WEBVTT 00:01.710 --> 00:09.470 Professor Robert Shiller: Today's lecture is about 00:09.471 --> 00:17.931 behavioral finance and this is a term that emerged into public 00:17.925 --> 00:23.325 consciousness around the mid-1990s; 00:23.330 --> 00:25.390 before that it was unknown. 00:25.390 --> 00:28.730 The term "efficient markets" is much older; 00:28.730 --> 00:32.760 I mentioned the idea goes back to the nineteenth century and 00:32.758 --> 00:35.078 the term goes back to the 1960s. 00:35.080 --> 00:39.900 But behavioral finance is a newer revolution in finance and 00:39.897 --> 00:44.297 it's something that I have been very involved with. 00:44.300 --> 00:48.590 I have been organizing workshops in behavioral finance 00:48.587 --> 00:52.307 ever since 1991, working with Professor Richard 00:52.308 --> 00:55.218 Thaler at University of Chicago. 00:55.220 --> 00:57.810 We've been doing that for eighteen years; 00:57.810 --> 00:59.930 amazing, that's a long time for you, right? 00:59.930 --> 01:06.030 When we started we were total outcasts, we thought; 01:06.030 --> 01:08.160 nobody appreciated us. 01:08.159 --> 01:11.719 I had tenure so I could do it but the problem is, 01:11.719 --> 01:16.019 you don't want to do things that are too out of fashion. 01:16.019 --> 01:18.489 Fortunately, we have a system that allows it 01:18.491 --> 01:21.021 to happen and I'm very happy to have that. 01:21.019 --> 01:26.779 What behavioral finance is a reaction against extreme--some 01:26.779 --> 01:32.339 extremes--that we see in efficient markets theory or also 01:32.339 --> 01:35.119 in mathematical finance. 01:35.120 --> 01:39.760 Mathematical finance is a beautiful structure and I admire 01:39.759 --> 01:44.149 what the people have done and I've worked in it myself, 01:44.153 --> 01:46.273 but it has its limits. 01:46.269 --> 01:50.209 Eventually--you know the way a paradigm develops--it goes 01:50.210 --> 01:52.040 through a certain phase. 01:52.040 --> 01:55.430 When mathematical finance was new, say in the 1960s, 01:55.434 --> 01:58.964 it was the exciting thing and nobody wanted to work on 01:58.961 --> 02:01.421 anything else; you wanted to be doing the 02:01.416 --> 02:04.716 exciting thing. As the '70s and '80s wore on, 02:04.715 --> 02:07.845 it got to be a little bit overdone; 02:07.849 --> 02:11.029 people run with it too far, they think that's all we want 02:11.030 --> 02:14.210 to do, and we don't want to think about anything else. 02:14.210 --> 02:16.860 Then they start to get sometimes a little crazy. 02:16.860 --> 02:23.370 Than we had to reflect that, well, things aren't perfect. 02:23.370 --> 02:29.680 The world isn't perfect and we have real people in the world, 02:29.681 --> 02:33.891 so that led to the behavioral finance. 02:33.889 --> 02:37.959 Behavioral finance really means--what does it mean? 02:37.960 --> 02:40.810 It's not like behavioral psychology. 02:40.810 --> 02:43.710 It doesn't mean behavioral psychology applied to finance. 02:43.710 --> 02:46.150 It really means something much more broad than that. 02:46.150 --> 02:49.750 It means all of the other social sciences applied to 02:49.753 --> 02:53.703 finance. The economics department is 02:53.702 --> 03:01.032 just one of many departments in the university that teaches us 03:01.027 --> 03:05.467 something about how people behave, 03:05.469 --> 03:09.829 so if we want to understand how people behave we can't rely only 03:09.832 --> 03:11.912 on the economics department. 03:11.909 --> 03:14.879 I think that it's coming around to a unifying of our 03:14.878 --> 03:17.748 understanding. Since then--since the 03:17.754 --> 03:21.924 beginnings in the '90s, our behavioral finance 03:21.921 --> 03:25.441 workshops have grown and grown and, 03:25.439 --> 03:28.019 of course, so many people are involved in it now; 03:28.020 --> 03:30.120 it's now very well-established. 03:30.120 --> 03:34.240 Before I get into that, I want to give some additional 03:34.238 --> 03:36.878 reflections on the last lecture. 03:36.879 --> 03:40.949 I have this chart, which you saw last 03:40.948 --> 03:47.618 time--actually it's an Excel spreadsheet that--I also put it 03:47.615 --> 03:54.955 up already on the classes V2 website so you can play with it. 03:54.960 --> 03:59.430 I just want to reflect again--I know I'm repeating myself a 03:59.430 --> 04:02.360 little bit, but it's very important. 04:02.360 --> 04:07.860 What we have in this chart is the blue line, 04:07.857 --> 04:14.627 which is the Standard & Poor Composite Stock Price 04:14.633 --> 04:20.773 Index going back to 1871--from 1871 to 2008, 04:20.769 --> 04:26.329 right now--so that's like 130 years of data. 04:26.330 --> 04:28.070 That's the blue line. 04:28.069 --> 04:30.329 You can see the--do you know what that is there? 04:30.329 --> 04:36.209 That's 1929 and that is the Crash of 1929. 04:36.209 --> 04:40.229 Well, actually it extended to 1932 and you can see other 04:40.233 --> 04:41.773 historic movements. 04:41.769 --> 04:47.639 There's the bull market of the 1990s--a very big upswing--and 04:47.636 --> 04:51.836 then there's the crash from 2000 to 2003. 04:51.839 --> 04:54.799 I don't know if you remember these things, 04:54.799 --> 04:58.479 they were big news, not as big as the 1929 crash, 04:58.480 --> 05:00.810 but the upswing was just as big as the 1920s upswing, 05:00.811 --> 05:03.421 wasn't it? Here's the 1920s upswing and 05:03.416 --> 05:07.546 here's the 1990s upswing--huge upswing in stock prices. 05:07.550 --> 05:12.410 This is in logs, by the way, so that means that 05:12.405 --> 05:18.315 everything--the same vertical distance refers to the same 05:18.315 --> 05:22.005 percentage change in the price. 05:22.009 --> 05:27.999 Then I had, as I said last period, I have a random walk 05:28.001 --> 05:31.331 shown--that's the pink line. 05:31.329 --> 05:36.429 The random walk is generated by the random number generator. 05:36.430 --> 05:39.000 I fixed the random number generator, so I made it truly 05:39.004 --> 05:39.914 normal this time. 05:39.910 --> 05:45.340 It slows it down a little bit, but if you press F9 we get 05:45.341 --> 05:50.191 another random walk, but it's always the same stock 05:50.190 --> 05:53.380 price. This is a random walk with a 05:53.375 --> 05:57.405 trend that matches the uptrend of the stock price. 05:57.410 --> 06:04.770 I can press--it kind of looks similar, doesn't it? 06:04.769 --> 06:09.029 It kind of shows that in some basic sense the stock market and 06:09.026 --> 06:11.186 the random walk are the same. 06:11.190 --> 06:14.860 06:14.860 --> 06:19.250 Here we have the crash of--here we have the market peak of 1929 06:19.252 --> 06:23.362 except it turned out in this simulation to have occurred in 06:23.360 --> 06:24.990 1910 or thereabout. 06:24.990 --> 06:28.220 Then we have the--that's The Depression of the '30s except 06:28.224 --> 06:29.364 it's not the '30s. 06:29.360 --> 06:32.340 I can just push a button and we get something else. 06:32.340 --> 06:34.360 I find this amusing. 06:34.360 --> 06:36.670 I don't know. Unfortunately, 06:36.665 --> 06:43.345 we live through only one of these in our lifetime. 06:43.350 --> 06:47.420 There's a TV show about parallel universes, 06:47.415 --> 06:49.805 right? What's the name of that show? 06:49.810 --> 06:50.840 I can't remember it. 06:50.840 --> 06:52.980 Don't you know this show? 06:52.980 --> 06:56.290 Where they go in some kind of time machine and they emerge in 06:56.289 --> 06:59.709 another parallel universe where history took another course. 06:59.709 --> 07:02.529 Well anyway, these are parallel universes 07:02.532 --> 07:05.642 that we see. In some of these universes, 07:05.640 --> 07:10.120 Jeremy Siegel would write his book, Stocks for the Long 07:10.116 --> 07:12.776 Run, and in some of them he would 07:12.778 --> 07:16.118 not because–well, this one he might not because 07:16.119 --> 07:19.649 in this case the stock market was just declining for the 07:19.652 --> 07:21.582 better part of a century. 07:21.579 --> 07:26.339 The thing I don't see in these charts and I think we haven't 07:26.338 --> 07:31.178 captured it perfectly with just the standard random walk is I 07:31.177 --> 07:35.127 don't see any crash as big as the 1929 crash. 07:35.130 --> 07:36.220 It's hard to get them. 07:36.220 --> 07:40.620 I keep pushing F9--this just seems to dominate, 07:40.624 --> 07:43.144 right? There's nothing as big 07:43.135 --> 07:47.575 here--press F9 again--you can keep pushing and pushing, 07:47.579 --> 07:51.249 maybe you'll get one but you have--you get the idea that 07:51.245 --> 07:54.705 there's something anomalous about that crash from the 07:54.710 --> 07:57.510 standpoint of this random walk theory. 07:57.510 --> 07:59.410 I'm not getting one, right? 07:59.410 --> 08:04.820 That's something that we'll talk about. 08:04.819 --> 08:08.699 I would--I'm not--I can push for a long time and I don't 08:08.699 --> 08:11.309 see--well there's a pretty big one. 08:11.310 --> 08:16.960 Isn't that just about as--not quite as sharp as the 1929 08:16.956 --> 08:20.546 crash, but it's hard to get them. 08:20.550 --> 08:23.660 I think that one thing–there are a couple 08:23.663 --> 08:26.103 of things that we'll come back to. 08:26.100 --> 08:28.480 One is--I think I've already mentioned it--fat tales. 