[MUSIC] So in this video, we're going to talk about loss aversion. Loss aversion is a very key I guess, behavioral bias. There's a lot of evidence that a lot of people are subject to this, and it has real effects on everything from firm decisions, from the housing market, to how people trade stocks. So it's kind of worth talking about. The basic idea behind loss aversion, is that losses hurt more than gains make you feel good. Prediction is that people will be risk-loving when dealing with losses and risk-averse dealing with gains. So that would lead to the prediction, that if you have a loss, you're going to kind of hold onto it awhile, hoping prices will rebound, you're risk-loving. And if you have a gain, you want to cash this in, because you like the feeling of kind of being a winner, and you would hate the feeling if that gain would lose value and turn into a loss. So the benchmark around which you measure losses and gains Is extremely important. We'll talk later how financial companies, the government, will work to kind of shift this benchmark, okay? Because it's very important in how a customer, how a taxpayer will view things as whether it's a gain or a loss. And this is based on work by Kahneman and Tversky, they're very famous for their work on prospect theory. Kahneman actually got a Nobel prize in economics for this, kind of very path breaking behavioral finance. Applying psychology to how people are making economic decisions, okay. So let's draw a graph of welfare that we might think about from a standard economic model. And then we'll see how loss aversion, what that graph of welfare looks likes. So you might simply think, if we're doing a graph of welfare, let me be very very simple here, where we have, I'm not going to call this utility. We'll leave that for the people who are teaching econ, but let's just give it a real name, let's just call it welfare. How happy are you? We'll put welfare in the y-axis and we'll have how much money you have on the x-axis. You might think that's a straight line or, you might think well maybe, that's a concave function. But you just simply are looking at your welfare as a function of how much money do you have, okay? So what does this welfare look like, this welfare graph, with loss aversion, okay? So here we have welfare, Okay, here we have what's important in loss aversion, is people are doing mental accounting for different transactions or different assets. And what's key to them is this asset at a gain or a loss. So on the right hand side, this is a gain, on the left hand side, it's a loss. So you can think of your welfare function being something like this, which is called an S-curve. Okay? So as you go from right here is where you're basically, there's not a gain or a loss. The asset has basically the same price as when you purchased it. As the asset goes up in value, you're happier, but there's kind of diminishing returns. Okay, if the asset loses a little bit, you're very unhappy. Your welfare falls dramatically with just a little loss, but once it's lost a certain amount, losing more doesn't affect your welfare too much. So you can see there's this gigantic increase in welfare going from a small loss to a small gain. That's this S-curve utility or welfare function that loss aversion is known for. Okay so, Andrea Frazzini in a paper, drew some nice graphs to illustrate how loss aversion would affect whether people sell a stock with a gain or sell a stock with a loss. So rather than me recreate these, I thought why don't I just kind of use them, these come from Frazzini's 2006 paper. Other people draw similar graphs. We're looking at a hypothetical stock investment. And first we're going to consider what happens when it's fallen ten dollars, and then we're going to consider what's happens when it rises $10? So let's focus on losses first. So here you have a stock investment, it's fallen $10, so you can decide either to sell it now at this $10 loss, or to let it ride an additional year. In which case you think there's a 50% likelihood it will recover $10, so you're back to even, or it will lose an additional $10, in which case, your total loss is $20. So here we have this S shape utility curve or welfare curve. So what does that imply about the decision you'll make? So if you decide that, hey, let me cash in this stock at a $10 loss. Your welfare is down here, just taking this $10 loss. Going down, seeing what that means for utility. If you sell and realize a loss, your welfare is here. What if you let the loss ride? Okay, so you basically are going to hold it, see what happens. It'll either go up $10, so you're back to break even, or it will fall an additional $10 and you've lost $20 in total. Well then you have to take the average of your welfare in the two states. If you lose an additional $10, here's your welfare from the $20 loss. If it happens to recover and now you'll break even, your welfare is all the way up here. So if you take the average of the 50% likelihood of the breaking even, and the 50% likelihood of having now a $20 loss, that average of these two welfare levels is right here. So your expected welfare from not selling, from holding the loss, is actually higher than your welfare from selling it, so we would predict, people that have a losing investment will let it ride, will continue to hold it. So, when we focus on losses, what are some of the key takeaways? Greater expected welfare from holding a losing investment than selling it, so that's a key prediction. The opportunity for the loss to reverse, and get back to your benchmark, is so psychologically important, you don't really care that much about the possibility of further loss, it's kind of all that matters to you is breaking even or being a slight winner. That is so important to you, that further losses don't really affect you that much. If we want to speak like an economist here, we'd say your risk-loving over the loss region. You let the losing investments ride, hoping that they can maybe recover in value, because psychologically, that would be so important to you. Okay, now let's focus on gains. When you have a $10 gain, what's the prediction? So let's zoom in on this picture, again, taken from Verzini, Andrea Verzini's, Andrea Verzini, a 2006 paper here. Other people draw similar charts, I thought why not just copy what he has. So your stock has gone up $10 in value, okay. So if we simply cash that in, sell this winning investment, our welfare will be up here, right? Just following the curve, $10 gain. Our welfare is right here, okay? What if we let this stock ride? And we're assuming it's 50% likely that if we let the stock ride another year, you'll either lose $10, fall back to break even, or, it will rise an additional $10, and go up to 20. Okay, go up to a $20 gain. So those are the two possibilities. Let's see what the average welfare is, if we let the stock holding ride. It either falls 10, or goes up an additional 10. Well if it falls 10, our welfare is down here, if it goes up 10, then our welfare is up here. So you can see the gain in welfare, associated with the further $10 increase in the stock price, is not that much, but there's