Richard Thaler was recently announced as the recipient of the 2017 Sveriges Riksbank Prize in Economic Sciences in Memory of Alfred Nobel (aka The Nobel Prize) “for his contributions to behavioural economics.” Thaler is essentially the father and most noted proponent of the field of behavioral economics as well as its offshoot, behavioral finance. By most accounts (and my own reading of many of his published works and his memoir, Misbehaving), Thaler is also quite a nice guy – and rather un-arrogant. A refreshing contrast from most economists. But is he deserving of the Nobel?
The main criticism of the work of Thaler (and of other behavioral economists) is that its conclusions are plainly obvious, if not to economists blinded to reality, then to generations of psychologists, marketers and hucksters. And very true, such work would probably not be quite as agreeable to the Nobel committee for awards in physics, chemistry or medicine.
But leaving aside the question of whether this silly prize should even exist, I’d say Thaler is deserving. In helping to introduce psychology into economics, and in shepherding behavioral economics from its backwater infancy to mainstream acceptance and political prominence, Thaler has done more to influence the field of economics than perhaps any other academic economist in the past few decades. A lifetime achievement award as it is. Congratulations.
Obvious or not, the key insight of behavioral economics is that we the people, do not behave the way economists and their models thought we did. Or think we should. In a nutshell, we do not always make decisions that maximize our wealth. And by showing evidence of this fact time and time again using simple experiments, questionnaires and financial data, a Noble Prize was won. I have no problem with that.
But I do have a problem, and the problem is this. By not conforming to the way economists think we should and by not consistently maximizing our wealth in all our decisions, we, human beings, are labeled “irrational.” And this is a conclusion stated in virtually all articles about behavioral economics, whether in the mainstream media, or in economic journals. In fact as we’ll see shortly, evidence of “limited rationality” was one of the stated reasons justifying Thaler’s prize by the Nobel committee.
Irrationality is nonsense and the need to “correct” irrationality should not be used to justify government intervention in the economy or in our daily lives.
Rationality and utility
Now we must get slightly technical. What exactly does it mean for an individual to behave rationally? The colloquial definition is something like, “to make decisions using reason.” Let’s first make note of the fact (which we’ll return to later) that rationality implies making a decision. Next, let’s ponder what it means to use reason (i.e. sound judgement, good sense, logical arguments) to make a decision, or to make a “reasonable decision.” I’d say a reasonable decision (or a decision based on reason) is one that I believe or expect will be in my best interests.
Take note again, that I use the term “believe or expect” to indicate that at the time of a decision, my information is incomplete. That is, I don’t know the future. So, rationality does NOT imply a decision that winds up resulting in a good outcome. It solely implies that my intent in making a decision was in my best interests given available information. Equivalently, irrationality does not imply making a decision that winds up being a bad one, provided that I thought it’d be a good decision at the time I made it.
Finally, and most importantly, what do we mean by the words, “in my best interests?” Here we will use a term very familiar to, though often misinterpreted by economists. And here is where we will really begin to deviate from the economic mainstream. The economic term for “my interests” is “utility.” Correspondingly, the economic term for “in my best interests” is “maximizing my utility.” So what the heck does “utility” represent?
Now I must be less precise because nobody, neither philosophers nor economists, have agreed upon what exactly constitutes utility. Some say it is a measure of happiness or pleasure. Some punt and just say, it is whatever it is that I maximize. I think we can shed a bit more light.
I’m going to say that utility is an aggregation of all the stuff that evolution and biology have made us humans desire. These include, first, basic life necessities such as water, food and good health. For example, other things equal, my utility at any given moment is higher if I’m not thirsty, not starving, and not sick. Given that humans are social animals (with strong incentives to reproduce), our utility is also made up of social desires such as love, friendship, companionship, sex and status (status is something to which we will very importantly return again and again). And while not necessarily an exhaustive list of the various components of utility, I’ll propose a third category of all sorts of entertainment (or that which prevents me from being bored).
At any given moment of time (or least when we are conscious), each of us have some level of utility. The components of utility will clearly vary from person to person and within a given person, from moment to moment. Even though utility is not necessarily quantifiable, each of us are capable of judging or estimating whether a given action will likely result in greater or lesser utility to us.
When we say that we humans make decisions in order to “maximize utility” what we precisely mean is that we make decisions in order to maximize the present value of the sum of our probability weighted future utility over our lifetimes (or longer, if you believe in an afterlife). So two final but crucial wrinkles that we need to discuss.
By the term “present value,” we mean that the same amount of utility today is worth somewhat more than that amount of utility tomorrow. How much more depends on some “discount rate” which will also vary from person to person and may vary from moment to moment. And lastly, note that we weight future utility by the likelihood of the events occurring that would result in that level of quantity of utility. Our brains do this implicitly, kind of the same way a baseball player can calculate the optimal path to run in order to catch a fly ball without knowing the first thing about physics or parabolas.
Before we move on, I will admit a few things. First, from the standpoint of philosophy (rather than economics), nothing about my idea or loose definition of utility is all that controversial. Second, my definition is essentially what is known as a tautology. That is, a sane adult will only make decisions they believe to be in their best interests. In other words all decisions made by sane adults are by definition, rational. Third, up until now, I am mostly talking about definitions and semantics. But that is not really the purpose of this article. Its purpose is to argue that a bad definition should not be used to justify government action. Fourth and finally, the concept of utility is clearly nebulous and extraordinarily difficult, if not impossible to quantify. Hence, economists don’t use it in their models or when analyzing experiments. And herein lies the problem.
So what do economists use to predict decision making and to pass judgement on whether a particular decision or set of decisions is rational or irrational? They do one of two things. Mostly they use a much more quantifiable metric as a proxy of utility: money (or wealth or income). That is to say, rather than maximize the sum of the present value of probability weighted utility, I should maximize the amount of money that I have. If I don’t consistently maximize my wealth, then I am making bad (irrational) decisions. As we’ll see shortly when we discuss Thaler’s research, it is this error of using money as a proxy for utility that mostly accounts for economists’ misinterpretation of rationality.
Then there is the second methodology that economists use to analyze decision making and to judge (ir)rationality. Based on the work of game theorists, economists sometimes utilize an alternative and non-colloquial definition of rationality. Rather than rationality being defined as a reasonable decision made in my best interest (utility maximizing), economists set forth a set of technical rules for which rational decisions must satisfy. For example, decisions must be logically consistent (if I prefer coffee over tea and hot chocolate over coffee, I must always prefer hot chocolate over tea). Decisions must also be time consistent (if today I prefer pizza over a hamburger, I must also prefer pizza over a hamburger tomorrow). Moreover, it is assumed that individuals have a perfect and instantaneous ability to calculate mathematical probabilities. If any such axioms are violated, the individual is thus judged irrational.
There are a number of problems with this approach to rationality. First, as we spoke about when we made our own definition of utility, there is no reason to expect that preferences need be static over time, including the discount rate (how we value utility today versus utility in the future) with which we implicitly use to discount future utility. Clearly it is not irrational to decide to have pizza today and a hamburger tomorrow. Nor is it irrational to today forgo dessert (value my future health more than the instant satisfaction of fat and sugar) but tomorrow partake in dessert (value instant satisfaction more than future health).
It is also not at all reasonable to assume that individuals are perfect calculators. Very few people get perfect SAT scores (some of what passes for the questions asked of participants in behavioral economic studies and research very much resemble SAT questions). Should not having a perfect score be viewed as irrational behavior? I think not. Neither miscalculation, mistake, lack of education nor even stupidity should be considered the equivalence of irrationality. Remember, we said that a rational decision is one that I believe results in the best decision, not one that actually does.
Lastly, I want to strongly reiterate that this type of technical definition of rationality is not the colloquial one. I suppose that in a purely academic setting it is sort of okay to misuse common knowledge words, provided that you define your misuse. The problem, however, is that this alternative definition has not been confined to academic journals and seminars. Instead, it has carried over into the mainstream (media) where a now popular and pervasive belief in erroneous irrationality is used to encourage and support government policy to “correct” human irrational behavior in all sorts of markets and sectors of the economy.
Thaler and his Nobel Prize winning research
We’re finally ready to discuss Thaler’s research, and its application, or misapplication, to human rationality. As described by the Nobel committee, Thaler won his prize for his contributions to four components of behavioral economics: 1) limited rationality, 2) lack of self control, 3) social preferences and 4) behavioral finance. Let’s take a look at each in turn and examine whether it is appropriate based on Thaler’s research to conclude that we humans are indeed “irrational.”
Note that all of the quotations below, unless otherwise noted, have been taken from the Nobel Prize Committee’s press release or accompanying background material.
1 (a). Limited rationality: mental accounting
“Thaler developed the theory of mental accounting, explaining how people simplify financial decision-making by creating separate accounts in their minds, focusing on the narrow impact of each individual decision rather than its overall effect.”
One of Thaler’s research topics was, in his words, to try to understand, “how do people think about money.” Economists assumed that money is money is money, or in technical terms, “fungible.” Thaler noticed that people often think about money in a very different way, in what he called “mental accounting.” People might separate their savings into different pools of money, either purely mentally or with separate banking accounts or money jars. For example, individuals or families might have a separate pot of money for housing, food, clothing, vacations, long-term savings, etc. In fact, most of us do this in one form or another.
Most economists, Thaler included, view this kind of “mental accounting” as less than fully rational behavior. For example, let’s say my “food money jar” is running low and is insufficient to buy this week’s groceries for my family. Let’s also assume that there is plenty of money in the “vacation money jar” that won’t be needed for a while. Assuming money is money, economists would say the rational thing to do is take money from the vacation jar and use it for food. But this is not what many people do. Instead they might choose to work extra overtime this week, or take the time and sell some stuff on eBay in order to fill up the food jar so as not to take from the vacation jar.
Is this behavior irrational? If rationality means simply maximizing wealth than clearly the answer is yes. But I don’t believe that is what rationality means. Let’s think about how money impacts my utility. I cannot drink money or eat money nor does looking at it provide much entertainment. Hence, money does not directly result in utility. It indirectly results in utility in at least three ways.
Most obviously, money allows me to consume goods and services in the future that will result in utility down the road. In other words having money now increases my future utility, which (present valued) is what I am trying to maximize when making decisions. Second, having money now contributes to my feeling of “status” which I believe is one of the primary components of utility. That is, feeling wealthy (or wealthier) makes me feel superior (or less inferior) about myself. Third, having money provides a sense of freedom and reduces stress and anxiety. Technically speaking, having money increases the ability for me to make choices in the future, some of those choices which might lead to higher utility.
Let’s now return to the question of money jars. I don’t take money from the vacation jar to use for groceries because I’d feel badly or guilty doing so. In other words, the status or self-worth component of my utility would be lower if I did. I’d rather preserve my status or self-worth and give up some leisure time to make money some other way (e.g. overtime or eBay). In my view, this is perfectly rational behavior.