08:28.480 --> 08:32.730 Stock price movements have a tendency to show some extreme 08:32.731 --> 08:36.161 outliers that are not represented by the normal 08:36.162 --> 08:39.982 distribution. But also, there are variations 08:39.976 --> 08:43.316 in the variance. So, in this period here--in the 08:43.322 --> 08:46.882 '20s and '30s--the stock market was extremely variable on a 08:46.882 --> 08:49.762 day-to-day basis; it was way beyond anything 08:49.762 --> 08:51.162 we've observed since. 08:51.159 --> 08:55.279 So, that's why it seems to be more volatile in that period 08:55.281 --> 08:58.971 because the accumulation of bigger random shocks. 08:58.970 --> 09:05.660 Anyway, we can play this game for a while but now I want to go 09:05.658 --> 09:12.128 and talk about--remember that the random walk that we see in 09:12.127 --> 09:17.387 stock prices is not the behavior of a drunk, 09:17.389 --> 09:20.639 even though you can describe a random walk as drunken behavior. 09:20.639 --> 09:25.639 The idea in the theory is that these movements only appear 09:25.641 --> 09:31.171 random because they're news and news is always unpredictable. 09:31.169 --> 09:34.449 If the market is doing the best job--this is efficient 09:34.445 --> 09:36.665 markets--in predicting the future, 09:36.669 --> 09:41.749 that means then that any time the stock market moves it's 09:41.750 --> 09:45.380 because something surprising happened. 09:45.379 --> 09:48.619 Like there might be a new breakthrough in science or there 09:48.623 --> 09:51.183 could be war or something outside--this is the 09:51.184 --> 09:54.774 story--outside of the economic system that disrupts things. 09:54.769 --> 09:57.679 The next question then–now, 09:57.677 --> 10:03.217 I've added something--it's on this little tab here--I've added 10:03.218 --> 10:07.668 something, which is a plot of present values. 10:07.669 --> 10:11.969 This is something that I published in 1981. 10:11.970 --> 10:13.700 That's a long time ago, isn't it? 10:13.700 --> 10:17.510 It was my first big success. 10:17.509 --> 10:20.699 Not everyone liked this article, but what I had--I got 10:20.696 --> 10:22.556 into a lot of trouble for it. 10:22.559 --> 10:26.869 I learned some people react with hostility when you offend 10:26.866 --> 10:30.636 their cherished beliefs, so I was on the outs for a 10:30.644 --> 10:32.764 while with this article. 10:32.759 --> 10:36.419 I said, it's kind of interesting to think that all 10:36.424 --> 10:40.394 these apparently random movements are really resulting 10:40.387 --> 10:44.427 in news about something that is fundamental--that's the 10:44.425 --> 10:46.215 efficient markets. 10:46.220 --> 10:50.680 Every time the stock market moves it's because there was 10:50.683 --> 10:52.553 some news about what? 10:52.550 --> 10:54.380 Well, it's about present value. 10:54.379 --> 10:57.519 The efficient markets theory, in its simplest incarnation, 10:57.517 --> 11:00.707 says that the price is the expected present value of future 11:00.709 --> 11:03.349 dividends. What I did, in a paper that I 11:03.354 --> 11:06.524 published in 1981, is I said, well let's just plot 11:06.521 --> 11:09.561 the present value of dividends through time. 11:09.559 --> 11:13.879 That's how I constructed this long time series back to 1871; 11:13.880 --> 11:15.410 nobody else was looking at it. 11:15.409 --> 11:20.279 Typically, researchers want the best data, the high quality 11:20.278 --> 11:24.138 data, and so they would look at recent data, 11:24.139 --> 11:28.209 which was the best data, and they would think going back 11:28.207 --> 11:31.607 to 1871 is crazy because that's so long ago. 11:31.610 --> 11:34.500 We have daily or minute-by-minute data by now, 11:34.500 --> 11:37.390 but we can't get it for that remote period. 11:37.389 --> 11:39.339 On the one hand, as I argued, 11:39.344 --> 11:43.114 the stock market is pricing things that occur over long 11:43.113 --> 11:46.583 periods of time. The present value formula is 11:46.580 --> 11:50.280 pricing dividends into the future, decades into the 11:50.277 --> 11:52.477 future--well, actually to the infinite 11:52.483 --> 11:55.003 future, but most of the weight is on the next few decades. 11:55.000 --> 11:57.890 So, we can't evaluate the theory by just looking at ten 11:57.892 --> 12:00.412 years of data we've got to get a lot of data. 12:00.409 --> 12:04.319 What I did then in that paper was I computed the actual 12:04.320 --> 12:08.520 present value of subsequent dividends for each year--that's 12:08.519 --> 12:12.429 on this tab--and compared it with the stock price; 12:12.430 --> 12:14.990 so that's what I did. 12:14.990 --> 12:20.470 This is an update of a plot that I showed in my 1981 12:20.466 --> 12:24.006 American Economic Review paper. 12:24.009 --> 12:27.429 The blue line--because when I published it I was right here. 12:27.430 --> 12:31.710 It's amazing how time goes by; it was 1979, I was right here. 12:31.710 --> 12:38.710 We had just come off from the big stock market drop of the--it 12:38.712 --> 12:45.372 was the '73-'75 drop and it was a couple of years later, 12:45.370 --> 12:47.420 so we were kind of bumbling around down here. 12:47.419 --> 12:51.139 We didn't have any idea whether this was coming at the time. 12:51.139 --> 12:53.619 What I did was I just, for each year, 12:53.615 --> 12:56.705 computed the present value of the dividend. 12:56.710 --> 12:58.440 I have a dividend series for every year. 12:58.440 --> 12:59.990 In fact, it's right over here. 12:59.990 --> 13:09.170 I have to--this is the data, so I have the--this is the 13:09.169 --> 13:15.519 S&P Price Index monthly, back to 1871, 13:15.519 --> 13:21.479 and here are the dividends they paid per share every year since 13:21.480 --> 13:24.440 1871. I just, for each year, 13:24.442 --> 13:30.322 I took all subsequent dividends and I priced them out at the 13:30.324 --> 13:36.704 present value formula and I used the constant discount rate of 6% 13:36.703 --> 13:41.233 a year. So, you see how we get what the 13:41.229 --> 13:43.269 present value was. 13:43.269 --> 13:47.139 Of course, there's a problem because we don't know dividends 13:47.143 --> 13:51.283 after 2007 because we don't have data on dividends past then. 13:51.279 --> 13:54.429 But, I just made some assumptions, so the value at the 13:54.425 --> 13:56.675 end is maybe a little bit arbitrary. 13:56.679 --> 13:58.999 It could be dragged up or down if I made a different assumption 13:59.003 --> 14:00.093 about dividends at the end. 14:00.090 --> 14:02.460 More or less, this is going to be what 14:02.463 --> 14:06.183 actually the present value of dividends was over this whole 14:06.184 --> 14:09.314 period. Here's the dilemma--this is 14:09.310 --> 14:14.060 what I said in my 1981 article--this is the thing that 14:14.061 --> 14:16.931 is supposed to be forecasted. 14:16.929 --> 14:21.199 That's the present value and the blue line is the forecast of 14:21.195 --> 14:23.205 that thing. Then you ask, 14:23.211 --> 14:26.321 does this look like a good forecast? 14:26.320 --> 14:30.820 Were people doing a good job of forecasting the red line with 14:30.820 --> 14:33.660 the blue line? Now, that may be a loaded 14:33.662 --> 14:36.322 question; but, I think that you get the 14:36.322 --> 14:39.912 impression that there's something possibly wrong here 14:39.911 --> 14:43.991 with efficient markets because the red line is just a smooth 14:43.985 --> 14:48.265 growth path like nothing happens to it and yet the stock market 14:48.265 --> 14:51.505 is going up and down all over the place. 14:51.509 --> 14:56.739 It's a little bit like if you had a weather forecaster and 14:56.739 --> 15:01.379 this morning he says, I predict today that the low 15:05.897 --> 15:08.717 says, I predict that the low today 15:10.220 --> 15:13.130 You would eventually start concluding that this weather 15:13.128 --> 15:16.138 forecaster can't be trusted because we never get to those 15:16.144 --> 15:18.884 temperatures. That's sort of what the stock 15:18.875 --> 15:21.445 market is doing; it's fluctuating much more than 15:21.448 --> 15:22.858 the thing that's forecasted. 15:22.860 --> 15:26.870 You've got to be careful; I ended up with so many critics. 15:26.870 --> 15:30.360 There are lots of issues here that--some people said, 15:30.356 --> 15:34.376 well people don't know where the red line was last period. 15:34.379 --> 15:38.699 And other people said, well you just are showing one 15:38.701 --> 15:41.581 reality for the realization--you're 15:41.582 --> 15:47.262 showing--they kind of get back to this parallel universe story. 15:47.259 --> 15:50.429 There must be another universe where there's another Earth and 15:50.432 --> 15:53.502 where everything looks the same except that the red line did 15:53.501 --> 15:56.941 something very different; that could be. 15:56.940 --> 15:59.030 So, people are saying, you never know, 15:59.028 --> 16:02.078 there could have been a communist revolution in America 16:02.075 --> 16:05.455 in the 1930s and they could have nationalized the whole stock 16:05.462 --> 16:08.572 market and then the red line would be down at zero--they 16:08.566 --> 16:10.876 would have taken the whole thing. 