And then the sell-side analysts, who are supposed to be objective interpreters of what's going on with a particular company, are actually not so objective, it turns out.

In this view, everything is aligned against you trying to beat the market.

Raghavendra Rau: Exactly.

- - - - - - -

Markets Today

Aiden Singh: When you look at markets today, do you see anything that is not explainable by traditional finance theory, but is of interest to you as a behavioral finance researcher? 

Are there areas in the market today where you see mispricings or momentum playing a role? Are there areas that you are looking at and thinking, ‘this could be the topic of a research paper one day’?

Raghavendra Rau: AI name changes is an obvious example. Any company which has anything to do with AI today will have the word AI right there in the name. 

Or even if it doesn't have anything to do with AI, some companies will use the buzzword AI. I recently saw an example where somebody was talking about a new set of hearing aids using a phrase like ‘edge AI’ or something like that. I was like, is there AI in your hearing aid? I mean, it's a hearing aid! 

But that's an example of using the phrase AI in marketing right now.

So that's one example of something in the markets which is directly related to my line of research.

For our last paper on mutual funds, we saw a bunch of companies changing their names to have ‘China’ in them. Then blockchain was really hot. Now AI is really hot. So every couple of years, there's some new fad coming along.

- - - - - - -

Further Reading

Investor Under Or Overreaction

“Investor Reaction to Corporate Event Announcements: Under-reaction or Overreaction?”, (with P. Kadiyala), Journal of Business 77 (2), 357-386, 2004. 

Corporate & Mutual Fund Name Changes

“A Rose.com by Any Other Name”, (with M. Cooper and O. Dimitrov), Journal of Finance 56 (6), 2371-2388, 2001 (Reprinted in The Psychology of World Equity Markets, Werner De Bondt, ed., Edward Elgar Publishing, Cheltenham, UK, Volume II, 500-517)

“Changing names with style: Mutual fund name changes and their effects on fund flows”, (with M. Cooper and H. Gulen), Journal of Finance 60 (6), 2825-2858, 2005.

Majority Shareholders & Tunnelling

“Buy high, sell low: How listed firms price asset transfers in related party transactions”, (with Y. Cheung, Y. Qi, and A. Stouraitis), Journal of Banking and Finance 33 (5), 914-924, 2009.

“Tunneling and propping up: An analysis of related party transactions by Chinese listed companies”, (with Y. Cheung, L. Jing, T. Lu, and A. Stouraitis), Pacific-Basin Finance Journal 17 (3), 372-393, 2009. 

“Tunneling, propping, and expropriation: Evidence from connected party transactions in Hong Kong”, (with Y. Cheung and A. Stouraitis), Journal of Financial Economics 82 (2), 343-386, 2006. 

“Does the market understand the ex ante risk of expropriation by controlling shareholders?” (with Yan-Leung Cheung, Aris Stouraitis, and Weiqiang Tan), Journal of Corporate Finance 68, 101946, 2021. 

Sell-Side Analyst Coverage

“Is there life after the complete loss of analyst coverage?”, (with A. Khorana and S. Mola), The Accounting Review 88 (2), 667-705, 2013. 

“The impact of all-star analyst job changes on their coverage choices and investment banking deal flow”, (with J. Clarke, A. Khorana, and A. Patel), Journal of Financial Economics 84 (3), 713-737, 2007. 

——————