But, I don’t blame you if you’re not yet convinced. You might be thinking, whoa..wait a minute! I’m making a decision based on “feelings” and not “reason.” Isn’t that the textbook definition of irrationality? No. Consider the following.
Lots of people choose to drive a BMW instead of a Honda or a Chevy even though all three vehicles provide virtually identical utility when it comes to their primary purpose of an automobile, transportation. That is, they all will generally safely get you to the same place at the same time. BMWs of course cost more so should I be considered irrational to spend extra money to own one? No, because they provide additional utility beyond simply transportation. The owner of a BMW derives utility from the sportier drive, the plusher seats, the better sound system, and most importantly, from the status or superiority that such a luxury brand provides. In other words, they feel better in a BMW.
The same arguments can be made for the decision to live in 10,000 square foot mansion rather than a small house even though both can equally provide necessary shelter. Or the decision to wear fancy designer clothes rather than last season’s basic hand-me-downs even though both can equally provide necessary warmth and coverage. Or the decision to eat meals at 3-star Michelin restaurants rather than neighborhood diners even though both can equally provide necessary nourishment.
All of these instances of everyday human activity show that people make decisions all the time that choose “feelings” over wealth. Just like with the money jars. And if you are going to argue that all of these kinds of decisions are irrational than you will wind up maintaining that nearly all decisions that humans make are irrational and that humans should never spend any money at all except for the most basic necessities (or for investments that will provide future basic necessities). Which is also essentially saying that you have zero insight into human decision making. A useless endeavor.
In a similar analysis to the jars of money, Thaler pointed out evidence of mental accounting and concluded “limited rationality” from the observation that many people have both outstanding credit card debt AND money in their savings accounts. Isn’t it irrational to pay high credit card interest rates when you could partially or fully pay off your debt by using your savings? Again, the answer is not necessarily.
Thaler has argued that perhaps the reason for this is self control, or lack thereof. That is, having credit card debt prevents me from spending even more. If I paid off my credit card debts with my savings, I might spend excessively and once again rack up credit card debt. This is a plausible argument and one that, in my opinion, is evidence of rationality, not irrationality. However Thaler would likely argue that the fact that we humans do have issues with self control is in and of itself irrational. We will return to the topic of self control shortly.
I will propose some alternative reasons for why people might carry high interest credit card debt while having money in their savings account. Perhaps it is considered socially acceptable to have credit card balances (something encouraged by credit card companies) but not socially acceptable to have zero money saved. I might feel guilty (lower social status, lower self worth) telling a friend or family member, or even just knowing that I have a zero savings balance. I do not necessarily feel the same level of guilt having credit card debt. Since I view social status as I primary component of utility, I view this behavior as perfectly rational.
Alternatively, I might rationally view my savings as more valuable than my credit card debt. To an economist this would not make sense since money is money and net worth is net worth. But my credit card can be maintained indefinitely provided I pay the minimal interest payments. Once the savings is gone, its gone. I may feel that I have more certainty, more flexibility or more of a safety net knowing that I have savings AND that I can continue to maintain a credit card balance.
A final rationale for having credit card debt and savings is that many people don’t realize or don’t understand the high interest rates they are paying to service the debt. To me, this is pretty stupid behavior. But it is not irrational behavior. Remember that to be rational is to “believe or expect” that a decision is in my best interests. Not understanding the amount of interest I must pay is certainly dumb but also means that it cannot be considered an irrational decision. Thaler and other behavioral economists would argue that in this sort of situation government needs to step in (see “nudging” below). If you want to make this argument (I would not), at least be honest. Government is intervening to correct stupidity, not irrationality.
A third often cited example of so-called mental accounting and limited rationality is a study performed of New York City taxi drivers. The study showed that drivers tend to target a certain amount of income each day (what Thaler refers to as a “reference point”). If it is a good day and they can make their target income early, they stop working. If it is a bad day they work later until they meet their target. This behavior is viewed by Thaler and others as irrational since drivers drive less on days with high demand and more on days with low demand, contrary to the laws of basic supply and demand learned in Econ 101. But is it?
First, understand that the supply part of “supply and demand” refers to firms not individuals. The implicit assumption is that firms always maximize profits and when experiencing high levels of demand, firms will expand their production capacity and new firms will enter the industry until some “equilibrium” is met. However, here we’re dealing with individuals, not firms, and the key insight of this article is that individuals do not necessary maximize profits (wealth). Moreover, unlike textbook supply and demand circumstances, taxi prices are not allowed to rise (or fall) with changes in demand (unlike Uber, for example, with its surge-pricing), nor can industry capacity (more taxis) easily be increased.
So, supply and demand clearly has limited relevance here. But what about the fact that economists call taxi driver behavior irrational because they are valuing money (work) and leisure (non-work) differently on different days, depending on the demand for taxi rides. Or in other words, why aren’t they increasing their own capacity (driving more hours) in response to high demand and lowering their capacity in the face of low demand?
Thaler maintains that the study of taxi drivers shows that they tend to have an income goal (a “reference point”) for each day’s work. I agree. The question is, is there a possible rational explanation for this? I think so, and once again I refer to the component of utility comprised of status or self worth. On high-demand days, I feel like I got a good deal. I can go home early and enjoy my leisure time, unlike other drivers (or any other workers) who may still be working. On low-demand days, I feel good because I worked harder (longer) and still made my target income. Had I not worked longer, I might feel like a quitter or even a failure. Moreover, I’d argue that for taxi drivers, having a daily income goal (rather than just maximizing income) in and of itself contributes to utility because it makes a stressful and lonely work day more palatable (or less unenjoyable).
Before moving on, I’ll say one more thing about the taxi driver study. Indulge me for I will now make a completely unscientific guess. I will speculate that taxi drivers with spouses exhibit this income goal (reference point) behavior more than those taxi drivers without spouses at home (though they still exhibit it too). If I return home late, I am met with something like, “Why are you late for dinner?” If, on the other hand, I come home with low pay, I am met with an even more serious, “Why didn’t you make more money today?” Either way, I am made to feel guilty (those without spouses make themselves feel guilty, but to a lesser extent). The feeling of guilt is tied to low status/self worth and lower utility. In my view, it is perfectly rational to sacrifice a small amount of income, or work a longer day, to avoid being made (or making myself) feel badly.
The final area of Thaler’s study on the subject of mental accounting that I wish to discuss is the observed fact that individual stock investors are more likely to sell winning stocks and hold on to losing stocks. This is known as the “disposition effect” and is related to some of what we will discuss later on when we turn our attention to behavioral finance. Part of the insight of behavioral economists is that investors treat each stock as its own mental account, rather than try to maximize their entire portfolio or net worth as they assume a rational being should.
I have little doubt that this so-called disposition effect is indeed correct. And once again, the behaviorists have successfully shown that we humans are not simply wealth maximizers. But surprise, surprise, they have not shown that we are irrational. As an individual investor, I get utility from owning a stock and especially from making a winning trade that goes far beyond the monetary value of the gain. I feel smart, brilliant even! I tell all my buddies at the bar, and strangers at cocktail parties, and my spouse, how great a stock-picker I am! Utility is not money. As we’ve talked about many times, utility includes my feelings of status and self worth. I’ll gladly (and rationally) trade a few bucks in exchange for the world to think I’m the greatest investor since sliced bread. No difference from trading a few extra bucks to be seen in a BMW instead of a Chevy.
Of course, if I wait too long to sell the stock, there is a chance its value might go down. I may even lose money! Better to book the gain and be brilliant than risk losing my perceived brilliance. Are my friends going to think I’m more brilliant because I have a 32% gain rather than a 28% gain? Probably not. Better to leave a little money on the table rather than risk losing all the gain (and all my perceived brilliance). In other words, booking the brilliance provides me more utility than the additional monetary gain. And what happens if I hold on too long and the gain is lost? Now I will feel all sorts of guilt (low self worth) from myself and others for not being smart enough to get out when I should have. As we’ll discuss shortly in the section on loss aversion and the endowment effect, this hurts even more.
Similarly, I have lower utility from a loss beyond its monetary value. As long as I hold on, there’s always the chance of reversing the loss. As soon as I sell, I’m an idiot and my utility is lower. But as long as I hold on, I’m not. The chance to not be an idiot outweighs the monetary loss of a further decline in the stock price. For this reason, and another of the subject of Thaler’s research (though not directly cited by the Noble Committee) sunk costs can be viewed as perfectly rational behavior.
1 (b). Limited rationality: the endowment effect
“He also showed how aversion to losses can explain why people value the same item more highly when they own it than when they don’t, a phenomenon called the endowment effect.”
In 2002 a psychologist named Daniel Kahneman was awarded the economics Noble Prize for a set of ideas he (along with the deceased Amos Tversky) derived about human decision making called prospect theory. The most important component of prospect theory is something called “loss aversion” or the idea that losing something lowers our utility a greater amount than obtaining the object had raised our utility. In other words, an asymmetry exists whereby losses are more painful than gains are pleasurable.
Richard Thaler had been the first economist to apply prospect theory, and specifically loss aversion, to the realm of economics. In fact, he essentially relabeled the idea as the “endowment effect,” noticing that people tend to value things they own more than they valued them before they owned them. Loss aversion and the endowment effect are no doubt correct. But are they indications of irrationality, as Kahneman and Thaler and many others would have us believe?
One of Thaler’s most well known pieces of research on the endowment effect is his coffee mug study. Briefly, he gave half of a class of college students free mugs (retail value $6) and allowed them to trade with the other half of the class that did not receive the mugs. As it turned out, very few trades were made and it was observed that the median price at which sellers wanted to sell a mug was about twice as high as what buyers were willing to pay for a mug. In other words, those students who were given mugs valued them twice as much as those who were not given mugs. The endowment effect in action!
There are at least three reasons why I think it is perfectly rational to value something more when I own it, compared with its value before I owned it. The first and most important reason is that once we own an object, we now gain additional utility from the good memories, sentiments and emotional attachments that the object brings. For instance, every time I drink coffee out of that mug, or even see it sitting on my dorm room shelf, I will derive utility from the memories that I won a free mug from an economics professor! How fortunate! How cool! How many students can say that?
In other words, valuing the mug before I owned to after I own it is not an apples-to-apples comparison. While the physical mug has not changed in any way, its usage has changed, and hence its value. It is no longer simply a receptacle for hot liquids. It is also a receptacle of good, and status-increasing memories. It is no longer a mug. It is now my mug. It is no longer identical to hundreds of other mugs sold in the campus bookstore. It is unique.
I know what you might be thinking. What a ridiculous argument. Why should an everyday object magically change in value from one moment to the next just because I own it now? To be clear, it is not that the object has changed, it is that the object’s value has changed to me. Consider the following.
Let’s say you are a huge basketball fan. Let’s say you are walking down the street late one evening and you run into Lebron James, and he’s so friendly that he gives you a jersey that he wore in that night’s game. Assume that nobody witnessed the gift and that you have no certificate of authenticity so you could never sell it to a collector as a game-worn jersey. Would it be worth more to you than the same #23 Cleveland Cavaliers jersey you could buy in any sporting goods store? An economist would say no. In fact, an economist would probably say its value (and hence, utility) as a piece of clothing is actually lower than a brand new shirt, precisely because it has been washed and warn, and therefore has a shorter useful life.