16:10.879 --> 16:14.079 Or there could have been some good news, some great 16:14.083 --> 16:17.743 breakthrough that we haven't discovered yet but in another 16:17.735 --> 16:19.525 reality they could have. 16:19.529 --> 16:22.709 So, all this noise in the stock market could have somehow been 16:22.711 --> 16:25.321 new information about things that didn't happen. 16:25.320 --> 16:31.610 I think we're getting kind of philosophical when we go to 16:31.613 --> 16:34.233 that. The point is that we've never 16:34.226 --> 16:38.066 seen any movement in the present value of dividends that would 16:38.068 --> 16:39.578 justify the movement. 16:39.580 --> 16:42.290 If we knew the future with certainty, according to this 16:42.287 --> 16:45.247 model, then the stock market would behave like the red line, 16:45.246 --> 16:46.596 not like the blue line. 16:46.600 --> 16:52.640 16:52.640 --> 16:54.870 Well, anyway. For example, 16:54.868 --> 16:57.758 let's look at the Great Depression of the 1930s, 16:57.758 --> 17:00.958 at the very least I think this chart will reveal some 17:00.956 --> 17:03.536 misconceptions that some people have. 17:03.539 --> 17:07.059 The Great Depression of the 1930s was awful, 17:07.055 --> 17:09.005 right? I mean you hear these stories; 17:09.010 --> 17:10.320 I assume you hear these stories. 17:10.319 --> 17:15.929 We had 25% unemployment at the peak--it sounds really bad. 17:15.930 --> 17:17.880 We had people selling apples on the street; 17:17.880 --> 17:19.590 you must know these images, right? 17:19.589 --> 17:21.929 It sounds awful, but look what happened to 17:21.927 --> 17:23.977 p* in the Great Depression. 17:23.980 --> 17:26.010 I can hardly see anything. 17:26.009 --> 17:29.509 Well, what actually happened was businesses continued paying 17:29.506 --> 17:33.116 their dividends right through the whole Depression and some of 17:33.121 --> 17:37.821 them cut their dividends, but it was only for a few years. 17:37.819 --> 17:42.579 The present value of--the value of stock depends on what it pays 17:42.579 --> 17:45.449 out over decades not just next year. 17:45.450 --> 17:49.040 The stock market--if people knew the--even if they knew the 17:49.036 --> 17:52.236 depression was coming, they shouldn't have marked down 17:52.236 --> 17:54.946 the stock market so much, according to this simple 17:54.950 --> 17:57.610 efficient markets story--according to the present 17:57.608 --> 18:00.288 value story. At the very least, 18:00.287 --> 18:06.367 I think that this diagram helps you to see what is wrong or what 18:06.370 --> 18:09.170 simple theories are wrong. 18:09.170 --> 18:12.500 So, it must be that if the stock market is reacting to new 18:12.502 --> 18:15.252 information over all this century of history, 18:15.250 --> 18:17.880 it must have been new information about things that 18:17.883 --> 18:18.993 just didn't happen. 18:18.990 --> 18:23.830 It could be that an asteroid almost struck the Earth and then 18:23.834 --> 18:27.794 it just missed and so the stock market crashed. 18:27.789 --> 18:30.129 Then when it missed it came back up again, 18:30.125 --> 18:32.855 so we don't see any interruption in dividends. 18:32.859 --> 18:34.689 But it has to be something like that. 18:34.690 --> 18:39.830 The problem is, I can't think of anything like 18:39.830 --> 18:42.500 that. I don't think that any asteroid 18:42.500 --> 18:45.760 came close to the Earth--not close enough to be worried 18:45.759 --> 18:49.199 about--and I can't think the communist revolution had much 18:49.199 --> 18:52.519 chance of taking place in the United States--but you can 18:52.518 --> 18:54.508 imagine--so we don't know. 18:54.509 --> 18:57.679 Behavioral finance kind of tends to reach the opposite 18:57.682 --> 19:01.452 conclusion: that this volatility in the stock market is the sign 19:01.453 --> 19:04.433 of something else; it's some social force, 19:04.428 --> 19:08.038 some speculative bubbles, some activity that is not 19:08.035 --> 19:10.555 related to anything fundamental. 19:10.559 --> 19:14.379 The reason I got so much hostility when I wrote these 19:14.379 --> 19:16.949 papers is I was striking a nerve, 19:16.950 --> 19:20.190 I guess, because many people have developed these beautiful 19:20.189 --> 19:23.539 mathematical theories that said that the stock market was the 19:23.541 --> 19:27.061 optimal predictor of everything and I was saying the emperor has 19:27.060 --> 19:31.380 no clothes; so there were others like that. 19:31.380 --> 19:33.080 What is happening? 19:33.079 --> 19:35.719 What I'm coming around to think--maybe it's my cynical 19:35.717 --> 19:37.357 view, I've always been a cynic. 19:37.359 --> 19:40.439 I don't know if you are cynics or not, but I think people 19:40.436 --> 19:42.246 convinced themselves of things. 19:42.250 --> 19:45.420 They--people think they understand things better than 19:45.420 --> 19:47.360 they do. You spend your whole life 19:47.362 --> 19:50.562 looking at this one picture of the stock market and you think 19:50.560 --> 19:53.760 you have an explanation for all of it--all rational good--but 19:53.758 --> 19:56.368 it's just over-confidence that's doing that; 19:56.370 --> 19:57.390 it's an illusion. 19:57.390 --> 20:01.560 I want to talk about over-confidence and I thought 20:01.555 --> 20:04.865 I'd try–there's also no eraser. 20:04.870 --> 20:14.600 20:14.600 --> 20:22.710 Can you find another eraser? 20:22.710 --> 20:23.840 There's probably one in this closet. 20:23.840 --> 20:31.310 20:31.309 --> 20:35.789 I wanted to try an experiment of asking you a series of a few 20:35.786 --> 20:39.166 short questions. It's a game we'll play, 20:39.170 --> 20:42.630 which I'll need your cooperation with. 20:42.630 --> 20:45.750 So, these are questions about over-confidence. 20:45.750 --> 20:52.380 Actually, I just want you to try to give me 90% confidence 20:52.376 --> 20:57.836 intervals for the answers to these questions. 20:57.839 --> 21:01.199 Do you know what a 90% confidence interval is? 21:01.200 --> 21:07.970 It's, for example, if I were to ask you what--how 21:07.965 --> 21:13.175 many people are there in New Haven? 21:13.180 --> 21:17.630 I want you to not just give me a number, I want you to give me 21:17.630 --> 21:21.570 a range such that you're 90% sure that you're right. 21:21.569 --> 21:25.459 I could say, well, it's between 90,000 and 21:25.462 --> 21:29.642 100,000 people and I'm 90% sure I'm right. 21:29.640 --> 21:32.600 If you give me a true 90% confidence interval, 21:32.600 --> 21:35.890 then you should be right 90% of the time, right? 21:35.890 --> 21:40.700 What I'm going to do is give you a few questions and ask you 21:40.698 --> 21:45.338 for a 90--ask you to write down--you have a piece of paper 21:45.344 --> 21:48.364 there--a 90% confidence interval. 21:48.360 --> 21:51.920 I have five questions; this is just an experiment. 21:51.920 --> 21:58.040 The first one is about the Statue of Liberty. 21:58.040 --> 22:02.560 22:02.559 --> 22:07.369 What does it weigh in pounds--in tons? 22:07.370 --> 22:14.310 Good, thank you. So, weight. 22:14.310 --> 22:19.240 22:19.240 --> 22:21.500 Incidentally, just to remind you, 22:21.503 --> 22:24.363 a U.S. ton is 2,000 pounds--not a 22:24.361 --> 22:28.061 British pound, which is 2,240 pounds--and a 22:28.064 --> 22:30.184 ton is 907 kilograms. 22:30.180 --> 22:36.340 Can you write down on your paper your 90% confidence 22:36.341 --> 22:38.651 interval? For example, 22:38.652 --> 22:41.192 I won't use realistic numbers. 22:41.190 --> 22:43.830 If you thought it was--you might write down, 22:43.833 --> 22:47.403 it's between one pound and three pounds and that you're 90% 22:47.398 --> 22:49.548 sure it falls in that interval. 22:49.550 --> 22:53.330 I didn't say--its tons, tons. 22:53.329 --> 22:56.549 I'm asking--it's more than a pound--I'll give you a hint, 22:56.548 --> 22:58.558 it's in tons. Let me also say, 22:58.558 --> 23:00.638 we're not weighing the base. 23:00.640 --> 23:05.830 The Statue of Liberty stands on a tall edifice; 23:05.829 --> 23:08.769 we're not counting that but we're counting also the steel 23:08.765 --> 23:11.225 reinforcing that they put in a few years ago. 23:11.230 --> 23:14.110 Remember, the Statue of Liberty was getting weak and they were 23:14.109 --> 23:16.139 worried that something might topple down, 23:16.140 --> 23:18.610 so they reinforced it and we're counting that. 23:18.609 --> 23:23.439 So, it's a copper structure with steel reinforcement. 23:23.440 --> 23:28.640 Can you write down on your notes a range within which 23:28.640 --> 23:33.640 you're 90% sure that the statue weighs, in tons? 23:33.640 --> 23:38.580 23:38.579 --> 23:41.039 If you could do that--I'm going to come back--what I'm going to 23:41.038 --> 23:43.338 do is come back and see how often you are right so we'll go 23:43.338 --> 23:44.248 back through these. 23:44.250 --> 23:46.820 Have you all written down a weight for the Statute of 23:46.824 --> 23:47.324 Liberty? 