Tuesday, June 21, 2016

Misbehaving: The Making of Behavioral Economics

Richard Thaler’s book, Misbehaving: The Making of Behavioral Economics, is both a fascinating look at the way humans make economic decisions and an interesting account of the history of this field within economics. Despite being an easy read, this book teaches important lessons.

I had no idea that claiming humans are not completely rational was once controversial in the field of economics. The term used for a person who makes perfectly rational financial decisions is “Econ.” The entire field of economics was built on the notion that we are all Econs. Of course, we aren’t. Economists used to deny that our irrationality had any serious impact on markets, but Thaler devoted his career to showing how our mistakes are an important part of economics.

One interesting finding is that “people who are threatened with big losses and have a chance to break even will be unusually willing to take risks, even if they are normally quite risk averse.” Perhaps this is the source of the common offer of “double or nothing.”

One part of the book was particularly tough on dividend investors. Thaler says that “in a rational world,” preferring dividends to capital gains “makes no sense.” “A retired Econ could buy shares in companies that do not pay dividends, sell off a portion of the stock holdings periodically, and live off those proceeds while paying less in taxes.” He goes on to describe a preference for stocks paying high dividends as “silly.”

Given a chance to take a fair coin flip to win $200 or lose $100, many people who would not do this once say they’d be happy to do it 100 times. Thaler explains why this is irrational, but notes that people think this way anyway.

Other research shows that people become less willing to take on risk if they see frequent losses. Combining this fact with the up-and-down nature of the stock market, we find “that the more often people look at their portfolios, the less willing they are to take on risk, because if you look more often, you will see more losses.”

Thaler explains a number of problems with strong forms of the efficient markets hypothesis, but notes that despite mispricings, “investors who attempt to make money by timing market turns are rarely successful.” This seems to be a common theme in the stock market. We can find anomalies, but it’s hard to profit from them.

The book contains humour as well. In a section analyzing the value of NFL draft picks, Thaler tells the story of the Redskins trading away several picks over multiple years to the Rams to get quarterback RG3. At the beginning of a game between the two teams, “the Ram’s coach sent out all the players they had chosen with the bonus picks to serve as team captains for the coin toss that began the game. The Rams won the game 24-0 and RG3 was sitting on the bench due to poor play.”

Thaler discusses the importance of “nudges,” which are ways to make it easier for people to do the right thing. Translating to the Canadian tax system, he suggests that RRSP contributions would increase if people could direct their tax refunds to their RRSPs and have the contribution count “for the return being filed (for the previous year’s income).”

Another amusing and likely effective nudge is when people need to “leave for higher ground before a storm strikes ... offer those who opt to stay a permanent marker and suggest they use it to write their Social Security number on their body, to aid in the identification of victims after the storm.”

It’s tempting to laugh at the kinds of mistakes people make and feel confident that we don’t make such mistakes. Of course, this isn’t true. It makes sense to try to make the choices an Econ would make, but we are doomed to fail sometimes. Hopefully, by pointing out the ways we make mistakes, Thaler is helping us make better choices.

31 comments:

  1. I love behavioural economics and think everyone would benefit from at least some understanding about our relationship with money (both general and specific). I'll even go as far as saying there is more value in studying behavioural econ than studying "the markets".

    But this will not happen any time soon. People, in general, have a difficult time facing themselves. It's much easier to try and control (and blame) externalities. Far easier to change investments than behaviour. Thus, we have 85,000 books on "personal finance" and less than 5,000 books on "behavioural economics".

    ReplyDelete
    Replies
    1. @SST: It's easy to despair when you make only a small dent in the world, but I'm content if I think I've helped even just one person, especially if that person is me :-)

      Delete
  2. Dividends as a way of life is SILLY?!?! HERETIC!!! Unbeliever!!!

    As long as humans are involved in a system, it is always entertaining.

    ReplyDelete
    Replies
    1. @Alan: Hopefully, the imminent book-burning won't include e-books.

      Delete
  3. "Thaler says that “in a rational world,” preferring dividends to capital gains “makes no sense.” "

    The reason people take dividends is precisely because it's not a "rational world". In a rational world, the company which retains earnings (i.e. does not pay a dividend) would take the most rational action with their money...as we all know, even companies act irrationally. The question is, who will do the most rational thing with the money -- the dividend-reaping investor or the earnings-retaining company?