But of course, you, a huge basketball fan, will value it much higher than a new shirt. Wearing it will make you feel special even if strangers have no idea who once wore it. Your friends will be jealous. You can daydream about passing it down to your future kid some day and telling him or her the story of how you obtained it. Point being, you get much more utility from the jersey than any other otherwise identical shirt you could have bought at a store. And because of that, its value is greater to you. And that is totally rational. Same for the coffee mugs.
A second reason why the endowment effect can be considered rational behavior has to do with how much time and effort you spend predicting an object’s value to you before and after you possess it. Before you own something, there is, obviously, a less than 100% chance you will come to own it. It is therefore rational to limit how much time and effort you expend anticipating the object’s use to you. Naturally, the greater the chance of ownership, the more time and effort you are likely to expend. Once you own an object, it now makes sense to expend additional time and effort to analyze the object’s potential usage.
Yes, I know that’s a bit confusing. Let’s use the mugs to make it clearer. Before I own the mug, I may give it a quick thought and say, “That would be a great mug for coffee or tea or hot chocolate.” Once I own the mug, I may upon further thought say, “not only can I use the mug for beverages, but it would also be a great holder for pens or spare change, or be a paperweight, or just look pretty on my desk, or maybe I can re-gift it…” I see more possible uses for the mug, more future utility, and hence more value.
I would argue that this value discrepancy pre-ownership and post-ownership due to the differential certainty of ownership is exacerbated when making decisions about objects of small value. Think about it. How much time is it really worth investing in thinking about the uses for a mug before I own it? In my view, this is especially true in academic behavioral economics research where questionnaires or even artificial trading do not involve real decisions, only theoretical ones. We’ll return to the issue of the applicability of research studies very shortly.
The third reason supporting the rationality of the endowment effect relates to the role of “status” in determining one’s utility. This is something we’ve discussed previously and will return to again and again. Thaler’s coffee mug study does not simply involve the respective valuations of the mug by owners (sellers) and potential buyers. It also involves the decision on both sides of whether or not to make a deal.
If I was one of the student’s fortunate enough to receive a free mug, I’m probably going to be reluctant to sell it for much less than its $6 retail value for fear of being viewed (by other students, or even by my own self) of making a bad deal. Now, think about the mindset of a potential buyer. I know that the student who received the mug got it for free. Why should I pay full price for an item that my fellow student received for free? I could just as easily buy it for full price from the campus bookstore. Just like with the seller, I don’t want to feel like or be deemed by others a “loser” for making a bad deal. There is also an element of “fairness” involved. Why should I pay for something someone else got for free? We will talk about more about fairness when we discuss Thaler’s research on social preferences. However, I think this holds true even if the seller of the mug had to pay for the mug in the first place.
All of us, and economics students especially, are trained to “buy low, sell high.” The perception I have (my status or self-worth) as a good trader (making a good deal or avoiding a bad deal) might be worth more than the few dollars differential of a coffee mug. I might even think that that unstated purpose of the exercise is indeed to measure my trading prowess, further biasing the analysis of mug value.
This is one of the limitations of artificial academic studies (something we will return to very shortly). Decisions about buying and selling mugs do not only capture the value buyers and sellers place on mugs. The study’s results are also affected (biased) by other factors, notably in this instance, how participants view themselves as smart traders vis a vis their fellow classmates. This is similar to our discussion of locking in stock gains and avoiding stock losses so as to be viewed as smart, something that contributes to my utility.
Think again to automobiles. Most of us know that the value of an automobile drops significantly as soon as it leaves the dealer’s lot. The car hasn’t really changed, other than perhaps a handful more miles on it. Why then, would you not buy it from me for anything close to what I paid? Naturally, I probably wouldn’t sell it to you either for anything less than I paid. Kind of like the mugs. The point here is that buy low, sell high is ingrained in most of us. When we violate this, we feel like idiots, which lowers our utility.
Finally, just like we talked about sentimentality which increases the value of the mug because it is my mug, and not just any old mug, there is also sentimentality to how I obtained it. Say for example that I acquired it in a trade from fellow student for a low price. I got a great deal! Going forward that mug will bring me utility as I will recall the brilliant trade I made with another (i.e. inferior) student. This is distinct from the utility I might get remembering that I got it free from a professor.
The fourth and final rationale I will make with regards to the rationality of the endowment effect relates to the issue of sunk costs. As I mentioned earlier, taking into account sunk costs when making decisions is viewed as irrational behavior by economists. I disagree. Admitting a loss lowers my feeling of status or self worth and thus my utility. I don’t want to sell the mug once I own it unless I get a very high price, because doing is an admission that I made a mistake. I would rather tradeoff a small amount of lost money rather than be viewed (or viewing myself) as making an error. In my view, therefore, sunk costs can therefore be considered a component of rational decision making.
In any case, enough talk of coffee mugs. Let’s move on to death and disease. A second well known study on the endowment effect is one that Thaler is a survey given to students in a classroom setting. The following two questions were asked in the survey:
A) Assume you have been exposed to a disease which if contracted leads to a quick and painless death within a week. The probability you have the disease is 0.001. What is the maximum you would be willing to pay for a cure?
B) Suppose volunteers would be needed for research on the above disease. All that would be required is that you expose yourself to a 0.001 chance of contracting the disease. What is the minimum you would require to volunteer for this program? (You would not be allowed to purchase the cure.)
A typical answer given by students (in 1980 dollars) was about $20 for the first question (how much would you pay for a cure) and $10,000 for the second question (how much would you require to be exposed to the disease).
As I’m sure you have noticed, the probabilities of your death are equal in both questions. Either way you have a 0.001% chance of dying. So from a purely mathematical standpoint, valuing death the same, one should give the same answer for answers A and for B. Thaler and others concluded that an endowment effect is at work. That is, people are willing to sell health for way more than will spend to buy health. And of course, they infer that this large discrepancy is strong evidence of irrationality.
I want to hold off on addressing the question of rationality for a moment. We said a moment ago when discussing mugs that there is an element to bias in an academic study of decision making that renders conclusions questionable or even invalid. In the mug case, my skill as a trader and the utility gained from making a good deal might trump the economic value of a mug. Here, the study is much more unrealistic. In fact, this study is so far fetched that I’d argue its conclusions are essentially meaningless.
Let’s say you were participating in this research study. What might go through your mind as you read the two questions? I know what would go through mine. If I choose A, why can’t I still get the cure if and after I learn I have the disease? If I choose B, why can’t I also get the cure? How do you know the disease will be fatal? How do you know that the disease will kill me in exactly one week? How do you know it will be painless? How do you know the exact probabilities of contracting the disease?
The point I am trying to make is that a survey like this is so unrealistic, so unbound to reality, that deciding between questions A and B has little relation to real-life decisions. The even more important point I want to make is that the choice of A or B has zero effect on my utility, other than perhaps a small impact if I infer the survey is some kind of test of my intelligence (as we saw with mugs). When I take such a survey, I am more likely to think, what is the right answer? Which answer will make me look smart? What is the point of the survey? What I am probably not thinking all that much about are the realistic possibilities of my own death, which is of course the intent of the study.
My criticism of this type of behavioral research study is not unique. Many others before me have shared the view that much of the research in behavioral economics is unrealistic and does not require the test taker to make a true decision. Thus, how can it be used to opine on the question of rationality? Certainly studies have shown that people might be bad at calculating probabilities. But as we’ve said before, a lack of math skill is not the equivalent of irrationality.
Having said all that, for the sake of argument, let’s take the survey and its conclusion at face value, that loss aversion or the endowment effect is absolutely true. The question that follows is why might it still represent a rational decision? The answer, in my opinion, is that I would feel like an idiot, and others would consider me an idiot if I caused my own death. I know what you are thinking. That makes no sense because either way I made a decision that resulted in my own death (either by not paying for the cure or by risking the disease). I don’t think this is exactly true.
To use one of the favorite tools of a behavioral economist, let’s restate or “re-frame” the two choices.
A) You may have a disease. Do nothing and you will probably live
B) You don’t have a disease. Do something and you might catch it and die
Do these feel equivalent? Of course, I’ve simplified the statements and left out the probabilities, but the point I am making is that how a question is framed has an enormous impact on most people’s decisions. Behavioral economists would no doubt agree, as framing is one of the most researched areas of decision making. But, what a behavioral economist would conclude is that the way a question is framed should not affect a rational individual’s choice as long as the outcomes are equivalent. If it does affect my choice, I am irrational. As I’m sure you can guess, I disagree.
There are a number of alternative ways I could have framed this choice but the point I am trying to make is that in the first choice, either I already have the disease or I do not. I am not giving myself the disease. I am only deciding whether it is worth to pay for a cure. In the second choice, I do not have the disease. I am making the choice whether to risk being exposed. In other words, I make the choice whether to give myself the disease or not. In B, I kill myself. In A, I don’t kill myself, I just don’t save myself. Yes the outcomes (death) are the same, but from a decision making standpoint they are not at all equivalent.
Why does this matter? Let’s think about what happens to me given both choices if I do get sick and do not have the cure. Either way, my last week on Earth is going to suck. Presumably I am quite upset, facing certain death. But there’s something else. My guilt will be far greater had I chosen to risk exposure (choice B) than had I opted to not pay for the cure (choice A). Think of all those wrenchingly sad goodbye conversations with my loved ones if I chose B. “How could you have risk exposure to a deadly disease for a bit of money?!?!?” Whereas with choice A, it seems perfectly reasonable to not pay for the cure given the very low chance of disease. The point is that since guilt is a key component of status or self- worth, and since status or self-worth is a key component of utility, my utility will be lower in choice B, than in choice A. And since my utility will be lower for that last miserable week, it makes perfect, rational sense to require a lot more money to make that choice, exactly what the study showed.
Let’s discuss one final example of the endowment effect. In one of his early research papers, Thaler wrote about a gentlemen referred to as “Mr. H.” who mows his own lawn. A neighbor’s son offers to mow Mr. H.’s lawn for $8 (1980 dollars) but Mr. H. continues to mow his own lawn. Mr. H. is then offered $20 to mow his neighbor’s equivalently sized lawn but Mr. H. declines.
On the one hand, Mr. H. is saying that mowing a lawn is worth no more than $8. On the other hand, Mr. H. is saying that mowing an equivalent lawn is worth no less than $20. How can this be? Naturally, Thaler concluded that there is an endowment effect going on, that the price a person is willing to buy a good or service can be significantly lower than the price at which they are willing to sell the same good or service. No disagreement here. But what about the issue of rationality?