23:47.320 --> 23:51.120 23:51.119 --> 23:55.969 Population of the country, Turkey. 23:55.970 --> 24:03.070 24:03.069 --> 24:08.659 Since I don't have the current population, I want it in the 24:08.662 --> 24:11.372 year 2000. I didn't get the latest 24:11.367 --> 24:15.127 estimate; so, how many people were there 24:15.131 --> 24:17.091 in Turkey in 2000? 24:17.089 --> 24:21.299 And again, put down a range, a low and a high, 24:21.297 --> 24:22.977 that's 90% sure. 24:22.980 --> 24:27.440 24:27.440 --> 24:29.670 Third, the Sahara Desert. 24:29.670 --> 24:34.190 24:34.190 --> 24:42.640 How many square miles in the Sahara Desert? 24:42.640 --> 24:46.760 Remember that a square mile is 2.6 square kilometers; 24:46.759 --> 24:49.439 just in case you think in terms of kilometers, 24:49.436 --> 24:52.996 you can devise your answer in kilometers and then multiply by 24:53.004 --> 24:57.544 2.6. Again, write down a range. 24:57.540 --> 25:02.200 Enrollment at Yale. 25:02.200 --> 25:07.330 25:07.329 --> 25:09.469 By the way, I should have asked, you have to be honest 25:09.466 --> 25:12.026 with it. You could game me by writing 25:12.034 --> 25:16.524 really wide intervals for nine of ten questions and then an 25:16.516 --> 25:19.836 extremely narrow interval for the tenth. 25:19.839 --> 25:23.549 I'm expecting some sincere cooperation here--then you would 25:23.554 --> 25:26.954 guarantee that you were right exactly 90% of the time, 25:26.949 --> 25:29.239 right? I mean, you could say the 25:29.241 --> 25:32.611 Statute of Liberty weighs between zero and a hundred 25:32.610 --> 25:35.650 quintillion tons and you know you're right. 25:35.650 --> 25:38.770 Then you could deliberately say the population of Turkey is 25:38.771 --> 25:42.001 between one and two people and then you know you're wrong. 25:42.000 --> 25:43.690 You could do--you're not supposed to do that. 25:43.690 --> 25:50.130 I want the enrollment in Yale; I don't have the latest 25:50.127 --> 25:53.477 number--2005. That's the total number of 25:53.478 --> 25:58.218 students at Yale University in 2005, including Yale College and 25:58.218 --> 26:00.358 all the graduate schools. 26:00.359 --> 26:09.409 The fifth question is about the Pulitzer Prize. 26:09.410 --> 26:14.080 26:14.080 --> 26:15.250 Do you know this prize? 26:15.250 --> 26:20.100 It's a prize that journalists win for writing great articles 26:20.104 --> 26:21.964 or books. I want to know, 26:21.958 --> 26:25.688 what is--how much do you get in cash if you win the Pulitzer 26:25.690 --> 26:31.060 Prize? I have it for last year, 2007; 26:31.059 --> 26:35.249 it might be different in 2008, so I'm asking for the 2007, 26:35.253 --> 26:36.213 in dollars. 26:36.210 --> 26:42.510 26:42.509 --> 26:45.609 I hope you were honest in putting confidence intervals. 26:45.610 --> 26:47.130 Have you gotten them? 26:47.130 --> 26:51.210 Now, what I'm going to do, if you've answered all five 26:51.210 --> 26:55.520 questions, I'm going to tell you the answers--the correct 26:55.521 --> 27:00.291 answer--and then ask for a show of hands of how many--please be 27:00.294 --> 27:03.224 honest and don't be embarrassed. 27:03.220 --> 27:09.770 Raise your hand if you were right, meaning that if my answer 27:09.769 --> 27:15.319 falls within your 90% confidence interval, okay? 27:15.319 --> 27:17.879 Let's go to the Statue of Liberty. 27:17.880 --> 27:24.450 The Statue of Liberty weighs 252 tons. 27:24.450 --> 27:29.430 So, can I have a show of hands--how many people here have 27:29.432 --> 27:31.392 252 in the interval? 27:31.390 --> 27:35.760 You're doing fairly well; what fraction–keep your 27:35.762 --> 27:40.552 hands up--looks like it's about, what would you say, 27:40.551 --> 27:44.601 20-25%? Thank you for being honest and 27:44.603 --> 27:49.013 not gaming me. It should have been 90% who 27:49.005 --> 27:53.815 were right. What is the 2000 population of 27:53.822 --> 28:00.382 Turkey? I'll give you the exact number 28:00.382 --> 28:08.802 that I got from their statistic: 65,666,677. 28:08.799 --> 28:12.189 That's a little over sixty-five million. 28:12.190 --> 28:16.060 How many people have that in their interval? 28:16.060 --> 28:21.400 That's better; that's like 40%--40% or 50%. 28:21.400 --> 28:25.140 You're doing better but it's still not 90%. 28:25.140 --> 28:27.820 How many square miles in the Sahara Desert? 28:27.820 --> 28:32.160 3.5 million. Can I get a show of hands--how 28:32.159 --> 28:34.449 many were right on that? 28:34.450 --> 28:38.970 Well, this one really got you, that was like 5%. 28:38.970 --> 28:42.710 Was anyone right on all of them so far? 28:42.710 --> 28:48.550 Nobody. Enrollment in Yale, Fall 2005? 28:48.550 --> 28:52.700 11,483 students. How many were right? 28:52.700 --> 28:58.120 Okay, that's about 40%--right close to 40%, 28:58.119 --> 29:01.359 I'd say. Finally, how much do you win 29:01.363 --> 29:05.483 if--how much do you receive if you win the Pulitzer Prize? 29:05.480 --> 29:08.580 $10,000. Can I have a show of hands? 29:08.579 --> 29:12.769 That was really low, that's like 5%. 29:12.769 --> 29:14.799 I knew that was a trick because you've heard about the Nobel 29:14.804 --> 29:16.844 Prize. Those are both prestigious, 29:16.835 --> 29:19.035 right? Nobel Prize gives you something 29:19.038 --> 29:22.308 on the order of $1,000,000 and the Pulitzer Prize gives you 29:22.311 --> 29:24.851 something like--it only gives you $10,000. 29:24.850 --> 29:25.960 So how can that be? 29:25.960 --> 29:29.840 I sort of picked something that I thought you might be wrong on. 29:29.839 --> 29:31.939 That reveals something about human behavior. 29:31.940 --> 29:35.030 It's a choice in life. 29:35.029 --> 29:36.679 You go into different walks of life. 29:36.680 --> 29:38.920 This is something that's fundamental to economics: 29:38.924 --> 29:41.264 there are just different expectations about how much 29:41.260 --> 29:42.680 money you're going to make. 29:42.680 --> 29:46.230 If you go into the news media--and I think that's a 29:46.232 --> 29:49.932 wonderful career--you're not going to make much money 29:49.926 --> 29:53.026 probably. The whole thing is just scaled 29:53.030 --> 29:56.960 down and I think there's something revealing about this 29:56.955 --> 30:01.235 that we just have social norms for how much someone is to be 30:01.243 --> 30:04.003 paid. If you were to give Stephen 30:03.998 --> 30:08.428 Schwartzman a $10,000 prize, it would be more like an insult 30:08.432 --> 30:11.792 than anything. But if you are working for The 30:11.792 --> 30:16.102 New Haven Register and you get this prize, it's a life-changing 30:16.095 --> 30:18.135 event, not because of the $10,000 30:18.140 --> 30:20.490 maybe--even they get more money than that. 30:20.490 --> 30:25.930 Anyway, the point was that people tend to be overconfident. 30:25.930 --> 30:28.540 Incidentally, it's not just males; 30:28.539 --> 30:31.839 females are well-known to be overconfident too. 30:31.839 --> 30:35.689 There is a thing about macho males--"know it all"--but 30:35.690 --> 30:39.540 experiments prove that women have the same problem. 30:39.539 --> 30:43.369 That's why I think that when we look at charts of the stock 30:43.367 --> 30:46.597 market, we see things that we think we understand, 30:46.602 --> 30:48.452 especially young people. 30:48.450 --> 30:52.380 They get deluded into thinking they understand more than they 30:52.375 --> 30:56.455 really do. I wanted to talk about some 30:56.458 --> 31:02.688 authors that I admire who have written about this. 31:02.690 --> 31:10.320 These are books that I don't have on the reading list, 31:10.324 --> 31:14.074 but they're fun to read. 31:14.069 --> 31:18.579 There's a professor at The Harvard Business School, 31:18.577 --> 31:24.027 Rakesh Khurana; he has a book on the search for 31:24.032 --> 31:26.312 charismatic CEOs. 31:26.310 --> 31:32.770 31:32.769 --> 31:36.659 It's not just overconfidence in yourself; 31:36.660 --> 31:38.740 we tend also to put overconfidence in leaders. 31:38.740 --> 31:45.160 We have a sense that some people are just natural geniuses 31:45.157 --> 31:50.217 and know everything, so we think that they can 31:50.223 --> 31:54.843 transform our lives or our companies. 31:54.839 --> 31:58.819 So, boards of directors are constantly looking for a CEO who 31:58.816 --> 32:02.856 is a genius and they keep getting fooled and disappointed. 32:02.859 --> 32:06.359 They bring someone in and this person often messes things up 32:06.358 --> 32:09.798 more than helps because this person realizes that he or she 32:09.797 --> 32:12.107 has to live up to this genius role, 32:12.110 --> 32:13.460 so they better do something. 32:13.460 --> 32:15.430 So, they do something in a flailing way, 32:15.433 --> 32:18.323 not understanding what they're doing, and they mess up the 32:18.318 --> 32:20.528 whole company. Really, a lot of what 32:20.532 --> 32:24.502 happens--good things that happen in human society--are the result 32:24.503 --> 32:27.923 of lots of people doing their own special things and all 32:27.915 --> 32:29.275 working together. 32:29.279 --> 32:34.839 There's no great genius but there's this idea in our mind 32:34.838 --> 32:38.708 that we are going to be such a thing. 