Finance 527: Lecture 37, Psychology of Investing V4

Transcription

1 Finance 527: Lecture 37, Psychology of Investing V4 [John Nofsinger]: Welcome to the fourth video for the psychology of investing, and we are going to talk about the representativeness bias and the familiarity bias. This is a bias of stereotypes is what we see here. And stereotypes I know is a word that culturally has a lot of negative connotations. But from a psychology point of view, it s taking characteristics we know and putting them onto things that we don t know so well just because they seem similar. And that s the representative bias. Sure saves us a lot of time. But just like in the society term of stereotypes, they often lead us in the wrong direction. Psychology of Investing V4 John Nofsinger Chapter 8 Representativeness and Familiarity [Image of Stereotypes-A Real Time Saver] [John Nofsinger]: So these representativeness bias and familiarity are what I call mental shortcuts. Psychologists call these heuristic simplification that s so why I call them mental shortcuts. And essentially the idea is that the brain uses mental shortcuts to reduce the complexity of information that we have to make a decision, right. At any given point in time, you have thousands of colors and things going on in your eyes, you hear lots of noises, smell things. You have lots of information coming into your brain from our senses. And we need to cut through most of it, use what s important, and then make our decision. And so these shortcuts allow the brain to do that. So we need these shortcuts to function. However, they sometimes especially in dealing with decisions with complexity with uncertainty and with risk, they can lead us into the wrong decisions. Mental Shortcuts The brain uses mental shortcuts to reduce the complexity of analyzing information These shortcuts allow the brain to generate an estimate of the answer before fully digesting all the available information Examples are representativeness and familiarity [John Nofsinger]: So let s focus on the representativeness bias. Again, it s one of kind of like stereotypes. So here s an example of a car. I don t know much about car engines. And I look at this car and it s clean. Right it smells nice the outside of the car. So I know those characteristics 1

2 are good for the inside of the car. And so I project them and I assume that the engine must be a good engine as well. I don t know that but that s what we have a tendency to do. Every used car dealer in the world knows this. Right don t fix the engine. Just clean it up, make it smell nice, and you have a better chance of selling it. So that s one kind of representative bias. Another one has to do more with extrapolation kind of thing. I do not know where the stock markets gonna go tomorrow or the next week or the next month. But I do know what happened yesterday, last week, or the last month. And I take what happened then and I extrapolate into the future so we end up chasing whatever the hot mutual funds last month. That s where all the money goes this month. Wherever the stock gainers were that s what we want. We jump into the market after its already been going up as we ll see again later. Representativeness Bias The brain makes the assumption that things that share similar qualities are quite alike. A clean care must be a well running car First example-past returns represent future returns Investors chase the hot mutual funds, the high stock price gainers, and a rising stock market [John Nofsinger]: Roulette anyone? Roulettes a game where the numbers spin in one direction and the ball spins in another direction and eventually the ball will stop on a number. And you can bet on a number or you can bet whether it s gonna be red or black. There s also a green one in there. So there s the same number of red as there is black. And so you can bet and try to win Roulette Anyone? Roulette is a casino game where a wheel and a ball are spun in opposite directions and eventually, the ball falls into a space. Bets can be placed on whether the ball will fall into a red, black, or green space. There are 18 red spaces, 18 black, and 1 green. [John Nofsinger]: If you were in Vegas and walked up to a Roulette area and you saw a couple of Roulette tables. And you watched and one table, these were the numbers that it had stopped on for the last five spins. And another table, these were the numbers that were recently had. So you have a question. Which table would you feel more comfortable placing a bet on black? K which one-you re gonna put a bet on black-which table do you pick? Well most people will end up picking this table. 2

3 Question: You re in Vegas and walk up to roulette tables with the 5 previous spins posted as: Which table would you feel more comfortable placing a bet on black? [John Nofsinger]: The representativeness heuristic that we re talking about part of that includes a belief that even in small samples, they should be representative of the total population. And we know that in general half should be black and about half should be red. And since you get a lot of reds, then a lot of people think with all these reds in a row that a black is due. Right it somehow has a greater probability of occurring next time because somehow we know there has to be more blacks now because there s been a lot of reds. Well that s not true. In any small sample, the chance of getting a black is exactly the same as it was before and is exactly the same as getting a red. Right it s Well there s a green one out there so we ll say it s So that s the Gambler s Fallacy where we think that the large sample population characteristics have to apply to small samples. Representativeness Heuristic Representativeness Heuristic Belief that even small samples should be representative of the population People also think that the recent past is indicative of the future The Gambler s Fallacy says: 5 reds in a row in a fair game black is due! [John Nofsinger]: This is a chart where we see the net new cash flow into different kinds of mutual funds-actual stock funds, equity funds. And what these bars represent is the tens of billions of dollars that have net that go into a stock fund. The red line up here is the S&P 500 and it s measured over here on the right hand side. And we know from the dates here that this represents the top of the market, which is kind of known as the tech bubble era. And what are we doing at the top of the market? We are buying stocks like crazy. We ve seen the stock market go up for much of the 90s especially the last half of the 90s. And so we ve seen that, we extrapolate 3