To understand why Mr. H.’s behavior can be considered perfectly rational we need to think first about the consequences to utility from mowing one’s own lawn versus not mowing one’s own lawn. When I mow my own lawn, there’s a sense of pride and accomplishment in my beautiful lawn-mowing job, which contributes to my status. I also avoid the negative status that stems from the guilt I receive from my spouse’s disappointment in me not mowing the lawn. I might want to demonstrate to my children the responsibility of chores. I also may avoid the guilt that I feel if I shirk my responsibilities as a homeowner. Finally, perhaps there is entertainment value in the actual mowing, being alone and with nature. All of these may contribute to my utility and may be worth far more to me than a small amount of money. Perhaps they are even priceless. That is, my neighbor might offer to mow my lawn for free, but I would still mow it myself.
Next, let’s discuss why I might not want to mow my neighbor’s lawn. I don’t get the same status from it. I don’t care that my neighbor’s lawn is beautiful. I don’t feel the guilt from my spouse or from myself for not doing it. It’s not my job as a homeowner since its not my home. Mowing my neighbor’s lawn is just a business transaction. Mowing my own lawn is not. Hence, they are decidedly not equivalent, even if the time and effort required for mowing are. In short, it makes perfect sense to mow one’s own lawn given that I get additional utility from it and it makes perfect sense to not mow my neighbor’s lawn for more money since I don’t get the same utility.
Before we leave the topic of the endowment effect, I want to point out two important conclusions that have been demonstrated by research. First the endowment effect is much weaker, if it exists at all, for goods that have easily defined and known monetary value. This should make sense since 1) cold hard cash or its equivalent has little sentimental value, 2) we think about the value of money all the time so there should be little difference in our predictions for it use before and after it is obtained, and 3) it is highly unlikely for a trade to be considered good or bad when the value of the item to be traded is obvious to both parties. The second conclusion of behavioral research is that professional traders generally do not exhibit the endowment effect. This also makes much sense since professionals tend not to become sentimentally attached to the objects they trade.
2. Lack of self-control
“Thaler has also shed new light on the old observation that New Year’s resolutions can be hard to keep. He showed how to analyse self-control problems using a planner-doer model, which is similar to the frameworks psychologists and neuroscientists now use to describe the internal tension between long-term planning and short-term doing.”
The second area of study that the Noble Prize committee cited as reasons for Thaler’s award is his research on the lack of self-control, and in Thaler’s opinion, government’s responsibility to correct people’s lack of self-control. Here, more than anywhere else in the article do I disagree with Thaler and his followers.
Thaler has stated that one of the things that first got him interested in studying decision making was the somewhat bizarre behavior of his academic colleagues and friends that tended to occur at dinner parties. That curious behavior involved bowls of nuts. Specifically, cashews.
Here I quote from Thaler’s book, Misbehaving:
“Some friends come over for dinner. We are having drinks and waiting for something roasting in the oven to be finished so we can sit down to eat. I bring out a large bowl of cashew nuts for us to nibble on. We eat half the bowl in five minutes, and our appetite is in danger. I remove the bowl and hide it in the kitchen. Everyone is happy.”
Thaler used anecdotes like this, and later, research studies to conclude that human beings lack self control, a conclusion that is surely true. But he also concluded that this lack of self control represents irrational behavior. Dinner guests say they are happier when the cashew bowl is removed. They knew it was ruining their appetite for dinner. Yet, they could have just stopped eating! This certainly seems irrational. Moreover, how could anyone be happier with less choice (no cashew bowl)? This is something known by behavioral economists as the “paradox of choice.”
I am now going to give two very different explanations for why I think that eating cashews should not be considered irrational behavior. I think the first is the stronger argument. As you’ll see, it is also a very different argument than I have used so far in this essay.
Eating the cashews is not an irrational decision because it is not a decision at all. Think about yourself in a similar circumstance. Do you decide you’re going to have another cashew and then eat it? Or does your body just do it without you deciding? Hand goes to bowl. Hand picks up nut. Hand goes to mouth. Repeat. Did you consciously make a decision to pick up a nut and put it in your mouth? No. This kind of action is unlike, for instance, deciding how much money to buy or sell a mug or whether to mow your lawn. Those require thought. Eating cashews from the bowl in front of you does not. Simply put, there is no decision being made.
Recall our definition of rationality: to make a decision that I believe to be in my best interests. Absent a decision, we cannot conclude rationality or irrationality. Eating a cashew in this case is little different from breathing, an involuntary activity of your body. You could also call it an addictive behavior. Either way, it is not a conscious decision, and therefore not an irrational one. It is also not an example of the “paradox of choice.” Because I don’t make a choice when eating the cashews, removing the bowl is not the same as removing a choice.
Thaler also noted that when the cashew bowl is far away (say, at another table on the other side of the room), people do indeed refrain from eating the nuts. They do not get up, walk across the room and grab a nut. That would require a conscious decision, and therefore could be considered an irrational one. But people don’t do this.
That was the first answer for why cashew eating is not irrational. A second possible answer is that eating the cashews actually does increase my utility even if I don’t want to admit it. People might say that they would rather eat a healthy dinner than a bowl of nuts, but perhaps they are lying. They might even be lying to themselves. Why would they lie? Because it is not socially acceptable to ruin one’s appetite by eating unhealthy snacks. It is not considered acceptable to have a dinner of nuts. That is considered by society to be weak and childish. And who wants to be considered weak and childish by one’s peers? Or even one’s self?
Evolution has given us humans a desire to eat fatty and salty foods. Hence my utility is higher. Similarly, why ever eat ice cream? Surely I can get the equivalent calories in a healthier package. But the fat and sugar of ice cream makes me feel good, it increases my utility. Admit it or not, cashews for dinner might not be the healthiest decision, but there’s no reason it can’t be the rational one.
Take your pick whether you prefer the answer that cashew eating is not a decision or that cashews are better than meatloaf. Both are probably true.
The planner-doer model
“Thaler used his research on self control to propose a model of human behavior he called the planner-doer model.”
Thaler hypothesized that a person has two selves, the planner and the doer. The planner tries to maximize the present value of lifetime utility. The doer is only concerned with current utility. Naturally there is conflict between the planner and the doer, but sometimes the planner can override the doer if sufficient willpower (some kind of cost) is used.
In my view Thaler’s planner-doer model is Ptolemaic, or maybe Freudian. It is confusing, unnecessary and wrong. First of all, if I’m deciding between a healthy fruit cup for breakfast or a chocolate doughnut do I really have a devil (doer) on one shoulder and an angel (planner) on the other? How exactly do they duke it out to make a decision? Second, how long does the “doer” have to make a decision until the “planner” kicks in? Is it instantaneous? What exactly is meant by current utility in this context? Isn’t breakfast in the future anyway?
Are chocolate doughnuts always the choice of the doer? Are they always disallowed by the planner? Do they always reduce my long-term utility? What about a chocolate doughnut once per week? Once per month? May I eat one once a year even? When I’m old can I eat one? Age 60? 70? 80? On my deathbed? Ever? What if I plan to eat a doughnut so it’s not an impulse decision? Would that be okay? Yes, I think I’ll plan to eat one for breakfast every day, starting tomorrow. That must be allowed since it’s a decision made by the planner in me, not the doer.
I’m obviously being a bit silly here. But the point I’m trying to make is that when you really think about the planner/doer model, it completely falls apart. There’s no obvious way to differentiate between the two decision makers unless you say something like the “doer” makes me fat, unhealthy and poor and the “planner” keeps me thin, healthy and rich. But that’s not a useful or valid model for an economist or for any other half-intelligent person.
We humans maximize the present value of our (probability weighted) future utility. That’s how we make decisions. How we weight the difference in value between current and future utility is exactly measured by the discount rate we implicitly use. No angels or devils, planners or doers needed. Of course, how we derive our discount rate is a good question, but one that I will not address except for this. Our discount rate can, and will change from time to time, contrary to the assumption of economists. Remember as we’ve stated before, there is nothing irrational about having that fruit cup today and that doughnut tomorrow.
Before moving on, let me cut Thaler just a bit of slack here. His definition of the “doer” is very vague. But, to the extent that the “doer” is a proxy for our cashew-eating involuntary decision making system of the brain, then I agree. Neuroscience research has indeed shown that there are multiple decision making systems of the brain. At the very least, there is one that involves voluntary (thinking) decisions and one that governs involuntary actions. However, as we stated above, actions made by this second involuntary system should be considered neither rational nor irrational since they do not involve conscious decisions.
Nudging
“Succumbing to shortterm temptation is an important reason why our plans to save for old age, or make healthier lifestyle choices, often fail. In his applied work, Thaler demonstrated how nudging – a term he coined – may help people exercise better self-control when saving for a pension, as well in other contexts.”
Thaler used his planner/doer model to infer that individuals tend not to act in their own best interests. That is, they favor the short-term over the long-term. Thaler co-authored an influential popular book called Nudge and a paper entitled, “Libertarian Paternalism is Not an Oxymoron” where he argued that individuals should have choices made for them by governments (or other entities) in situations where they make decisions believed not to be in their best long-term interests.
The concept of “nudging” has been applied to a number of areas where academics (and government officials) believe people make irrational decisions that favor short-term benefits in lieu of long-term interests. These include healthy eating, smoking cessation, education and organ donation. However, the area that has had more research and probably the most real world implementation is one that Thaler is most known for, retirement savings.
Specifically, Thaler posited that most people undersave for retirement since they do not have the (planner) willpower to override their (doer) urges to spend the money now. Implicitly, Thaler’s view is that this constitutes irrational behavior. Thaler’s research on retirements savings also demonstrated that many people do not participate in voluntary employee or government sponsored retirement programs, and thus miss out on valuable tax deductions and/or employer matching funds. This too he considered irrational.
To compensate for such short-term, irrational thinking, Thaler (and others) suggested that enrollment in retirement funds be made automatic. That is, instead of people having to fill out paperwork and choose to enroll in a savings plan (“opt-in”), they would be automatically enrolled unless they filled out paperwork stating their desire not to enroll (“opt-out”). Further, Thaler argued that funds be automatically invested in some sensible diversified portfolio (a default portfolio) rather than the individual having to choose the investments since individuals tend to pick irrationally. Thaler also suggests that contributions to retirement plans automatically increase as an employee’s salary increases.
Thaler labeled this libertarian paternalism or the more user-friendly, “nudge” and successfully advocated many companies and governments in the U.S. and U.K. to adopt such plans. On the surface, libertarian paternalism or nudging feels reasonably benign. It is not coercive because individuals can always opt-out. In Thaler’s view, a fully rational individual should be indifferent to a traditional opt-in retirement plan or a nudging opt-out plan since, either way, they have the ability to make the same choice (to save or not to save).
I, however, find four significant issues with this concept of nudging. First, how does Thaler or the government or anyone else know that my utility is higher if I save more? Second, if government does assume that long-term interests always trump short-term ones, there are an infinite number of situations where nudging could be applied. Where do we stop? Third, how do you prevent special interest groups from co-opting otherwise well-intentioned policies? Fourth, is nudging (libertarian paternalism) really consistent with liberty and freedom, at least as recognized in the U.S.?