a big drop off in welfare if the price falls back to break even. So if we take the average to the break even point, the stock fell $10, and the $20 gain point, the stock goes up $10, we get, kind of, average utility right here. So we can see our average welfare from cashing in this $10 gain, is higher than our expected welfare, when we let the stock ride, and it either turns into a $20 gain or break even. That average welfare's here, so a prediction is sell, and realize the gain, cash in the gain. So psychologically, cashing in this gain, being a winner is very important, or another way to say it, you are terrified of the prospect of having this winning investment potentially turning into a losing investment by holding it longer. So when we focus on gains, what are some of the key insights? Greater expected welfare from selling a winning investment than holding it. Exactly the opposite of what we had over the loss, the opportunity for the gain to reverse and fall back to your benchmark is so psychologically painful, that you just don't want to do it. You don't care much about the possibility of further gains weighing more heavily on your mind is, I don't want this gain to reverse and turn into a loss. So again, if we want to speak like an economist, we would say we're risk-averse over the gain region. We don't want to kind of let the stock ride, we want to kind of cash in the gain while we have the opportunity to do so. So now let's think about this kind of loss aversion and let's look at a lottery question. So this is an actually a very popular lottery question here. Why don't you think what your response to this question would be. There's no right or wrong answer, just based on your personal preference. So, suppose you have a one time opportunity to take part in a lottery that's determined by flipping a fair coin. So it either comes out heads or tails and it's 50% likelihood for each. There's no cost to enter the lottery. Now, if you flip the coin and it's heads, you lose $50, okay? If you flip the coin and it results in tails, you'll win some amount of money. Now the key question is for you to be willing to have this coin flipped. And you know if it turns out heads, you're going to lose $50. What's the minimum amount of money you need on the upside, okay, if If it turns out to be tails? So obviously, if you tell me well flip the coin, if it's heads, you lose $50. If it's tails, you get a million dollars, everyone's going to say I wanted to do this gamble. The question is, what's the minimum you needed to be on the tail side, for you to be willing to do the gamble? There's no right or wrong answer here, just based on your personal preference. So what's the minimum upside you would need if you flipped the coin and it turns out tails to compensate for the $50 you lose, if it turns out to be heads? So think about that and then I'll tell you what the common response to this survey question is. So this is a simple survey question that is done to measure loss aversion. It was used by Tversky and Kahneman in their 1992 paper, so I thought I'd repeat it for you to kind of do this own test. What's a potential gain you need to offset a $50 loss. So in most surveys the common response is needing $100 gain to compensate for a $50 loss. So you need twice the upside. To compensate for the $50 downside. And this is extremely robust. Every class I do this, whether it's MBA students, MSF students, executive MBA students, Chinese bankers kind of visiting from, visiting Champagne from China. It's always the case that the median and modal answer to this survey question, will be this $100 gain. So this could be a measure of loss aversion. So take the 100 divided by 50. That kind of gives you a loss aversion measure of two. If you said you need $200 to compensate for the $50 loss, that would mean you need kind of $4 gain for every $1 of loss. Now from economic perspective to do the net present value, if it turns out, heads is a $50 loss, tails, should just be $51. And you're willing to do the bet, because you're making money in expected value here. But, this simple survey question suggests that people don't like losses, and they need higher gains to compensate for certain Loss. So key, that given this loss aversion turns out to be a pretty pronounced effect and a lot of different individuals is, what is the natural benchmark that people use to evaluate whether something is at a gain or at a loss, okay. So you know this is really key to loss aversion. What is the benchmark, okay? It's crucial to identify this benchmark if you want to test for how does loss aversion affect individuals, as well as does loss aversion have affect on broader markets, like the stock market, or the real estate market. If you want to do the test, well, you first have to identify ahead of time, what would be the natural benchmark that investors use. So for example, we're looking at people investing in stocks, probably their natural benchmark is what did I pay for the stock, what did I pay for the house, would be a natural benchmark. So loss aversion has so many effects on behavior of individuals and then broader markets. I thought I'd break this up into separate video segments. So, first we're going to talk about loss aversion and tax motivation for stock trades, particularly sales, so loss aversion has a prediction that people will be cashing in gains, holding on to losses. Tax motivation has exactly the opposite. Remember, Module 1, we talked about you should be holding on to gains, selling losses to get tax deductions, so how do those two counter when you look at the trades of individual investors. Personal connection with the asset. So, loss aversion based on emotion. Some assets, you might have more a personal connection like your house. Some you might have less connection with, like a mutual fund, okay. How does that affect the loss aversion effect on trading? When you look at assets, we have a stronger versus weaker personal connection. Corporate finance decisions, we'll look for evidence, are CEOs affected by loss aversion? Does that affect firm decisions? And if you're doing an acquisition, do the shareholders of the target company, are they maybe subject to loss aversion? And if so, how does that affect the type of price you offer to acquire this company? Can loss aversion be a potential explanation for the momentum strategy? So we talked about in the US, if you look historically at returns, if stocks have gone up the past year, continue to drift up for the next month, those have gone down the last year, continue to drift down the next month. Could loss aversion be a potential explanation for this momentum affect? And then finally, the importance of the endowment. And this is really just kind of marketing in terms of how do people set up this benchmark that people will then evaluate whether something is a gain or a loss. And by manipulating this endowment point or this benchmark point, you can really have strong effect on people's behavior. This is done by government, by financial companies. It's kind of marketing 101. So a lot of good stuff coming ahead in our discussion of loss aversion and its various effects.