    ReplyDelete
    Replies
    1. @SST: There's another important factor: taxes. For those in an accumulation phase, realizing taxes hurts returns. So, just believing you're more rational than the companies you invest in is not a good enough reason to prefer dividends.

      Delete
    2. A friend once told me he preferred dividend payers because they were less likely to be cooking the books.

      ie, with an 80% dividend payout, it's hard to be bluffing. With no dividends... who knows what the true story is.

      Delete
    3. @Anonymous: There is certainly a limit to how long you can bluff if you're paying a substantial dividend. The trouble is that once the bluff gets found out, the company is bankrupt. I think that dividend or no, you're facing similar risks. If Nortel had paid a bigger dividend, they would have imploded sooner, but investors who reinvested their dividends would have lost it all either way.

      Investors worried about conspiracies sometimes focus on a company's cash flow. It's harder for creative accountants to play games with cash flow than with reported revenue and earnings.

      Delete
  4. I really enjoyed this book as well. You might enjoy "Predictably Irrational" (and anything else by Dan Ariely) and "Your Money and Your Brain" by Jason Zweig if you liked this.

    ReplyDelete
    Replies
    1. @John: I enjoyed both books. Unfortunately, I didn't do full reviews back than and just picked out one part to write about:

      http://www.michaeljamesonmoney.com/2008/09/be-first-to-order-in-restaurant.html

      http://www.michaeljamesonmoney.com/2008/10/two-bad-stock-market-days-in-row.html

      Delete
  5. So why is the "win $200-lose $100" 100 times vs 1 time irrational? This seems like the kind of thing that I might agree to do... :(

    ReplyDelete
    Replies
    1. @Anonymous: If you're unwilling to take this bet once, then why would you do it the 100th time after the first 99? And if you're not willing to do it the 100th time, then the 99th is out as well. Proceeding like this, you're not willing to do it at all.

      There are various behavioural quirks at play that make us willing to take the bet 100 times, but not once. We hate to have a chance of losing money even when we could win twice as much. And once we've won some money (with high probability in the first 99 trials), we're more relaxed about playing with "house money." These things are irrational, but they're part of how we think.

      Delete
    2. The way I understood the offer to flip 100 times, choosing to stop after 99 was NOT an option. To be a looser at this game, you got to loose 67x over 100 attemps wich is very unlikely to happen if it's a fair flip. But really, the 100th flip is not rational, neither the 99th and so on.

      Delete
    3. @Le Barbu: That would be a different game. Let's say I can't decide whether I'll let you flip once or flip any number of times up to 100. Many people would reject the offer of just one flip, but would choose 100 flips in the second case. This is irrational, but feels right to many of us.

      Delete
    4. I probably do not understand the rules of this game. If you give me the option to flip once, my outcomes are 200$ or -100$. If I flip 100x I will likely end up with anything between 3,500$-6,500$ wich is a lot better than 200$.

      Delete
    5. @Le Barbu: You don't get to choose the version of the game; I do. Suppose I offer one flip to someone and he says no. But he didn't know I planned to then offer any number of flips up to 100. Now he says he'll take 100 flips. If the 100 flips are a good idea, then he should have taken the offer of only one flip.

      Delete
    6. Ok, I get it now!

      Delete
    7. Yeah, I'm not sure I'm convinced. If you line up a hundred people to play each game, roughly half of the single-flip folks will have lost money. Of the hundred people playing the 100-flip game, how many would you expect to have lost money? Not half of them... maybe even none of them.

      Delete
    8. To use an analogy that's closer to what you normally talk about in your blog to illustrate: following your logic, if you're happy putting all your retirement savings in a portfolio of 5000 stocks, you should be perfectly happy putting all your savings in 1 stock.

      No?

      Delete
    9. @Anonymous: Focusing solely on whether or not you've lost money and ignoring how much money you've won or lost is an irrational behavioural quirk that many humans share.

      The attempted analogy to diversification of stocks does not apply, I'm afraid. If I'm investing $500,000 in 5000 stocks, that's $100 per stock. It makes sense to say that if I'm willing to put the $500,000 into 5000 stocks, I should be willing to put $100 into one stock. But it does not follow that I should be willing to put the whole $500,000 into one stock.