1. Is utility really higher?
The most crucial assumption that Thaler and other nudgers make is that favoring the short-term over the long-term is a mistake. That, for example, spending today instead of saving for tomorrow is irrational. Is it really?
There’s no question that other things equal, people prefer to spend money rather than save it. To use the technical term, people have a high discount rate when present valuing their future utility. In Thaler’s view, this discount rate is far too high (“hyperbolic” in his words). Hence, utility today is valued too high, and the value of utility tomorrow (or, say, 30 years from now in retirement) is too low.
Let’s start with something easy. It may indeed be the case that most people do not understand how much money they will need when retired and hence how much they need to save for retirement. They may not understand the concept of compound interest. They may not understand the financial markets at all. Thaler’s view is that stupidity equals irrationality. Said differently, anyone that lacks the necessary education or knowledge or mathematical ability to make the same decision that would be made by a highly educated PhD economist should be considered an irrational being. As you know by now, I do not share this view. A decision should only be considered irrational if it is made knowing that it won’t be in your best interest. Not not knowing.
Second, let’s consider what Thaler would consider an irrationality “no-brainer.” If I don’t contribute to my 401k retirement plan, for example, I lose the tax deduction that the federal government (in the U.S.) grants me. In Thaler’s view, this is money lost and why would any rational human ever choose to lose money? But it’s not that simple. If I put money into a 401k, there are significant limitations and penalties if I want to use the money before I retire. The money is not free for me to spend as it would be if I put the same money in a savings account, or a normal (non-retirement) investment account. So yes, I lose the tax deduction but I retain access to my money. There is a trade-off. It is not necessarily irrational to give up the tax benefit in order to keep my own savings accessible.
The next question to ponder is why is it irrational to value the certainty of consumption now a lot more than the uncertainty of consumption some decades down the road? The answer of course is, that it’s not. Will I be alive in 30 years? Don’t know. Will my social security checks be sufficient to meet my financial needs? Maybe. Will I even care about status when I’m old the way I care about status now (remember that most consumption is really for the purpose of increasing our social status or self worth)? No idea.
Now we must get a bit philosophical. Take you, dear reader. I’m going to take 1% of your income and force you to save it for retirement. You can consume less now and you might be able to consume more later, much later. Is the present value of your utility higher now? Well, is it? Are you better off? I have no idea. Maybe it is. Maybe it isn’t. I don’t know. And that’s the key point. Neither does Richard Thaler.
Thaler says we should save, not consume (incidentally, this is the exact opposite behavior encouraged by the Keynesian policies espoused by nearly all mainstream economists). When we reach retirement, is it okay to consume then? Why? Should we save even longer? Just like with breakfast, is it ever okay to eat the chocolate doughnut? Do I ever get to enjoy my savings? Should I wait until I’m too old and feeble? What is the point of wealth if not to spend it? How do you know that spending later is better for me than spending now?
The same arguments we can make about retirement savings we can make about other areas for which nudging has been advocated. Take healthy living, for instance. Thaler would argue that individuals should be nudged to live healthier lives. But how does he really know that a person’s utility is indeed higher giving up soda or junk food or even cigarettes just to potentially live a little longer? Why would anyone assume that maximizing life expectancy is the equivalent of maximizing utility? Clearly evidence points away from this. All humans engage in behavior that reduces life expectancy in exchange for near-term utility. Some just go further than others. Where do you draw the line?
Recall also that addictive behavior, like our cashews, is neither rational nor irrational because it does not represent a decision. As much as I personally find smoking to be abhorrent (and addictive) behavior, I do not recognize it as irrational behavior.
Lastly, I want to address the point that Thaler makes that a fully rational individual should be indifferent to opt-in or opt-out. His view is that either way, an individual can participate or not participate. Therefore, from a utility standpoint, they represent identical choices. I find this argument unpersuasive. Firstly, many people might not know they have the ability to opt-out. To Thaler this, in and of itself, is stupid and irrational. To me, only stupid. Second, many will feel pressure to not opt-out since big-brother (either government or their employer) has made participation the default option. Peer (or big-brother) pressure affects status and self worth and hence utility so even though opt-in and opt-out both allow participation nor non-participation, their effects on utility are not necessarily equivalent.
2. The slippery slope
There is no question that we humans make choices that favor the short-term over the long-term. If government views this is bad, how far should government go to correct this behavior? Let’s return to retirement savings. Perhaps by default, 5% of my salary should be saved for retirement. Maybe 6% would be better. Or 7%? 10% of my income? Maybe 20%? Where does government draw the line? This is the slippery slope problem. Once you start down the nudging path, where do you stop?
How about healthy living? Clearly many people eat too much dessert and drink too much alcohol. We get fat, we get diabetes, we get liver problems, we don’t live as long as we might have otherwise. Maybe government needs to nudge. Perhaps when I go to a restaurant, it should be illegal for the restaurant to present to me a wine list or dessert menu unless I specifically ask for one. Perhaps there should always be a default order: green salad and grilled chicken. That’s what I get, unless I specify otherwise (maybe in writing, to make it even harder) for the steak and fries.
Maybe grocery stores should be mandated to put unhealthy foods on high, out-of-reach shelves. Maybe they should be in a separate section of the store, a section to which I need to (in writing again!) ask for entrance. Perhaps I should be automatically enrolled in a health club membership. Maybe a personal trainer should automatically stop by house every day to encourage me to exercise. Maybe they should even have a key to my house so they can get me out of bed in the morning to exercise.
Frankly speaking, these are not unreasonable debates. But the key point I am trying to make here is if you are going to advocate nudging, how do you decide where to nudge, and how do you decide how much to nudge?
3. Nudging and special interests
Now I am going to talk about an issue that affects all government intervention, not just nudging. That issue is special interests. For every government action, some entities are helped and some are hurt. There are always unintended consequences, and very often (perhaps even 100% of the time), those unintended consequences ultimately dwarf the intended ones. Said differently, when government gets involved, the cure is often (usually) worse than the disease.
Let’s return to retirement savings. If retirement funds increase, who benefits? Where is my 401k money going? To a money manager. To Wall Street. To financial services firms. To the stock market. Nudging retirement accounts has the effect of subsidizing Wall Street and financial markets. Is that really a good thing? Might it not lead to more power to Wall Street and the financial markets? Might it not lead to a greater likelihood of Wall Street bailouts down the road since government is really made the decision to put my money into Wall Street? Now they can’t let it decline?
Might now the financial industry lobby for even more nudging of retirement savings since these firms benefit? As I wrote earlier, if 5% savings is good, why not 6% or 10% or 20%? And of course, in all of this some industries have to lose. Perhaps traditional local and community savings banks where I would have otherwise put my money to save. Perhaps retail stores or restaurants where I would have otherwise spent my money.
Let’s say government wants to nudge towards healthier eating. Encourage certain foods, discourage others. Some companies gain, others lose. But which foods are even the healthy ones, which ones the unhealthy ones? Frankly, scientists have no idea. Eggs used to be good for us, then they were bad for us, now they are good for us again. Butter is bad, margarine is good. Now margarine is bad, butter is good. Fat kills, so eat carbs. Now, carbs kill so eat fat. And its not just food. The entire healthcare system suffers from such uncertainty. So why should government take sides, unless the evidence is absolutely overwhelming (as it is with smoking).
The real problem is that government involvement is ripe for decisions encouraged by special interests. Big companies with big lobbying budgets at the expense of small businesses without. These special interests almost always trump the best interests of the people. And that assumes that government officials and politicians even have the best interests of the people at heart, something of which I am skeptical.
4. The oxymoron
As I mentioned above, Thaler co-authored a paper called “Libertarian Paternalism is Not an Oxymoron.” I am by no means the first to argue that this title is emphatically wrong. Thaler clearly does not understand what the term libertarianism truly means. The essence of libertarianism is not that I will do something to you unless say no. The essence of libertarian is that I will not do to you unless you want it done.
Allow me some latitude to solidify this argument. Consider the issue of sexual consent. I will have sex with you unless you say no. Or the alternative: I will not have sex with you unless you say yes. At least in the U.S., both societal norms and the legal system have moved towards the latter statement. That is, sex requires affirmative consent. When it comes to the violation of our bodies, most of us clearly seem to prefer it this way. However, Thaler’s idea of libertarian paternalism espouses the former (consent is assumed absent a “no”). Just some food for thought.
Before moving on from nudging, let me say three final things. While I personally would not often advocate nudging by government, there are arguments that can be made in favor. Government is (for better, or worse) collectively the largest health insurer in the U.S. (through Medicare, Medicaid, public employees, veterans, etc.). It is therefore reasonable to argue that nudges in favor of healthy living, and hence lower medical expenditures are warranted, given the government’s economic stake in our health. Similarly, it is not unreasonable to argue for nudging with decisions that affect children, since the decision making processes of children’s brains are not yet fully developed. But what I do ask of those who, like Thaler, advocate for nudging is that they not base their arguments on human irrationality. For that is a fallacy.
Secondly, I am in agreement with the behavioral economists that yes, most people make lots of mistakes. They certainly do make decisions that wind up being not in their best interests (though they do not realize this at the time of the decision and thus they are still rational decisions). For the most part, we need to let people make mistakes, not have government correct them. People learn from making mistakes, and that’s how society improves. Obviously there are limitations here. But, I would argue government should only get involved not simply when the benefits are greater than the costs, but when the benefits are an order of magnitude greater than the costs. The bar must be set higher. The special interests, the unintended consequences, the inefficiencies of government involvement are just too great in too many circumstances. The cure must never be worse than the disease.
Lastly, I point out that to the extent a case for government involvement in markets or personal lives is overwhelming, government has four different ways in which to act. First it should educate. Only if that education fails should it incentivize, for example through sin taxes (to discourage undesirable behavior) or tax credits (to encourage desirable behavior). Only if incentives fails should it nudge. Notice that nudging is the third option, not the first. And finally, only if nudging fails should government force or coerce behavior.
3. Social preferences
“Thaler’s theoretical and experimental research on fairness has been influential. He showed how consumers’ fairness concerns may stop firms from raising prices in periods of high demand, but not in times of rising costs. Thaler and his colleagues devised the dictator game, an experimental tool that has been used in numerous studies to measure attitudes to fairness in different groups of people around the world.”
The third area of study cited by the Noble Committee is what they refer to as “social preferences,” which mostly means fairness. That is, people don’t always act selfishly, as naive economists, or at least their models, think they should. Another example of irrational behavior. Not so. Let’s talk about evolution for a moment.
Evolution works at the level of genes and our genes have one primary purpose – to replicate themselves. But genes can’t reproduce on their own. They are dependent on their host (e.g. us humans) to reproduce. Fortunately they have quite an influence on their host, as they provide their carrier with its basic programming. In other words, in order to maximize the chance that a gene reproduces, it programs its host to seek food and avoid danger and attract a mate, among many other things. The host is rewarded. It “feels good.”