4 that it s gonna continue to go up so we buy like crazy. Well what happens then? It goes down right. And so what happens when we get around the bottom? We sell. Well buy high, sell low, is that how we make money? Of course not. But again we ve extrapolated this long negative trend in the market and so we decided that it s gonna continue to do that and so we sell. So that s a form of the representativeness bias or extrapolation bias some people call it. Now notice that the rest of the time period in here, we kind of learn some lessons a little bit. We didn t buy too strongly right before the financial crisis so that s what this period is known as-the financial crisis. We are selling like crazy at the bottom though of the financial crisis. So that s not good. And this tells me here that we re just full of fear. K the market ends up going back up and I don t have the latest numbers on here because the monthly mutual fund data always comes out later. So I haven t quite got that updated. But the market did end up reaching you know a new high. But we were afraid because of all these other times we bought at the top. Extrapolating: Buy High, Sell Low? [Graph of time vs. Net New Cash Flow into Equity Mutual Funds and S&P 500 Index Value] [John Nofsinger]: Another way to test extrapolation is to-is being done by the American Association of Individual Investors. This is a group that is an association for us individuals. And you get a nice magazine every month and you get other literature. If you live in a big city, there might be a local chapter where you can meet once or twice a month and have little presentations and things like that. I ve actually presented at the Denver one a couple of times. But this organization will survey its members. And it will ask you: hey are you a bull or a bear on the market? You know what s your asset allocation like? What are you planning to do? And so they investigate that and they do say that people-there s sentiment about the market. If they think it s a bull market, they do buy stocks. But if they think it s a bear market, they sell. So their sentiment does lead to action. However when you look at that, action do they seem to be able to time the market well? That is are they getting in and then a nice bull market? And then when they get out, do they follow that up with a bear market? The answer is no. Their market timing is neutral at best maybe poor. But if you investigate their sentiment, are you a bull or a bear? And you then look backwards in time, you see that oh, they are bullish after the market has already had a great return. And their bearish after the market has already gone down. So another example of extrapolation. Extrapolation Weekly and monthly survey of its members Bull/bear, asset allocation 4

5 Asset allocation changes over time Outcome: Sentiment about the stock market does lead to asset allocation changes No abnormal returns (poor market timing) Sentiment is backward looking in trends [John Nofsinger]: The other representativeness form of the bias is investors-and this is more like the car example-investors confuse a good company with a good investment. K whatever a good company means to you-maybe strong financials or growth, maybe you like the products that the company produces, maybe you d like to work for it, maybe it s a green company or has a lot of sustainability. There s all kinds of definitions you might come up with why that company is good. But that doesn t mean it s a good investment. Those are two very different things. In fact, a lot of times, good companies make poor investments. Apple computer or Google are often considered really good companies. But there are plenty of times when they have some pretty bad returns. Why? Because they re already overpriced. So it s a good investment is one that goes up in price-makes you a lot of money. They re not necessarily always the same thing as a good company. Second Example (Representativeness bias) Investors confuse a good company with a good investment Good companies have strong earnings, growth, good products, good to work for, etc. Good investments increase in price more than alternatives. But good companies are often overpriced and make poor investments! [John Nofsinger]: Alright let s jump over to the other mental shortcut-familiarity. People think that the things that they re familiar with are better. Mkay they prefer things that are familiar. And better is an interesting word because better implies low risk is better. High risk is bad. High return is better though right. And so that s interesting because high returns better and low risk is better but they don t go together right. Expected return and risk are positively related not negatively related. So that can trick us, and we ll talk more about that. Familiarity Another mental shortcut People prefer things that are familiar to them In many cases, people believe the things they are familiar with are better than those things of which they are not familiar 5