      Delete
    10. Ok, I'll accept your point about the analogy.

      But, I'm not ignoring how much I win or lose. Since your expected return is positive, you really only care about how much you can lose and how likely you are to lose (unless you could be making more money doing something else, but let's assume that we're flipping coins quickly enough). However, the more flips there are, the better off you are. One flip, and you have a 50% chance of being up $200 or down $100. Two flips, and you have a 25% chance of being down $200, and 75% chance of being up either $100 or $400 (either is a win, so I don't really care which it is). So you could lose twice as much as before, but the probability of that happening is half as much. Three flips, and you could lose $300, but there is only a 12.5% chance of that happening, and an 87.5% chance of breaking even or better. Your maximum loss is increasing linearly, but the probability of that happening is decreasing at a much faster rate. Meanwhile, your expected return is also increasing linearly. At 100 flips, yes, you could at worst lose (or win) $10k, but the probability of losing (or winning) that much is 7.9e-31... AND the probability of losing ANY money at all is 0.044%. Meanwhile your expected return has also grown to $5000. Increase the number of flips to 1000, and the probability of losing ANY money at all is down to 1.07e-26, and your expected return is $50k.

      So the more flips, the more likely you are to come out ahead (and therefore less likely to lose anything), and the more you flip, the higher your expected return. So the more flips you offer me, I think it becomes more and more rational to accept.

      Delete
    11. @Anonymous: Your reasoning is a good example of prospect theory, which is a model for how people make these types of decisions. Unfortunately, this is a model of our irrationality.

      I'm not sure I can think of another way to explain all this. It is definitely true that the more flips you take, the better off you are. However, the process is very close to linear. 100 flips are a good idea because each flip is a good idea. The proof that each flip must be a good idea comes from the fact that if you were given the option to stop after 99 flips, would you take it? After all, the result of the 99 flips is in the past. You're facing the 50% possibility of losing $100. If you take into account previous results to decide whether to take the last flip, this is irrational. If you make a rational decision to take the last flip, this means you should have been willing to take just one flip from the beginning.

      Delete
    12. Yeah, okay, I see what you're getting at. Thanks for taking the time. :)

      Delete
    13. @Anonymous: No problem. I find these issues similar to the visual puzzles where you can measure two lines and know that they are the same length, but you can't prevent your visual system from being "sure" that one line is longer. My gut might tell me not to take a certain coin flip bet even though I know from the math that I should take it.

      Delete
  6. "One flip, and you have a 50% chance of being up $200 or down $100. Two flips, and you have a 25% chance of being down $200, and 75% chance of being up either $100 or $400..."

    Is this correct?

    I'm no mathematician, but isn't there some kind of probability of independent events rule (i.e. it will always be 50/50 "to win $200 or lose $100" not matter how many flips)?

    No need to explain in detail, just yes or no. :)

    ReplyDelete
    Replies
    1. @SST: Both are correct. The flips are independent. The equally-likely possibilities are ++, +-, -+, --. This is 25% -$200, 50% +$100, and 25% +$400. Each individual flip remains +$200 or -$100 with 50/50 odds. Not sure if this answers your question.

      Delete
  7. Yup, thanks.

    As this thread has demonstrated, behavioural economics is a tough hurdle.

    ReplyDelete
  8. Sounds like an interesting book. How does it compare to Thinking Fast and Slow? Although not about economics you might want to check out Steven Pinker's book How The Mind Works. Very detailed and interesting.

    ReplyDelete
    Replies
    1. @Mark: I liked this book but I liked Thinking Fast and Slow more. They are quite different even though they cover some similar material.

      Delete
  9. I wonder how everyone's brehavioural breconomics misbehaved amid the Brexit chaos?

    Full disclosure: I sold precious metals and bought public equities.

    ReplyDelete
    Replies
    1. @SST: Before the vote results or after? In my case, I did nothing. But I am now bemoaning the fact that I have no appreciable amount of cash lying around to buy some beaten up ETFs. I don't keep "dry powder" available for market selloffs because I find the opportunity cost too high, but it would have been nice if I just happened to pick today to invest some cash that had been building up.

      Delete