As we stated at the very top of this article, it is this genetic programming that influences what constitutes our “utility.” Eating and being healthy and having sex clearly (other things equal) contributes to utility. And like many other animal species, we humans have been programmed to be social. That is, it is a lot easier to obtain food and stay healthy and find that mate if we interact with other members of our species. Long story short, while our genes might be totally selfish (they “care” only about reproducing), us humans cannot be. In order to survive and reproduce and raise our children, and have our children reproduce, we must interact with other humans. And very often engaging in social activities requires making decisions that appear to economists to not be in our best interests. But to someone who actually understands human behavior, these decisions are perfectly rational.
We chase wealth not for wealth itself but to attract a mate, to be “alpha-male” (or “alpha-female”), to feel strong and powerful and superior. We buy fancy cars and live in big houses and wear big jewels to signal our superiority to others the same way a gorilla pounds its chest or a peacock flouts its feathers. At the end of the day, it is not net-worth that contributes to our utility, as economists believe, but self worth. Net-worth is just a component of self-worth.
Of course, we cannot be solely selfish. Or at least most of us cannot. Society wouldn’t survive and most of us (and our genes) wouldn’t reproduce. We cannot chase wealth and power at all costs. If I steal from the grocer, true I may get a free meal. But if I continue to do so, the grocer may wise up. Forbid me from entering his store, one way or another. Now where will my food come from? Worse off will I likely be.
Selflessness also matters. We treat people kindly so they will return the favor. This is the oldest form of insurance there is. And our genes reward us for this. It is in their interest and ours. We feel good about it, and we are rational to do so. We punish the jerks among us (or at least try to) so that they will learn and correct their behavior and if not, leave our community altogether. And again, our genes reward us for doing this. It is in their interest and ours. And we feel good about it and we are rational to do so.
Let’s now take a look at some of Thaler’s research on human social behavior and fairness, beginning with price gouging. We will see how humans behave in a manner inconsistent with economics, but perfectly consistent with what evolution has made our genes, and with how our genes have programmed us.
In the 1980s, Thaler performed a study that showed that the majority (82%) of people found it “unfair” for a hardware store that sells snow shovels to raise the price of shovel from $15 to $20 the morning after a large snowstorm. Let’s examine two things. First, is it rational behavior for the hardware store owner to raise prices? Second, it is rational for snow shovel consumers to find this behavior “unfair?”
Economics 101 teaches us that prices should rise (other things equal) in circumstances of rising demand. The assumption here is that demand for snow shovels increases after a bad snowstorm. Hence, a simplified understanding of Econ 101 implies that the hardware store has justification for raising the price of shovels. But, as we’ll discuss next, consumers may very well be turned off by this “price gouging” behavior. Their distaste may lead them to avoid shopping at this store for any goods in the future. The may be so incensed that they arrange a boycott of the store. The point being that, from the shop owner’s standpoint, to raise prices is really a question of short-term gain versus potential long-term loss. A business that depends on steady, long-term relationships is probably best served (and rational) not to raise prices. A business that caters to one-time customers (say, tourists), may benefit from price gouging. There’s no right or wrong answer here, except that either decision can be considered “rational” (and long-term profit maximizing) depending on the circumstances. Of course, a socially-minded business owner (and much less likely a big public corporation) may decide that fairness trumps profits regardless. As we’ll see shortly, this too can be considered rational behavior (contrary to the beliefs of economists).
On to the more interesting question. Are consumers of snow shovels rational for finding this price gouging behavior unfair? I can certainly sympathize, for there is a feeling of being cheated. Why should the shop-owner benefit from the dumb luck of a random big snowstorm? Worse, why should the shop-owner benefit extra from my misfortune of having to expend time and money clearing my driveway? As we’ve described before, here’s an example where self worth trumps net worth. If I overpay for the shovel, I’ve been taken advantage of by a fellow human being. Put simply, I feel like a sucker. And I will continue to feel like a sucker every time I walk into that store from now until the end of time. Better to pay $20 to a neighborhood kid to shovel than to give an extra $5 to that greedy hardware store. Now every time I walk into that store, I’ll know, even if they don’t, that I didn’t let them cheat me! I’m happier, I feel better, my utility is higher, and therefore I’ve made an entirely rational decision.
Around this same time as his price gouging study, Thaler and his collaborators invented an experiment that has become known as the dictator game. In this study, students were asked to divide $20 between themselves and a random and anonymous fellow student. They had two choices:
1) Keep $18 and give away $2, or
2) Split the $20 evenly, keeping $10 and giving away $10
Clearly, the selfish, and (to an economist) rational decision is to keep $18. However, as you may have predicted, it turns out that the majority of students (76%) decided to split the $20 evenly, demonstrating that for many people, social considerations are more important than money. Why might this be?
First of all, I would suggest that the study’s assumption of anonymity is a faulty one. As a participant, might I not be questioning that assumption? What if the recipient somehow finds out that it was me, the greedy one, that only gave them $2? What if the teaching assistant finds out, or the professor? Do I want my T.A. or my professor thinking I am a jerk? If he or she thinks poorly of me, might that not affect my grade in this class? As I have mentioned before, this kind of artificiality, ambiguity, unrealism or bias, is why I find many of the conclusions of behavioral economics to be questionable.
But for the sake of discussion, let’s now assume that anonymity is not an issue. Let us assume that there is absolutely, positively, no way for anyone knowing who gave $10 and who gave $2 other than the individual who made the decision. Why might it still be rational to be “fair” and not “greedy?”
Here we return to the conclusion that self-worth trumps net-worth. My genes have programmed me to feel good to treat someone else fairly and to feel guilty to treat someone else unfairly, even if my actions aren’t known to others. As the study showed, most of the participants gave up $8 to feel good about themselves, and/or to not feel badly about themselves. As we’ve stated many times before, this behavior is little different than spending more money for luxury items to feel good about myself, or to give to charity, or even hold a door open for a stranger (which expends some small amount of energy).
Thaler and his colleagues decided to test another aspect of human nature. They extended the dictator game to include a second round with a third player. The third player had the following two choices:
1) Receive $5, give $5 to a (fair) student who had split the original $20 evenly in round 1 and $0 to a (greedy) student who had kept $18 in round 1, or
2) Receive $6, give $0 to a (fair) student who had split evenly in round 1 and $6 to a (greedy) student who had kept $18 in round 1
Just like in the first instance, the “economically rational,” wealth maximizing decision is choice #2, to take $6 over $5. But the majority (74%) of students choose #1, that is to give up the $1 difference in order to reward Round 1 players who were “fair” and punish Round 1 players who were “greedy.” Are we irrational beings because we are willing to sacrifice $1 to punish a jerk?
I don’t think so. Our body’s social programming has taught us that punishing a jerk is a type of investment. We are (or least attempting to) train the jerk to not be a jerk next time. Having fewer jerks in a community is a good thing. Perhaps the punishment now in an economics class will prevent the jerk from becoming another Bernie Madoff some day and getting more screwed later on. Certainly, these are long odds, but might it be worth $1 now to potentially save myself from getting cheated out of millions? Why not? Moreover, I get satisfaction (we call it schadenfreude) from the punishment. I feel superior, a better person. I have a higher sense of self-worth, and thus, greater utility.
Lastly on the subject of social preferences is a question that is frequently raised by economists who study decision making. Why do people voluntarily leave tips at restaurants they never intend to visit again?
Start with the fact that tips are not really voluntary. Gratuities may have started out that way, as a way to reward good service, but at least in the U.S., they have become a meaningful (sometimes majority) component of the pay of waiters and restaurant staff. That is, they are expected. But of course, there is no legal obligation to provide one, only a moral obligation. Not doing so violates a social contract.
As we’ve discussed, our genes reward us for maintaining the social contract and punish us for violating the social contract. For some, doing the right thing for its own sake feels good. I feel good treating people nicely. Others feel good knowing a stranger, as in the waiter, thinks highly of them. Yet others seek to avoid the guilty feelings that contribute to a sense of low status/low self-worth. I don’t want the waiter to think of me, until the end of time, as a cheapskate. Every time I go into a restaurant I may be reminded of my cheapness. What if I run into the waiter again someday, even if I don’t intend to return to that restaurant? Do I really want to take that risk? I will never get that out of my mind… For most people, all three of these social components of utility contribute to decision making. That’s why we tip.
4. Behavioral finance
“Thaler was one of the founders of the field of behavioural finance, which studies how cognitive limitations influence financial markets.”
The proliferation of computers in the 1980s allowed economists to become number crunchers. As a wise person once said, best to fish where the fish are. Best to number crunch where there are lots of numbers. And where are there lots of numbers? Financial markets. Specifically prices and trading data of common stocks and other liquid financial assets.
As economists turned their attention to financial market data, analyzing decades of stock market data with simple statistical tools, they noticed something. They noticed anomalies. Up until then, the prevailing assumption held by most economists and finance professors was that markets (at least highly liquid ones like the stock market) were perfectly efficient. That is to say, all available information is immediately priced into a security. The only way to outperform the overall market (or a market index) is to 1) take more risk, 2) have non-public information or 3) get lucky.
These anomalies seemed to show that by analyzing certain historical data, investors could in fact outperform the overall market without taking on extra risk. To most, this observation clearly contradicted the view that markets are perfectly efficient. It also led to a dramatic increase in the study of financial markets by academics (analyzing data on a computer in your office is a lot easier than running experiments on college students in your classroom or lab!) and spawned entire new asset classes such as quantitative hedge funds, smart beta funds, ETFs and factor investing.
Thaler co-authored one of the first prominent studies of market anomalies in 1985. Thaler compared stocks that had dropped in value over the prior few years (“losers”) with those that had increased in value over that same time period (“winners”). He found that the loser stocks subsequently outperformed the winner stocks. In other words, investors could generate positive risk-adjusted returns (“alpha”) by buying a portfolio of losers and selling short a portfolio of winners. Similarly, an investor could out-perform the overall market by simply buying a portfolio of losers.
Thaler offered up a behavioral explanation for the anomaly he uncovered. Based on earlier published psychological research, he posited that investors must “overreact” to information. In other words, after a company’s stock has declined because of poor financial performance (or some other bad news), investors hold the view too long that the stock is a bad one and that the poor performance will continue. Consequently they are too slow in reevaluating or reassessing if the news and/or the company’s financial performance has improved (or regressed to the mean). Similarly, investors overreact to good news and good financial performance and are too slow tempering down their positive views of a stock.
Over the years many academic papers on financial markets have been published and many such anomalies have been uncovered. However, most of these anomalies do not persist. That is, the out-performance tends to disappear after the publication, either because it was spurious to begin with (the product of data-mining or data-snooping) or because the strategy is traded upon and the profits get “arbitraged out” by investors once the anomaly becomes widely known. A prominent example is the so-called “January effect,” whereby stocks were thought to increase in price during the month of January after having fallen in December. This was believed to occur due to investors selling stocks in December in order to capture the tax benefits of capital losses (to offset capital gains). It is pretty much a given that the January effect no longer exists, if it ever even did.