32:38.710 --> 32:44.680 Related to that--and I wanted to mention, it's on the reading 32:44.678 --> 32:49.848 list--an article by one of my students in this class, 32:49.851 --> 32:51.941 who's now at MIT. 32:51.940 --> 32:59.030 He took this class about ten years ago, Fadi Kanaan and 32:59.033 --> 33:04.423 co-authored with another MIT professor, 33:04.420 --> 33:10.810 Dirk Jenter, again looking at overconfidence 33:10.807 --> 33:13.627 in our judgments. 33:13.630 --> 33:18.220 Again, they looked at CEOs, chief executive officers of 33:18.218 --> 33:23.148 companies, and they found that companies in industries that 33:23.146 --> 33:25.946 fail tend to fire their CEOs. 33:25.950 --> 33:29.370 This is unjust; this is an overreaction. 33:29.369 --> 33:33.029 You bring in this CEO who's supposed to be brilliant and 33:33.025 --> 33:36.675 then the business fails, so you fire the guy right after 33:36.681 --> 33:39.561 that. We're kind of manic-depressive 33:39.555 --> 33:40.915 about these guys. 33:40.920 --> 33:43.720 When the business fails, we think we were such a 33:43.723 --> 33:45.753 mistake. This guy had such promise and 33:45.754 --> 33:47.984 he just didn't live up so we get rid of him; 33:47.980 --> 33:50.990 but in fact, they found that the CEO gets 33:50.991 --> 33:54.381 fired even if the whole industry went down. 33:54.380 --> 33:58.480 So, you can't blame the CEO for the fact--if you're one company 33:58.483 --> 34:02.063 in an industry and the whole industry goes down--or the 34:02.057 --> 34:05.757 remaining industry even not including that firm--it's not 34:05.763 --> 34:09.083 the CEO's fault. We tend to be kind of wild and 34:09.078 --> 34:10.688 extreme in our judgments. 34:10.690 --> 34:13.990 You've seen that a lot--a lot of CEOs lost their jobs recently 34:13.988 --> 34:15.338 in the subprime crisis. 34:15.340 --> 34:17.180 Was it their fault? 34:17.179 --> 34:20.629 Probably not, but they get fired anyway. 34:20.630 --> 34:24.440 We go through this manic-depressive--we try to hire 34:24.438 --> 34:28.088 charismatic CEOs, then we get disappointed and we 34:28.094 --> 34:32.364 keep going through musical chairs one after another. 34:32.360 --> 34:37.770 Nassim Taleb, who lives here in Connecticut 34:37.767 --> 34:44.197 and I know him well, has a book called Fooled by 34:44.203 --> 34:49.923 Randomness, which was a best-seller and 34:49.922 --> 34:52.712 it's very fun to read. 34:52.710 --> 34:58.000 It's a story--he's a Wall Street--he had an investment 34:58.004 --> 35:03.004 management firm and he observed a lot of people. 35:03.000 --> 35:06.860 It's a book about how people over interpret--they tend to 35:06.859 --> 35:10.989 blame themselves for failures and congratulate themselves for 35:10.993 --> 35:14.783 successes too much and they don't realize that it's just 35:14.784 --> 35:17.084 random. Some guy who's in a 35:17.080 --> 35:20.180 business--business is succeeding--why is it 35:20.177 --> 35:22.977 succeeding? Because the guy came in, 35:22.978 --> 35:27.468 dumb luck at the right time and everything is supporting that, 35:27.472 --> 35:29.832 concludes that he's a genius. 35:29.829 --> 35:38.029 Then Taleb observes them later, after things don't go so well, 35:38.026 --> 35:42.456 and suddenly they're depressed. 35:42.460 --> 35:45.280 – I talked to stockbrokers before and after 35:45.276 --> 35:48.736 the '87 stock market crash and one of them told me--or maybe 35:48.738 --> 35:52.488 more than one of them told me--I can tell that the crash occurred 35:52.493 --> 35:55.843 from the tone of voice of the people when they call up the 35:55.838 --> 35:58.228 phone. When the market was soaring 35:58.230 --> 36:01.430 just before the '87 peak, he said they would call up and 36:01.428 --> 36:04.508 they were brash and rude to me and they would say, 36:04.510 --> 36:08.200 let's trade this, get this done--kind of just 36:08.197 --> 36:11.547 disparaging subtlety, the stockbroker. 36:11.550 --> 36:13.890 Then after the crash, when these people were sort 36:13.886 --> 36:16.606 of--many of them--wiped out, they'd answer the phone in a 36:16.612 --> 36:18.532 sheepish way. You could just tell in the tone 36:18.533 --> 36:19.913 of their voice that they were crushed. 36:19.910 --> 36:22.880 So, that's what happens. 36:22.880 --> 36:29.140 I also have down on this part of the reading list Irving 36:29.142 --> 36:33.472 Fisher, who was a professor at Yale, 36:33.469 --> 36:39.439 who was a very prominent economist in the first half of 36:39.441 --> 36:43.271 the century. He's another Yale graduate, 36:43.270 --> 36:45.720 Yale Class of 1895, I think. 36:45.719 --> 36:48.889 I'm sure he lectured on this stage because this building 36:48.894 --> 36:51.784 was--his office was in this building, I believe. 36:51.780 --> 36:55.230 He died around the mid-1940s but he's famous for 36:55.231 --> 36:58.861 overconfidence. In 1929, he was interviewed 36:58.859 --> 37:03.349 just before--two weeks before--the 1929 peak and--do 37:03.353 --> 37:06.353 you know what I'm referring to? 37:06.349 --> 37:11.259 He said he thought the stock market was on a permanently high 37:11.264 --> 37:16.104 plateau and he wrote a book in 1929--actually it came out in 37:16.097 --> 37:20.277 1930--with this extremely optimistic outlook for the 37:20.275 --> 37:24.925 market. He had a beautiful mansion; 37:24.929 --> 37:27.859 he was a wealthy man for a while, but he lost everything in 37:27.857 --> 37:29.117 the stock market crash. 37:29.119 --> 37:33.189 In fact he had to--he had borrowed against his home and he 37:33.185 --> 37:36.315 lost his house, so Yale University bought his 37:36.323 --> 37:39.393 house for him and rented it out to him; 37:39.389 --> 37:41.209 otherwise, he would be on the street. 37:41.210 --> 37:46.640 I have an article written by him in 1930--I think it's 1930 37:46.638 --> 37:50.568 or at the end of 1929--discussing the stock 37:50.568 --> 37:54.928 market crash. He still is unrepentant. 37:54.929 --> 37:57.659 This was our most brilliant professor here at Yale, 37:57.661 --> 38:00.011 but he just totally misjudged the market, 38:00.010 --> 38:02.730 He's just totally unrepentant--he just went back 38:02.733 --> 38:05.113 over his book. There are so many good 38:05.111 --> 38:08.541 reasons--the 20s were a spectacular era--so many good 38:08.543 --> 38:12.113 reasons the stock market will keep going up and he just 38:12.106 --> 38:13.686 wouldn't back down. 38:13.690 --> 38:16.610 In fact, what he actually did was he started borrowing from 38:16.608 --> 38:19.528 his relatives--he had wealthy relatives--and he lost all of 38:19.527 --> 38:21.667 it. He just couldn't have imagined 38:21.674 --> 38:25.264 that the stock market would go down--there was just no reason 38:25.262 --> 38:28.432 that he could think of--and that's what he says in the 38:28.432 --> 38:32.692 article. Anyway, I want to talk more 38:32.693 --> 38:37.213 precisely about how people behave; 38:37.210 --> 38:40.960 this is all general about overconfidence, 38:40.959 --> 38:46.489 but there's some other factors that I want to start with. 38:46.489 --> 38:53.699 The most important theory in behavioral finance is the 38:53.702 --> 38:59.012 Kahneman and Tversky Prospect Theory. 38:59.010 --> 39:02.050 39:02.050 --> 39:06.080 Danny Kahneman, who is now a professor of 39:06.078 --> 39:10.508 psychology at Princeton, and Amos Tversky, 39:10.510 --> 39:16.820 who died a few years ago--they wrote, I think, 39:16.824 --> 39:23.844 the most famous article on behavioral economics; 39:23.840 --> 39:25.680 it goes beyond just finance. 39:25.679 --> 39:36.589 The title of the article was Prospect Theory and that 39:36.594 --> 39:39.594 was 1979. This is, I think, 39:39.594 --> 39:42.374 the–actually, I think there was a ranking of 39:42.373 --> 39:45.323 economics articles--scholarly articles--by numbers of 39:45.322 --> 39:48.782 quotations and this was number two out of all articles written 39:48.781 --> 39:50.541 in the last fifty years. 39:50.539 --> 39:54.269 Number one, it was quoted for some other reason--I'm not 39:54.270 --> 39:58.340 sure--it was some statistical method that everyone quoted. 39:58.340 --> 40:00.700 In terms of an intellectual contribution, 40:00.696 --> 40:03.936 this is the most important economics article in the last 40:03.937 --> 40:06.507 fifty years, at least judged by how many 40:06.505 --> 40:07.585 times it's cited. 40:07.590 --> 40:10.950 Kahneman and Tversky are not really talking about 40:10.946 --> 40:14.646 overconfidence but something, well, perhaps related to 40:14.652 --> 40:16.822 it--something more general. 40:16.820 --> 40:21.070 It's how people make choices and there are two elements to 40:21.074 --> 40:28.644 this theory. It replaces expected utility 40:28.643 --> 40:39.833 and it has--what it does, it replaces the utility 40:39.826 --> 40:49.376 function with a value function--with value 40:49.378 --> 41:01.258 function--and replaces the probabilities with, 41:01.260 --> 41:03.840 what they call, weights. 