6 [John Nofsinger]: There is lots to choose from out there-thousands of stocks you can choose from in the United States. More outside the United States. Which stocks do we pick? We tend to pick stocks that we re familiar with. Familiarity Breeds Investment There are 10 s of thousands of stocks in the U.S. That many outside the U.S. as well Which stocks do we pick? The ones for which we are familiar [John Nofsinger]: For example, if you look at professional money managers portfolios and you look at where the company s they buy are located, the companies they buy are 100 miles closer on average to them than the average company is to them. So they buy companies that are closer geographically to them. Familiarity Bias A typical money manager s portfolio consists of firms that are located 100 miles closer to the manager s office than the average US firm [Image of map of New York and the surrounding area] [John Nofsinger]: We see the same things with individual investors. 30% of the individual stock portfolio is invested in firms within 250 miles of their home. So that s a substantial portion. The other portion tends to be companies that obviously aren t headquartered as close, but are big nationwide well-known products you know that everyone knows. And we see this effect in the United States and the Swedish markets and in Finland. But the interesting thing is that Finland itself has two languages. The official languages are both Finnish and Swedish in Finland. And so companies can report their earnings in either language. And as it turns out, investors prefer companies that speak the same language. So another familiarity type of thing. And here s another interesting one. An investor right lives in a city, has a portfolio tilted toward stocks from that area-that city. What happens when they move to a different city? It turns out that within the next year or two, they will rebalance their portfolio away from the old city companies to the new city companies. So as we get more familiar with the new companies, we end up thinking they re better and we switch to those. Local Bias 6

7 30% of investors stock portfolio is invested in firms located within 250 miles of their homes United States Sweden Finland Finnish investors prefer local and same language firms (Finnish versus Swedish) When an investor moves to a different city, they rebalance their portfolio Firms furthest from the new city are most likely to be sold Stock purchases are tilted to a new city local firms [John Nofsinger]: Another thing about familiarity is the international aspect to it. We think that we re more familiar with the companies in our own countries that we think that they re better. Now portfolio would suggest that you know if we wanted the global market portfolio, then we d have only half of our stocks or so in the U.S. and the other half from outside the U.S. But that doesn t seem to be the case. You know some studies show less than 5% of are assets are in international stocks. We call that the home bias. Familiarity Breeds Home Bias Only half of worldwide stock equity is in the United States Portfolio theory suggests 30% to 50% of a portfolio should be in foreign securities (depending on risk level desired) The average international holdings is less than 5% Called the home bias Considered due to the familiarity problem [John Nofsinger]: We can see that that occurs everywhere. This dark line shows that the United States has 43% of the world s stock market in the U.S followed by 9.7% of the Japanese stock market. And then the U.K. on and on. So that s the portion of the world s stock market in each country. Now you look at investors in that country invest. Well investors put almost 90% in the United States of their money in the United States even though that s double of what they should from a global market portfolio perspective. And of course even worse in a lot of other countries. The Spanish put 91% of their money in Spanish firms even though Spanish firms represent only 2 and a half percent of the world capitalization. So you can see that home bias effects everyone. Familiarity Breeds Investment [Graph of different countries vs. World Market Cap Share and Domestic Equity Share] 7

8 [John Nofsinger]: But where do you put your money if you are going to put it internationally? Here s what the Germans do. They put 13% of their international money in the United States companies. Almost the same amount in France, and then a lot in the U.K. So what explains that? Well it would make sense that you would put a lot in the United States. After all, the United States represents almost half of the world capitalization. But France is only 4% of the world s capitalization. Why would you put so much of your money there? Switzerland is only 2% and you put so much there. And the same with the United Kingdom. Well they are very familiar right. Here is the opposite. Japan represents about 10% of the world capital market, and they only put about 5% of their portfolio in there. So they underrepresent in Japan. Why? Because investors when they go outside their country still want to be familiar. They want a familiar geography, language, religion. These are all impacts. When you do invest outside your country, where do you invest? [Graph of different countries and their allocation of German Funds to Equities in Other Countries] Investors invest in countries that are familiar: geography, language, religion [John Nofsinger]: Here s another one. The companies that-well let me start over. The familiarity bias says better. We like things that are more familiar because we think they re better. And we think high return is better and low risk. But they don t go together. We know that and yet that s still how we behave. If you look at the Fortune 500-the Fortune Magazine surveys executives and analysts. And they asked about the reputations of different companies, and they asked them to rank them just on a scale from you know like one to ten. I guess zero to nine on what kind of return those companies are gonna get. And then they also asked them to do a risk score. Again it s just a scale of zero to ten. And so this represents you know an average this company they feel is pretty high risk but pretty low return. And the line I put in here kind of gives us like a regression line. And what is wrong with that line? What s wrong is that it s going in the wrong direction. It s suggesting that we believe that low risk companies are gonna earn a high return. Right the line should be this way. Where return and risk go together. But we don t really think that way. We think this company is better. Therefore it must have a high return and it must have a low risk interestingly enough. If we compare that also-those risk scores-to the Fortune longterm investment value index, we also get that it is on that it is a line going the wrong direction. High Expected Return and Low Risk? 8