There are, however, market anomalies that have seemed to persist even though they have been widely known for decades. The two most important are the value effect and the momentum effect. The value effect is an extension (and essentially a renaming) of Thaler’s discovery of overreaction discussed above. Recall that Thaler uncovered that stocks that had declined for the prior few years tended to outperform the market and stocks that had increased over that time period tended to underperform. Later research concluded that stocks that are cheap by some metric such as Price-to-Book Value or Price-to-Earnings (called “value” stocks) tend to outperform stocks that are expensive (usually referred to as “growth” or “glamour” stocks).
The second prominent anomaly still thought to be in existence is the momentum effect. Whereas Thaler looked at three years of historical data to determine whether a stock should be considered a winner or a loser, other researchers looked at shorter time frames, say 6-12 months. What they found was quite the opposite of Thaler’s conclusions. Stocks that have outperformed the overall market (i.e. increased) over the prior 6-12 months tend to continue to outperform (increase) over the next 6-12 months. Similarly, stocks that have underperformed the market (i.e. decreased) over the prior 6-12 months tend to continue to underperform.
Researchers labeled this the “momentum effect” and concluded that investors could do very well by buying a basket of short-term winners and shorting a basket of short-term losers. Similar to Thaler, researchers posited a behavioral explanation for their anomaly. Whereas Thaler said that investors overreact to longer-term good and bad news, momentum researchers argued that investors also underreact to shorter-term good and bad news. That is, it takes time for good news and good performance (and bad news and bad performance) to become fully appreciated by investors and hence, fully priced in to stocks.
Note that while the value effect and the momentum effect appear at first glance to be contradictory, this is not so. They are measured over different time periods. In fact, many quantitative hedge funds (and later ETFs and other “smart beta” products) have used the combination of these two “factors” as the basis of their investing strategies. Buy a basket of value (long-term cheap) stocks that have exhibited strong momentum (short-term gains) and short a basket of growth (long-term expensive) stocks that have exhibited weak momentum (short-term losses).
Why the long digression into quantitative investing factors and hedge fund strategies? As I stated, these two strategies or factors, value and momentum, along with a number of other less prominent strategies or factors seem to prove that market anomalies do exist. Here I agree.
The question then, and one we’ve asked many times in this article, is do the existence of such market anomalies prove that investors are necessarily irrational in their behavior, or as the Nobel Price committee stated, are they evidence of “cognitive limitations?” Most say yes. I say no.
There are at least five possible explanations for market anomalies. The first is that the anomalies are not real. They are spurious, the result of data mining or data snooping, the product of overeager PhDs desperate for an article to publish or an interview at a quant fund. No doubt many of the anomalies (factors) found over recent years fit this type. But, for factors such as value and momentum it is hard to make this case. These two anomalies have persisted for decades after discovery and have been confirmed in multiple asset classes (not only stocks) and in the financial markets of many different countries.
The second possible explanation is risk. That, for example, value stocks outperform growth stocks because they are inherently riskier (perhaps a greater risk of distress or bankruptcy). But, the argument goes, this implicit risk does now show up in traditional risk metrics such as volatility or Beta. In other words, yes, value outperforms growth but not on a risk-adjusted basis, if risk were measured properly. Some economists (especially those inclined towards efficient markets) have indeed made this argument in response to the research of Thaler and others.
I am sympathetic to this argument, though actually even more so for the case of the momentum anomaly than for value. I would argue that stocks with strong momentum are far riskier than traditional risk metrics imply. Said differently, what is considered to be “risk” is vastly underpriced. The primary reason for this is the implicit backstop of financial markets by central banks (something we will return to shortly when we discuss financial bubbles). Central banks over many decades have engaged in bailouts of financial markets repeatedly, and ever more strongly. Time and time again they have prevented prices from failing. Intuitively, the stocks that have risen the most (those with the strongest momentum) are likely to decline the most absent the backstop of central banks. Someday when a financial crises comes that central banks are unable (or unwilling, but much more likely unable) to curtail, the momentum anomaly will disappear. It will be shown that these momentum stocks were riskier all along.
The third explanation for market anomalies is one that Thaler has also extensively researched, and one for which he is heaped praise by the Noble Committee. This is something called the “limits of arbitrage.” One set of anomalies that researchers have unearthed occurs when two securities with the same underlying assets have prices that differ. This violates what is known as the “law of one price.”
One of the most famous examples of such an anomaly is one that Thaler discusses at length in his book, Misbehaving, the 3Com/Palm spinoff. Very briefly, 3Com was a tech company during the first dot com bubble. In 2000, 3Com decided in would spinoff a subsidiary, Palm, by initially selling a fraction of its stake in Palm to the public (about 5%). Then, months later, each 3Com shareholder would receive 1.5 shares of Palm stock so that 3Com would divest all of Palm. During those months between the IPO of the 5% shares of Palm and when 3Com divested the rest, the market value of Palm was substantially higher than the market value of 3Com. In other words, the stock market was valuing 3Com’s business excluding Palm at a substantially negative value! Given that the lowest a stock price can be is $0, this made no sense.
Thaler argues that, in situations such as 3Com/Palm, two things are going on. One, irrational (mostly individual) investors are driving up the price of Palm to irrational levels. Two, something prevents the smart (mostly institutional) money from correcting the mispricing. In the case of 3Com/Palm (and in many similar cases), even though everyone knew about the mispricing (it was widely reported on in the mainstream media), it was virtually impossible to short Palm stock given how few shares were outstanding. Without being able to short the stock, investors could not arbitrage the difference in value between 3Com stock and Palm stock and therefore could not “fix” the mispricing. This is known as a “limit to arbitrage.”
I want to hold off for a moment on the discussion of whether Palm stock was so highly valued because of “irrational behavior” or some other reason. But I do want to make the point that the violation of “the law of one price” is not in and of itself evidence of irrational behavior precisely because of the limits of arbitrage. That is, investors are trying to arbitrage out the value difference. They are trying to fix the mispricing. They are trying to enforce market efficiency. They are quite simply unable to do so because of structural limitations to markets (i.e. the inability to short a stock). We will come back to this issue again when we discuss financial bubbles.
The fourth possible reason for why market anomalies exist is indeed the one favored by behavioral economists like Thaler: irrational investor behavior. Economists tend to assume that most stock market investors are what they call “noise traders.” As Thaler has pointed out, another (less polite) name for them used by some economists is “idiots.” The idea here is that most individual investors do not buy and sell stocks based on fundamental data or rational analysis, but on emotions or animal spirits. I freely admit that being an idiot might qualify as a “cognitive limitation.”
It should not surprise you, however, that I do not concede that such investing behavior is necessarily irrational, for three reasons. First, as long as I believe that I can sell for higher than I bought, I am making a decision that, to me, is rational. I may not have done much, or even any, fundamental analysis. I may not even be capable of doing such analysis. I may not even know what fundamental analysis is. It doesn’t matter. I’m still making a trading decision that I believe to be in my best interests. And recall from our discussion earlier about the momentum effect. Momentum usually works. So losing myself to “animal spirits” by buying a high-flying stock, regardless of so-called fundamentals or valuation has very often been a profitable enterprise. We’ll return to this point.
The second reason why such investor behavior is not necessarily irrational is because maximizing my utility is not necessarily the same thing as maximizing my wealth, or the value of my stock portfolio. Other factors that makeup utility must be considered. I won’t say more about this yet, but we will also come back to this idea shortly.
To understand the third reason why economists are confused about irrational investor behavior requires busting one of the most basic myths of all of economics and finance. There is no such thing as fundamental value. All value is relative. The value of a stock, or any other liquid asset is what someone else is willing to pay for it. Even what is known as fundamental analysis (a discounted cash flow, for instance) requires implicit and explicit estimates of other market (or relative) variables.
I am going to give you what I believe to be a better answer for why market anomalies exist. But before I do that, one comment on the mainstream view of “cognitive limitations” or investor irrationality as an explanation for such anomalies. As I’ve said, prominent anomalies such as the value and momentum effects are believed to have persisted for a long while now. However, they do not always work. That is, there are long (multi-year) periods where one or both of these anomalies do not work. This seems to me to be inconsistent with the thesis of cognitive limitations. Why would investors be cognitively limited in only some years and not others? Our brains work in some years, not others? We do analysis in some years, not others? We are more emotional in some years, not others? I don’t get it.
Now, on to the fifth and final rationale for market anomalies, and as I stated just above, the one I find most compelling. I do believe that anomalies are indeed due in good part to investor behavior, consistent with mainstream theorists. I just don’t believe that this behavior should be viewed as irrational. This is of course consistent with the main theme of this entire article. Most of the studies published by Thaler and other behavioral economists are essentially correct. They rightly show that most decisions made by individuals are based on calculating factors other than purely what will maximize wealth. But for the umpteenth time, this is not irrational.
As we’ve alluded to a number of times, we can segregate investors into two broad groups, institutional investors and retail investors. Institutional investors are generally considered the “smart money” and retail investors, the “dumb money” or the “noise traders” or the “idiots.” In order to explain why investor behavior should be considered rational, I need to talk briefly about the motivations of each of the two investor types.
When we talk about institutional investors, we refer to entities that manage money for some group of investors. These include mutual funds, hedge funds, insurance companies, pension plans, endowments, and others. But at the end of the day, there is always a person (or persons) responsible for making the day-to-day decisions of what stocks (or other assets) to buy and what to sell. These are the portfolio managers and the analysts. Even in the case of quantitative funds, there have to be programmers who code the algorithms. The point here, and it’s a crucial one, is that “institutions” don’t make decisions. People make decisions.
What motivates people? They want to maximize their (present valued) utility of course. So what motivates portfolio managers? First and foremost, they probably want to keep their job. Second, they probably want to get paid a lot of money. Third, they probably want to feel smart (or not feel stupid) compared to their peers. Etc. Obviously, having really high stock market returns is likely to help you keep your job, get paid well and make you look smart. But, going after high returns generally involves taking a lot of risk. This is not the strategy employed by most professional money managers.
Instead, most portfolio managers act to minimize the risk of losing their jobs. And they minimize the risk of having their salaries cut by avoiding the kind of poor performance that gives reason for investors to pull their money out and have their assets under management (AUMs) decline. The end result is that the vast majority of institutional managers aim to hit a benchmark rather than maximize gains, and they become what we call “closet indexers” where their holdings mimic an index such as the S&P 500.
In other words, the primary goal of a portfolio manager is to do what everyone else is doing. That way, you keep your job, keep your AUMs and keep your nice salary. It’s okay to lose money as long as everyone else is also losing money. Similarly, it’s not worth taking high risk to shoot for the moon. Hence it is incredibly difficult to be a contrarian investor. You run too high a risk of losing the patience of investors, losing your AUMs, and losing your job.
This mindset of those that manage institutional money more than anything else probably explains the momentum anomaly. Because of their need to track benchmarks and to not be wrong, institutional money managers exhibit momentum investing probably even more so than “dumb” individual investors. Even professional investors want to look smart by holding a portfolio of winners, or avoid looking stupid holding a portfolio of losers.