41:03.840 --> 41:08.760 I'm going to explain what that is and we'll move on. 41:08.760 --> 41:14.230 Let me give a little story that leads up to it and it's a story 41:14.233 --> 41:18.213 that Paul Samuelson, who's a professor at MIT, 41:18.205 --> 41:21.595 told. Paul Samuelson was a highly 41:21.604 --> 41:27.304 esteemed--he is--I think he's ninety-two or about ninety-two 41:27.296 --> 41:32.406 years old now and still writing and still working. 41:32.409 --> 41:36.559 He was a--he is a mathematical economist, retired now, 41:36.564 --> 41:40.254 but he told a story that illustrates some of the 41:40.248 --> 41:44.418 beginnings of Prospect--in fact, he kind of anticipated Prospect 41:44.421 --> 41:47.581 Theory. This goes back to an article 41:47.580 --> 41:49.890 that he wrote in 1963. 41:49.889 --> 41:56.309 In 1963, he was having lunch with one of his colleagues, 41:56.313 --> 42:00.103 another economist; he doesn't name this other 42:00.098 --> 42:02.268 person because it would be embarrassing, 42:02.270 --> 42:03.830 but everyone knows it was E. 42:03.830 --> 42:05.780 Cary Brown, a professor at MIT. 42:05.780 --> 42:10.220 Samuelson, in a playful mood--he was always sort of a 42:10.224 --> 42:13.734 playful person--he said at lunch--he said, 42:13.727 --> 42:15.947 hey let's toss a coin. 42:15.949 --> 42:20.179 Let's make a bet just for the fun of it and if it comes up 42:20.183 --> 42:23.973 heads, I'll give you $200, but if it comes up tails, 42:23.970 --> 42:25.530 you give me $100. 42:25.530 --> 42:27.620 He said, let's do it I'm ready. 42:27.619 --> 42:29.899 This kind of took E. 42:29.901 --> 42:32.641 Cary Brown by surprise. 42:32.639 --> 42:35.429 That sounds like a lot of money, especially in 1963; 42:35.429 --> 42:40.219 prices were much lower--that's like $1,000 or $2000. 42:40.220 --> 42:41.640 It was big money. 42:41.639 --> 42:43.419 But of course, these professors could afford 42:43.422 --> 42:46.182 it; it's not that much money. 42:46.179 --> 42:48.869 So, let's just say it's $100 and $200. 42:48.869 --> 42:51.889 Do you feel like--if I were to offer that to you right 42:51.887 --> 42:55.187 now--let's do it because you don't have cash on you now, 42:55.190 --> 42:58.880 but you'd have to promise to pay me if you came out wrong. 42:58.880 --> 43:00.370 Do you feel like doing that? 43:00.369 --> 43:04.049 No, someone is answering me honestly. 43:04.050 --> 43:07.770 Introspect and think about it while this is suddenly thrust on 43:07.773 --> 43:08.303 you. E. 43:08.300 --> 43:11.660 Cary Brown said, come on I don't want to do 43:11.659 --> 43:15.819 this--Samuelson was being annoying by doing this. 43:15.820 --> 43:18.920 Then Samuelson thought--had another idea-he said, 43:18.920 --> 43:22.670 what if I offered--he didn't actually offer this--what if I 43:22.666 --> 43:25.246 offered to--let's do this 100 times. 43:25.250 --> 43:29.680 We'll toss a coin 100 times and each time it comes up heads I 43:29.680 --> 43:34.480 give you $200 and each time it comes up tails you give me $100. 43:34.480 --> 43:36.690 Well, E. Cary Brown, knowing mathematics 43:36.690 --> 43:39.070 of statistics and the law of probabilities, 43:39.070 --> 43:41.910 he said, well, if we do it 100 times, 43:41.914 --> 43:45.164 by the binomial theorem, I'm sure to win. 43:45.159 --> 43:48.829 I couldn't possibly--this is elementary--100 times is a lot 43:48.833 --> 43:51.183 of times. In fact, I'll make thousands of 43:51.180 --> 43:52.060 dollars. So, E. 43:52.057 --> 43:53.977 Cary Brown said, I'll do it. 43:53.980 --> 43:56.890 I would do it, but he didn't actually do it. 43:56.889 --> 44:01.589 Samuelson then said--he went back to his office and he wrote 44:01.588 --> 44:05.328 a paper--that's this 1963 paper--proving that E. 44:05.331 --> 44:07.721 Cary Brown was irrational. 44:07.719 --> 44:10.899 You cannot possibly say, I will take 100 of them but I 44:10.900 --> 44:12.400 won't take one of them. 44:12.400 --> 44:14.030 That's not rational. 44:14.030 --> 44:18.290 That was one of the motivating things in Kahneman and Tversky. 44:18.289 --> 44:22.489 What Kahneman and Tversky said is that people behave--if you 44:22.490 --> 44:26.620 can introspect and imagine why some of you didn't feel like 44:26.620 --> 44:30.960 taking this bet--people behave as if they have a kink in their 44:30.963 --> 44:34.563 utility. This may sound an abstract way 44:34.558 --> 44:37.498 of putting it, but expected utility 44:37.503 --> 44:42.273 theory--the traditional theory says that everybody has a 44:42.267 --> 44:46.767 utility function that they consistently refer to when 44:46.771 --> 44:49.111 making calculations. 44:49.110 --> 44:53.770 I'm going to put Kahneman and Tversky over here and I'm going 44:53.770 --> 44:57.110 to put Expected Utility Theory over here. 44:57.110 --> 45:03.210 45:03.210 --> 45:07.870 Expected Utility Theory says that I want wealth--and I'll 45:07.870 --> 45:12.700 call w wealth--and I get utility from wealth--that's 45:12.698 --> 45:16.058 U. My utility curve--it has maybe 45:16.062 --> 45:19.792 any number of shapes, but its concave downward and 45:19.789 --> 45:23.819 smooth, so you have what's called diminishing marginal 45:23.821 --> 45:24.811 utility; 45:24.810 --> 45:33.820 45:33.820 --> 45:35.750 that's Expected Utility Theory. 45:35.750 --> 45:40.000 What Expected Utility Theory means--the slope is always 45:39.997 --> 45:42.427 decreasing. Every extra dollar of wealth 45:42.434 --> 45:45.314 gives me less happiness but it always give me a little bit 45:45.313 --> 45:46.883 more, so I always want more. 45:46.880 --> 45:50.670 Expected Utility Theory would say that that's a two-for-one 45:50.669 --> 45:54.519 bet that Samuelson is offering and it's small compared to my 45:54.523 --> 45:58.503 lifetime wealth. My utility is essentially 45:58.495 --> 46:04.355 linear over the relevant range, plus $200 or minus $100, 46:04.359 --> 46:09.369 so I don't really concern myself about risk. 46:09.369 --> 46:12.109 I should just take every bet like that all the time. 46:12.110 --> 46:14.820 You should always be looking--if you are behaving 46:14.822 --> 46:17.142 this way--you should always be looking. 46:17.139 --> 46:19.479 Anyone who wants to make a bet with me anytime, 46:19.476 --> 46:22.566 I'll always take it if it's in my advantage--even a little bit 46:22.574 --> 46:25.794 in my advantage. People seem to like to gamble 46:25.793 --> 46:29.063 but they don't like to do it consistently. 46:29.059 --> 46:32.659 They like to go to--they end up going to gambling casinos where 46:32.655 --> 46:35.385 it's stacked against them, not for them, 46:35.387 --> 46:39.387 but it's somehow arranged as an entertainment. 46:39.389 --> 46:45.239 Well, Kahneman and Tversky said that people don't behave this 46:45.236 --> 46:51.176 way and it's as if they have a value function as a function of 46:51.180 --> 46:55.150 their money. Let's put in the middle of the 46:55.148 --> 46:58.118 value function, the reference point. 46:58.119 --> 47:04.059 Reference point means where you are today and your value--that's 47:04.062 --> 47:08.532 V, which is like utility, but now we're talking in 47:08.532 --> 47:11.322 psychological terms, so we give it a different name. 47:11.320 --> 47:17.670 The value function has a kink; it's something like that at the 47:17.674 --> 47:21.484 reference point. I'm trying to draw it--it's not 47:21.478 --> 47:26.138 necessarily--it looks here like two straight lines and that's 47:26.139 --> 47:28.469 not quite the way to do it. 47:28.469 --> 47:32.409 Let me try and do this again--it's curved downward a 47:32.410 --> 47:36.740 little bit, but it becomes much less--I don't--I'm having 47:36.737 --> 47:40.057 trouble drawing this on the board well. 47:40.059 --> 47:42.459 I don't want to ever--kind of going down. 47:42.460 --> 47:47.550 There's a kink here, where the slope--I think I've 47:47.550 --> 47:49.940 got it sort of there. 47:49.940 --> 47:53.430 It's concave down everywhere, just like the utility function 47:53.433 --> 47:56.633 is, but there's a discontinuity of slope right here. 47:56.630 --> 48:00.660 Where is that? That's where I am now. 48:00.659 --> 48:06.099 What it means is that I value losses much more than I value 48:06.098 --> 48:08.628 gains from wherever I am. 48:08.630 --> 48:12.170 There's a big difference between losing and winning, 48:12.170 --> 48:16.200 so when I reflect on this bet I'm thinking of--I could lose 48:16.196 --> 48:18.136 $100 and that scares me. 48:18.139 --> 48:21.329 It feels bad--the idea that--I would just feel bad. 48:21.329 --> 48:26.499 So gaining $200 is positive for me but it doesn't offset the 48:26.499 --> 48:28.689 loss that I might make. 48:28.690 --> 48:31.140 So, if I have equal probabilities, 48:31.142 --> 48:35.382 what you want to do is weight the gains and losses and the 48:35.378 --> 48:38.998 losses tend to dominate, so you don't want to take the 48:38.995 --> 48:41.025 bet. The weighting function 48:41.027 --> 48:45.047 incorporates Samuelson's lunch colleague's problem: 48:45.045 --> 48:49.215 that people don't want to take bets that are to their 48:49.224 --> 48:52.324 advantage. It goes back to a kink in the 48:52.324 --> 48:53.504 utility function. 