9 Each year, Fortune surveys executives and analysts on the reputations of companies. For the 2007 survey, investors were asked to rate these firms expected return and risk on a scale of 1 to 10. [Graph of Risk Score vs. Expected-Return Score]\ These risk scores are compared to the Fortune Score (Long-Term Investment Value). [Graph of Risk Score vs. Fortune Score] [John Nofsinger]: What company are you most familiar with? Probably the one you work for. Morningstar did a very interesting question on their webpage. They said which is more likely to lose half of its value? Your company that you work for or the overall stock market? Now we know that any one company has a lot of firm specific risk left in it, right. A diversified portfolio like the market is gonna have a lot less volatility. Therefore you should always say your company is more likely to lose half its value. It s also more likely to double its value than the overall market. But that s what that volatility means right. Only 16% of the people really understood that. They thought their company was as safe or safer than the stock market in general. Vanguard did a similar one and talked about more risky, less risky. And only a third of the people recognized that their employer stock is more risky than the overall market. So we re not understanding risk very well because we tend to think of it as better not as high risk, low risk. Familiarity Breeds Investment Problems What companies are people most familiar with? Probably their employer... A survey on Morningstar.com: Which is more likely to lose half of its value, your firm or the overall stock market? (1,000+ responded, only 16.4% stated their company) Survey to Vanguard pension clients, 2003 Would you say your employer s stock is MORE RISKY, LESS RISKY, or has ABOUT THE SAME LEVEL OF RISK as an investment in a diversified stock fund with many different stocks? Employer s stock is Riskier: 33% About the same risk: 39% Less risky: 25% 9

10 [John Nofsinger]: So we put our money and to invest our 401k plan money right. And a lot of times we have the option in investing in the company stock itself. And this is very dangerous obviously. If the company goes bankrupt, you d lose your job and your retirement money. But we tend to do it a lot anyway. We ve actually seen this occur when Enron and WorldCom and K- Mart and on and on. We put too much-42% of our money we put in of our 401k assets-in the company stock. Too much, right-way too much. That s even higher when our stock has done well in the past right. So there s an extrapolation bias as well. 401(k) Investments A survey of 242 SP500 companies found that 42% of the 401(k) assets were invested in the company stock! Danger! If your company files for bankruptcy, you are likely to lose your job and your 401(k) may be worthless Employees are more likely to invest in their company s stock if the stock has performed well recently Familiarity and representativeness [John Nofsinger]: This is how much for a large set of like S&P500 companies-how much-what is the percent of equities and total investment is your company stock? K so this represents that on average, half of the people-half of the money people have in their 401k stocks-half of that money is in the company stock right. And the other half is more of a diversified portfolio. Or you could say some bonds and bond mutual funds too. So you might want to compare the amount of employer stock they own to the entire 401k balance not just the equity portion. And that would be here. And so you can see that there s some volatility over time-bond markets go up. But we still have a very high portion of our money in the company stock. Then all of a sudden the news comes out about Enron. We see stuff like that all the time-other companies-how people are 55 years old fired because the company went bankrupt. And now they ve lost all their retirement savings because they had all of their money in Enron. K that was in this time period here. So what did we all do? Did we go run out there and go to our human resources division and say oh man, I gotta sell the company stock and diversify. I don t want to be like that Enron guy. Well no, not really. I mean the percentage goes down a little bit, but it s still within the range that we ve seen in history. So we don t really learn too well from other people s mistakes. Learning from other people s mistakes Percent of company stock in the 401(k) plan over time 10

11 Percent of equities and percent of total 401(k) account [Graph of years vs. percent of equities and share of all balances] [John Nofsinger]: So just to finish this up. This fourth video for the psychology of investing. We ve talked about the representativeness bias. We ve done a lot about familiarity as well. These biases help us pull the trigger. In other words, it does help us make decisions. We need mental shortcuts. But they lead us to under-diversification, too much in one stock, poor asset allocation type decisions. And it also therefore we end up taking too much risk for the level of return that we are receiving. So that finishes up the video. Familiarity Bias Outcomes Allows people to pull the trigger Can lead to under-diversification Too much in one stock Poor asset allocation Result: taking too much risk for the level of return received 11