48:53.500 --> 48:56.010 Now, incidentally, this is fundamentally different 48:56.013 --> 48:58.373 from--in economic theory, economists would say, 48:58.372 --> 49:01.042 well you can put a kink in the utility function. 49:01.039 --> 49:03.779 There could be some wealth level that's special to you. 49:03.780 --> 49:07.190 But a theory economist--that kink has to stay at a certain 49:07.192 --> 49:10.032 wealth level. With Kahneman and Tversky, 49:10.027 --> 49:13.677 this kink moves around with you, so whatever--it's 49:13.677 --> 49:18.517 whatever--you're always at the kink because it's not rational; 49:18.519 --> 49:20.789 this is not rational Expected Utility Theory. 49:20.789 --> 49:24.839 This is--I'm always looking at where I am now and exaggerating 49:24.844 --> 49:28.304 in my mind the importance of deviations from that. 49:28.300 --> 49:31.200 People are very concerned with small losses; 49:31.199 --> 49:36.709 that's what th kink in the value function is. 49:36.710 --> 49:40.630 Now, I want to talk about another Kahneman and Tversky 49:40.627 --> 49:43.507 thing, called the weighting function. 49:43.510 --> 49:51.230 The weighting function refers to the fact that people distort 49:51.230 --> 49:55.090 probabilities in their mind. 49:55.090 --> 50:06.700 50:06.699 --> 50:12.869 It's not that they don't know probabilities but they distort 50:12.866 --> 50:15.476 them in their thinking. 50:15.480 --> 50:19.510 I'll give an example that illustrates the Kahneman-Tversky 50:19.514 --> 50:23.764 weighting function and it goes back years before Kahneman and 50:23.760 --> 50:27.900 Tversky. It's a famous example from a 50:27.898 --> 50:31.658 French economist, Maurice Allais, 50:31.656 --> 50:36.116 and it's called the Allais Paradox. 50:36.119 --> 50:43.509 It illustrates thinking that violates Expected Utility 50:43.505 --> 50:46.945 Theory. I'm going to give you a choice 50:46.948 --> 50:50.888 between two "prospects," as Kahneman and Tversky called 50:50.886 --> 51:00.346 them. Suppose I offered you a 25% 51:00.354 --> 51:12.864 chance to win $3,000 or, alternatively, 51:12.858 --> 51:22.728 a 20% chance to win $4,000. 51:22.730 --> 51:29.640 51:29.640 --> 51:33.040 Maybe I can get a show of hands. 51:33.039 --> 51:35.609 This is like Samuelson's lunch colleague again, 51:35.605 --> 51:36.995 but a little different. 51:37.000 --> 51:39.460 Suppose I'm offering--I'm not offering this, 51:39.459 --> 51:42.949 but suppose I offered this--you have a choice between Prospect 51:42.947 --> 51:44.317 One or Prospect Two. 51:44.320 --> 51:48.780 Prospect One--I'm going to toss a four-sided coin and if it 51:48.779 --> 51:53.309 comes up with a probability of one-fourth, in a certain way, 51:53.314 --> 51:55.164 you will win $3,000. 51:55.159 --> 52:00.189 In Prospect Two, I'm going to give you a chance 52:00.193 --> 52:02.713 of 20% to win $4,000. 52:02.710 --> 52:06.120 Can you tell me which of these you'd pick if you had to pick 52:06.115 --> 52:07.265 only one of these? 52:07.270 --> 52:09.310 Do you understand the question? 52:09.310 --> 52:13.390 How many would pick number One? 52:13.390 --> 52:17.580 It seems like it's about 20%. 52:17.580 --> 52:20.140 How many would pick number Two? 52:20.139 --> 52:21.969 So, most of you would pick number Two. 52:21.969 --> 52:26.829 Now, let's do a variation on this question here--a very 52:26.829 --> 52:28.539 simple variation. 52:28.540 --> 52:31.560 Which would you prefer? 52:31.559 --> 52:36.309 This is the one that you picked--most people picked. 52:36.309 --> 52:48.679 Another prospect--100% chance of winning $3,000--or Two, 52:48.682 --> 52:59.482 that would be an 80% chance of winning $4,000. 52:59.480 --> 53:02.920 53:02.920 --> 53:06.230 Do you see the--if you pick Prospect One, 53:06.230 --> 53:09.790 you're going to just get $3,000 for sure. 53:09.789 --> 53:13.319 If you pick Prospect Two, you'll probably get $4,000 but 53:13.318 --> 53:14.728 an 80% chance of it. 53:14.730 --> 53:16.660 How many would pick One? 53:16.659 --> 53:20.039 That looks like the--how many would pick Two? 53:20.040 --> 53:22.240 Very few of you would pick Two. 53:22.239 --> 53:26.569 Have to reflect--so we picked One this time. 53:26.570 --> 53:28.360 Now, you might want to reflect on that. 53:28.360 --> 53:31.560 Why was it such a different--why did you pick One 53:31.563 --> 53:34.303 in this case and pick Two in this case? 53:34.300 --> 53:37.960 The thing I want to point out is that the number--the cash 53:37.959 --> 53:41.879 amounts are the same in the two examples but the probabilities 53:41.875 --> 53:43.925 are just multiplied by four. 53:43.929 --> 53:50.289 So the expected utility of the two is just four times as great, 53:50.294 --> 53:55.704 no matter what. They're the same--the utilities 53:55.696 --> 54:00.056 are the same, with the same numbers. 54:00.059 --> 54:04.349 All I've done is multiply your expected utility by four in this 54:04.353 --> 54:07.473 case, so you can't make a different choice. 54:07.469 --> 54:11.479 If you picked Two over here when comparing these two 54:11.477 --> 54:16.027 prospects you should also have picked Two when you compared 54:16.034 --> 54:17.924 these two prospects. 54:17.920 --> 54:21.620 Why didn't you? Most of you switched. 54:21.620 --> 54:24.090 Can you tell me why? 54:24.090 --> 54:28.600 Yes. Student: I would prefer 54:28.604 --> 54:31.304 not to gamble, so if I had the chance a--in 54:31.298 --> 54:34.568 the first situation, I would take the chance to make 54:34.570 --> 54:36.130 $4,000. Professor Robert 54:36.132 --> 54:38.012 Shiller: You would choose not to gamble. 54:38.010 --> 54:40.690 Does this mean it's like a moral judgment or-- 54:40.689 --> 54:43.129 Student: No, I prefer certainty. 54:43.130 --> 54:44.140 Professor Robert Shiller: Okay, 54:44.137 --> 54:44.707 you got it exactly. 54:44.710 --> 54:46.750 That's yeah--you got--you prefer certainty. 54:46.750 --> 54:54.400 There's some anxiety about maybe--you got it exactly right. 54:54.400 --> 55:00.820 I think people like certainty and ambiguity is difficult for 55:00.822 --> 55:03.002 them to adjust to. 55:03.000 --> 55:07.010 Kahneman and Tversky put it in this following way: 55:07.009 --> 55:10.199 it's a little bit like we're cavemen. 55:10.199 --> 55:13.309 It turns out, we were all taught to count and 55:13.309 --> 55:17.619 to do arithmetic but primitive people actually have difficulty 55:17.621 --> 55:20.431 counting. There's an old story that 55:20.426 --> 55:24.346 cavemen had only three numbers: one, two, and many. 55:24.349 --> 55:27.589 I used to disbelieve this story but I'm not--actually it was a 55:27.591 --> 55:29.771 psychologist at Princeton told me that, 55:29.769 --> 55:32.999 as a matter of fact, it's proven that there are some 55:33.004 --> 55:35.924 people whose languages have only those numbers: 55:35.922 --> 55:38.282 one, two, and many. 55:38.280 --> 55:41.180 For example, they're called--in Laos in 55:41.183 --> 55:45.313 Thailand, there's a very primitive group of people with 55:45.309 --> 55:47.219 primitive technology. 55:47.219 --> 55:50.409 I don't mean that they're primitive people but they only 55:50.411 --> 55:51.921 have one, two, and many; 55:51.920 --> 55:53.690 and there are others that have been discovered. 55:53.690 --> 55:56.720 Emotionally we're like that. 55:56.719 --> 55:59.639 I used to wonder how could they have only those numbers: 55:59.643 --> 56:00.763 one, two, and many. 56:00.760 --> 56:03.780 You ask a mother, how many children do you have? 56:03.780 --> 56:07.120 She couldn't answer; she didn't have the word three, 56:07.116 --> 56:09.496 but as a matter of fact they didn't. 56:09.500 --> 56:13.130 So I guess, if you asked the mother, how many children you 56:13.133 --> 56:15.813 have, she would probably just name them. 56:15.809 --> 56:17.609 She couldn't say, I have three children. 56:17.610 --> 56:21.750 But anyway, we're all kind of like that when we think about 56:21.750 --> 56:23.980 probability; that's Kahneman and Tversky. 56:23.980 --> 56:28.100 Kahneman and Tversky say what we do is that in our minds we 56:28.098 --> 56:32.218 weight the probabilities in a distorted way and this is the 56:32.216 --> 56:33.846 weighting function. 56:33.849 --> 56:38.169 So, we have the weight--that's weight not wealth here--against 56:38.170 --> 56:42.350 the probability and I'm going to exaggerate a little bit. 56:42.349 --> 56:47.689 This is zero and this is one because probabilities range from 56:47.685 --> 56:51.055 zero to one. The weighting function looks 56:51.058 --> 56:55.158 like this--I'm exaggerating a little bit so you can see 56:55.163 --> 56:59.043 but--and then it jumps up or it jumps down here; 56:59.040 --> 57:01.260 this is the idea. 57:01.260 --> 57:05.650 What Kahneman and Tversky said in their original 1979 article 57:05.651 --> 57:10.121 is, we act as--there's a wide range of probabilities here that 57:10.116 --> 57:13.406 are all kind of blurred and put together. 57:13.409 --> 57:15.529 We minimize and–emotionally, 57:15.531 --> 57:18.941 the difference between probabilities--they're all kind 57:18.939 --> 57:20.289 of in the middle. 57:20.289 --> 57:23.219 So, when I said twenty or twenty-five, in your mind you 57:23.218 --> 57:26.418 said, here's twenty and here's twenty-five but I don't think 57:26.419 --> 57:28.859 they're much different to me emotionally. 57:28.860 --> 57:32.450 The money sounds different but the probability sounds the same. 57:32.449 --> 57:35.469 It's like I have only three probabilities: 57:35.470 --> 57:37.390 can't happen, might happen, 57:37.386 --> 57:39.666 and it's certain to happen. 57:39.670 --> 57:42.880 You tend to be totally in to the certainty story, 57:42.883 --> 57:45.163 so you give it much more weight. 57:45.159 --> 57:53.649 The way--what people do then, summing up--in expected utility 57:53.651 --> 58:00.871 theory, you maximize the probability-weighted sum of 58:00.868 --> 58:02.848 utilities. 58:02.850 --> 58:10.970 58:10.969 --> 58:15.259 You maximize the summation of the probability of the 58:15.257 --> 58:19.877 i^(th) outcome times utility in the i^(th) 58:19.881 --> 58:25.261 outcome. But in Prospect Theory, 58:25.262 --> 58:34.772 you maximize the sum of the weights times the value 58:34.767 --> 58:39.327 function--the values. 58:39.329 --> 58:45.989 This is the Kahneman and Tversky variation on Expected 58:45.992 --> 58:50.832 Utility Theory. There's something related to it 58:50.832 --> 58:55.892 that psychologists talk about, it's called Regret Theory. 58:55.889 --> 59:00.669 It's a little bit different but it's essentially the same 59:00.665 --> 59:04.325 as–well, it's consistent with Prospect 59:04.331 --> 59:07.401 Theory. That is, people experience pain 59:07.395 --> 59:11.575 of regret and they do a lot of things to try to avoid the pain 59:11.578 --> 59:13.368 of regret. For example, 59:13.367 --> 59:17.847 when the stock market goes up they try to sell it and lock in 59:17.853 --> 59:21.823 the gain because they are worried that if it goes down 59:21.815 --> 59:25.325 again they will regret not having sold it; 59:25.329 --> 59:27.509 that's not a rational calculation. 59:27.510 --> 59:31.390 If you come to something, you just have it and then it 59:31.394 --> 59:33.524 escapes you, you feel pain. 59:33.519 --> 59:37.209 I guess that's what happened at the Super Bowl last night when 59:37.206 --> 59:40.826 the New England Patriots had a winning streak and they messed 59:40.832 --> 59:44.342 up at the very end--that's exceptionally painful and that's 59:44.337 --> 59:46.027 part of Regret Theory. 59:46.030 --> 59:49.230 I don't know how pained any of you are but it must have been 59:49.225 --> 59:50.575 painful to them anyway. 59:50.579 --> 59:58.779 I just mentioned some other things that are related to 59:58.783 --> 1:00:06.083 Prospect Theory. There's something called 1:00:06.081 --> 1:00:16.801 "mental compartments" that people--Expected Utility Theory 1:00:16.796 --> 1:00:20.246 says, your utility depends on your 1:00:20.248 --> 1:00:23.378 whole lifetime wealth, so you should be always 1:00:23.380 --> 1:00:27.560 thinking that everything that happens today is just part of a 1:00:27.557 --> 1:00:31.007 bigger story; I'm always thinking about my 1:00:31.013 --> 1:00:33.623 lifetime. I had you do an exercise at the 1:00:33.617 --> 1:00:37.077 beginning where I asked you to estimate the present value of 1:00:37.075 --> 1:00:40.295 your lifetime income and it probably came out to several 1:00:40.299 --> 1:00:43.099 million dollars. So, if you were behaving 1:00:43.103 --> 1:00:47.173 rationally you would always be weighing things against that big 1:00:47.170 --> 1:00:49.400 sum of several million dollars. 1:00:49.400 --> 1:00:52.730 That's why plus $100, minus $200--who cares, 1:00:52.730 --> 1:00:55.190 right? That's the way you should be 1:00:55.185 --> 1:00:59.115 thinking but you don't think that way because you're human. 1:00:59.119 --> 1:01:01.809 People put things in mental compartments, 1:01:01.812 --> 1:01:04.642 all different compartments in your mind, 1:01:04.639 --> 1:01:07.709 and you have separate values for things depending on which 1:01:07.712 --> 1:01:09.062 compartment they're in. 1:01:09.059 --> 1:01:11.599 For example, when you go to the gambling 1:01:11.598 --> 1:01:15.438 casino, the winnings and losses are completely different. 1:01:15.440 --> 1:01:17.950 You just put them in a game compartment and you think, 1:01:17.951 --> 1:01:19.991 I can accept these and it doesn't matter. 1:01:19.989 --> 1:01:23.549 Investors are that way too, they'll sometimes put part of 1:01:23.545 --> 1:01:27.285 their portfolio in "I can play with this" mental compartment 1:01:27.292 --> 1:01:30.152 and others in another mental compartment. 1:01:30.150 --> 1:01:37.520 Anyway, I just want to come back--I have maybe a little bit 1:01:37.515 --> 1:01:42.335 more to say about this, but let me come back and talk 1:01:42.344 --> 1:01:45.544 just about the problem set we talked about last period. 1:01:45.539 --> 1:01:50.429 Problem Set #3--you've got your second problem set here--Problem 1:01:50.431 --> 1:01:54.471 Set #3 is a stock market forecasting exercise and the 1:01:54.469 --> 1:01:58.969 spreadsheet that I have up here is one spreadsheet that you 1:01:58.973 --> 1:02:01.073 could use to do that. 1:02:01.070 --> 1:02:04.170 It's illustrated--I clarified it a little bit in the version I 1:02:04.172 --> 1:02:06.952 put up. So, you run a regression like 1:02:06.949 --> 1:02:09.669 that to predict the stock market. 1:02:09.670 --> 1:02:13.340 This is actually a hands-on experience that's supposed to 1:02:13.338 --> 1:02:16.608 help you eliminate your overconfidence by trying to 1:02:16.613 --> 1:02:18.123 predict the market. 1:02:18.119 --> 1:02:25.409 This is the example where I tried to use time as a predictor 1:02:25.408 --> 1:02:32.818 of the stock market and failed pretty decisively to do so. 1:02:32.820 --> 1:02:37.830 What I want to say is that I have this spreadsheet up here 1:02:37.831 --> 1:02:42.581 that has some data--it has monthly--this is my 130-year 1:02:42.578 --> 1:02:47.618 long stock price series, but you could add other data 1:02:47.617 --> 1:02:53.047 and whatever--if you can find data series somewhere it would 1:02:53.047 --> 1:02:57.647 be more fun to try to predict using other data. 1:02:57.650 --> 1:03:00.740 This is just for you to really try to do it. 1:03:00.739 --> 1:03:04.539 Some people do sports things, so if somebody wins the Super 1:03:04.535 --> 1:03:08.395 Bowl--I don't know what the story is--this is a famous story 1:03:08.396 --> 1:03:12.646 actually--the stock market goes up or goes--do you know that. 1:03:12.650 --> 1:03:15.920 I don't know this exact repeated story--so you could 1:03:15.922 --> 1:03:19.452 create other variables like a dummy variable for winning 1:03:19.452 --> 1:03:23.112 the--somebody winning the Super Bowl and put that in. 1:03:23.110 --> 1:03:27.070 There's a famous story--it goes back to the 1930s--about skirt 1:03:27.068 --> 1:03:29.078 lengths and the stock market. 1:03:29.080 --> 1:03:30.140 Do you know this story? 1:03:30.139 --> 1:03:33.599 In the 1920s, an unprecedented thing happened 1:03:33.603 --> 1:03:37.383 in women's fashion, never been seen before in the 1:03:37.382 --> 1:03:41.232 United States, maybe in--women started wearing 1:03:41.233 --> 1:03:44.063 short skirts and it was scandalous. 1:03:44.059 --> 1:03:47.119 They weren't quite mini skirts, but they were scandalous. 1:03:47.119 --> 1:03:51.809 The women's hemlines rose and peaked in 1929 and then the 1:03:51.813 --> 1:03:57.013 skirt lengths came down in the 1930s, right with the market; 1:03:57.010 --> 1:03:58.300 so that was noticed. 1:03:58.300 --> 1:04:02.400 Some people thought there was some euphoria that was driving 1:04:02.402 --> 1:04:06.162 women crazy or something about the 1920s--the optimism, 1:04:06.156 --> 1:04:09.566 the sense; it sort of happened again in 1:04:09.568 --> 1:04:12.118 the '70s. Remember, the mini skirts came 1:04:12.116 --> 1:04:13.286 in the 1970s, right? 1:04:13.289 --> 1:04:16.719 Then the 1970s--'74 crash didn't exactly--I don't know if 1:04:16.722 --> 1:04:18.012 hemlines came down. 1:04:18.010 --> 1:04:21.980 But anyway, I had one student who thought, well maybe there 1:04:21.977 --> 1:04:26.077 are other fashion things that explain the market and she went 1:04:26.082 --> 1:04:30.462 back to microfilm newspapers and measured the width of men's ties 1:04:30.460 --> 1:04:32.650 in fashion advertisements. 1:04:32.650 --> 1:04:35.430 She thought, wide ties are a sign of--it's 1:04:35.432 --> 1:04:39.572 like a short skirt I guess--a sign of optimism and excitement, 1:04:39.573 --> 1:04:42.563 so she collected data on widths of ties. 1:04:42.559 --> 1:04:46.099 She had a time series--this is a very good answer to a very 1:04:46.100 --> 1:04:49.580 good problem set--and she collected fifty years of data on 1:04:49.579 --> 1:04:52.689 the width of men's ties and correlated--to see if it 1:04:52.692 --> 1:04:54.342 predicted the market. 1:04:54.340 --> 1:04:58.890 Unfortunately it did not, but it was a wonderful choice. 1:04:58.889 --> 1:05:02.419 I'm hoping that some of you can think of interesting things to 1:05:02.417 --> 1:05:04.727 do to try to predict the stock market. 1:05:04.730 --> 1:05:09.000 Alright, I'll see you again in two days.