For example, it is widely known that many mutual fund managers will buy expensive popular stocks towards the end of the mutual fund’s fiscal quarter or year. Why buy high? So that investors, who look at the fund’s list of holdings (published quarterly or annually) will see winners and think highly of the brilliant portfolio manager, even if the fund did not participate in the price run-up of that stock. Are mutual fund investors irrational for behaving this way? Not necessarily since mutual funds don’t disclose all of their trades.
The vital point here is that economists assume that all market participants are trying to maximizes trading gains. To do anything otherwise is irrational. But that’s not how the “smart money” works at all.
Now let’s talk about the other big class of market participants, retail investors. Retail investors are regular people, middle class or wealthy, who invest their own money in the markets. Typically they buy stocks directly in the stock market or purchase shares of mutual funds. Yes, they are generally (though not always) less sophisticated than institutional investors (“idiots” remember?). But like the portfolio managers at institutions, they buy and sell securities for reasons other than purely maximizing performance. They try to maximize utility, not simply wealth.
To many, stock market investing (really speculating) is also entertainment, not simply a way to save. Like going to Vegas, I pay for the entertainment. But unlike Vegas, where whether I win or lose is mostly based on pure chance, with investing, it is at least perceived to be based on skill and smarts (whether it really is, is another story). So, I get utility from the entertainment of picking stocks. And I also get utility from the status effect of picking winning stocks.
Let’s now try to explain the value anomaly. Recall that over time, value stocks have often outperformed high growth/high glamour stocks. In my view, this is because investors, especially retail investors, get utility from owning such stocks. This utility more than compensates them for the small loss of money they could have had if they had owned boring out-of-favor stocks. It is fun to own the stock of Disney or Apple or Google or Facebook. These are stocks I can talk about at cocktail parties and water-coolers. Anybody want to talk about the insurance company or electric utility stocks that I own?
Even the smart money exhibits this behavior. They go to parties too, and industry conferences. More fun to talk about the high-glamour winners my fund owns than the boring losers. Plus, money managers sometimes get invited to corporate events hosted by the companies whose stock they own, or analyze. Which companies are likely to have more events and better events? The highly valued popular companies, or the struggling, perhaps cash-poor unpopular ones? Finally, the glamour stocks are more likely to be part of indices such as the S&P 500 which most funds track. Stocks that do poorly tend to get kicked out.
Let’s return to the 3Com/Palm situation. We explained that because of the inability to short Palm stock, the two securities could not converge to one price. This was the idea of “limits to arbitrage.” We did not, however, discuss why the price of Palm was so high to begin with. Should it be attributed to irrational investor behavior? No. Palm was one of the most glamorous of all the glamour stocks back at the absolute height of the dot com bubble. Importantly, there was a very limited number of shares outstanding (remember than initially, 3COM only issued 5% of PALM to the public).
In fact, a small number of shares was generally true for many of the technology stocks back then. Given that lots of people wanted to talk about owning these tech stocks at cocktail parties, there was high demand, and thus high valuations. This was a fun time. Investing it tech stocks was also a hobby. There was utility to be gained beyond just the monetary amount of the trading gains. And remember that it is not irrational to think I can buy high and sell higher. That was normal. And rational. We’ll return to this point shortly when we talk about financial bubbles.
Financial bubbles and irrationality
Before leaving the topic of behavioral finance, I want to address one final misconception believed by virtually all behavioral economists not to mention financial journalists. That the existence of financial bubbles proves the irrationality of investors. While financial bubbles are not really a topic of research by Richard Thaler, they have been a major research topic of another behavioral economist, Robert Shiller, who shared the Nobel Prize in 2013.
There is no question that financial bubbles exist. To list a few, there was the famous Dutch tulip bubble in the 1600s, the stock market bubble in the 1920s that preceded the Great Depression, Japan’s stock market and real estate bubbles in the 1980s. More recently, the late 1990s dot com bubble, the mid 2000s real estate bubble and today’s bubble in nearly every financial asset. Most economists admit that financial bubbles do happen (they have no idea why) but they deny that bubbles can be identified until after they have burst. I find this view ludicrous. The whole world knew we were in a tech stock bubble in 1999. Many wrote about a real estate bubble in 2007. And many more hold the view that virtually all financial assets are in a central-bank fueled bubble today. I do believe, however, that the timing of when a bubble will burst is impossible to predict.
As I said above, the mainstream economics view is that the existence of financial bubbles is proof of irrational behavior on the part of investors. I beg to differ. For at least four reasons, the existence of financial bubbles is absolutely not evidence of irrationality.
1. Market participation is NOT optional
In 2007, at the peak of the credit boom that shortly thereafter became the 2008-2009 financial crises and the Great Recession, Chuck Price, CEO of Citigroup famously said, “When the music stops, in terms of liquidity, things will be complicated. But as long as the music is playing, you’ve got to get up and dance.” What he meant was that as a major financial institution (one of the largest in the world), Citigroup had to take the same kind of risks, and go after the same businesses as its competitors. If it didn’t, its revenue and profits would lag its peers and investors and Wall Street analysts would call for Chuck Prince’s head. Though wildly criticized for his statement, Prince was right. Citigroup had to dance.
We have already talked about the mindset of institutional money managers. It is okay to be wrong as long as everyone else is wrong. It is not okay to under-perform one’s peers, or under-perform one’s benchmark. That is to say, if the market is going up and everyone else is taking the ride, as a portfolio manager, you need to take the ride too. You cannot sit in cash or other less risky assets. You play the momentum game or you lose your job.
Even more obvious is that fact that certain institutions are essentially legally obligated to take risks. Pension funds and insurance companies, being highly regulated for instance, must have investment returns high enough to meet future liabilities. But with safer assets at historically low (5,000 year lows, that is) rates of return thanks to the monetary policies of the world’s central banks, these institutions have absolutely no choice but to buy riskier assets such as equities or high yield bonds. This is even true for individual investors saving for retirement. While not legally obligated to take risk to meet return targets like pensions, individuals too cannot afford to invest safely in today’s environment. Earning 0 to 1% in savings accounts or CDs just doesn’t make the retirement math work. They too have no choice but to take more risk.
The point I am trying to make is that in nearly all cases, market participation is not optional. Even if you think valuation metrics are high, or asset prices are expensive, you still have to be invested in those assets. And there is nothing irrational about trying to keep your job, keep your salary, keep your retirement funds growing.
2. Buy high, sell higher usually works
We’ve already talked a lot of about how momentum investing, buy high, sell higher, usually works. For at least the past three decades, this has generally been true for nearly all financial assets including stocks, bonds and real estate. While past performance is no guarantee of future success as they say, past performance is an argument for momentum investing constituting positively rational decision making.
If I see my neighbor make a fortune flipping houses, why shouldn’t I do the same? I probably consider myself to be smarter, and perhaps more highly educated. When can my neighbor do that I can’t? Sure the music might stop some day, but probably not tomorrow. But by the time it does, I’ll be rich too. To you and me, this might seem like irresponsible behavior, but is it irrational? I don’t think so, especially since it usually works. And recall that when deriving utility, money is mostly just a proxy for status. Why not take the risk if my neighbor is taking the risk? If he or she gets rich and I don’t, I’ll regret it. My social status, and hence my utility will suffer.
3. Central banks always come to the rescue
As I’ve alluded to a number of times, one vital lesson that the last several decades has taught the world is that central banks always come to the rescue of the financial system and financial markets. Even more so than just low interest rates or printing money, it is this backstop and the promise of bailouts that encourages (subsidizes) risky behavior. And this is exactly the primary cause of financial bubbles. Why not take risk if there is little downside? Take the risk and make the millions (or billions). If things go bad, the government and the central bank will clean up the mess. What’s so irrational about that?
4. Contrarian limits to arbitrage
Earlier we discussed how limits to arbitrage can lead to violations of the “law of one price” and to what economists wrongly assume to constitute irrationality in financial markets. On a much larger scale the same concept of limits to arbitrage can prevent financial bubbles from being naturally curtailed and from forming in the first place. Recall that everyone in the world knew that the price of Palm stock was too high relative to its parent 3Com. However, nobody could effectuate the arbitrage because Palm shares were impossible to short.
Similarly, as I’ve mentioned before, plenty of participants in financial markets have recognized bubbles well before they have ultimately popped. But not knowing when exactly the bubble will pop (as I also said earlier, a forecast I believe to be impossible) prevents them from shorting the market and “correcting” the bubble. In simpler terms, it is virtually impossible to be a contrarian investor in today’s market environment (and in the market environments of the past few decades).
Being contrarian risks margin calls. I can’t hold my shorts if the market continues to go up. Being contrarian risks underperforming the market. I lose my AUMs as impatient investors pull out. I risk looking like an idiot compared to all the other brilliant managers playing the momentum game and closet indexing. Bottom line is that I may know the market is in bubble territory and will some day correct, but unless I know the timing of that correction (which I cannot), it is far too risky to bet against the market, and to “fight the Fed.” No less than 3Com/Palm, this should be considered a limit to arbitrage and is our forth reason why financial bubble are not evidence of irrational behavior.
That market participation is mandatory, that momentum investing works, that central banks subsidize risk, and that being a contrarian investor is virtually impossible, are together the four reasons that give us much of an understanding of the causes of frothy markets and financial bubbles. But as we’ve noticed, all four of these stem from rational actors in financial markets making what to each actor are perfectly rational decisions.
In 1996, Federal Reserve chairman Alan Greenspan made his famous “irrational exuberance” speech, commenting on the seemingly high valuation of common stocks (they were to go much higher, rising until 2000). Greenspan might have been correct about the “exuberance” part but he was wrong about it being “irrational.” Market participants were acting rationally, maximizing their own utilities. Investors were simply reacting to the incentives laid out, most importantly, and mostly unknowingly, by the unwise risk subsidizing policies of the Federal Reserve.
Conclusion
Until the ascendance of behavioral economics, mainstream economists held the naive and erroneous belief that human beings always set out to maximize their income or wealth. And so it was said that “homo economicus” was a rational creature. But then came Richard Thaler and others who showed that this belief was indeed naive and erroneous. The models were wrong. Human beings are not always, or even typically, wealth maximizers. We are instead, well…human beings, shaped by eons of evolution and biology. And in helping to show this, Thaler and the community of behavioral economists no doubt deserve some credit.
But the behavioralists made the same two mistakes as those economists they attempted to supplant. First, they forgot to question their own assumptions. They maintained the same non-colloquial definition of rationality, and kept the same proxies for utility they inherited, without a bit of thought as to whether they made sense, or had any basis in reality. Like the economists who came before them, they demonstrated that they too have very little understanding of what factors truly drive human decision making.
Then their work escaped the chambers of academia and became popular. Their studies were easy to understand, common sensical and not too mathy. Fun even. “Irrationality” made great headlines and journalists and the mainstream media ate it up. Bestsellers were published. TED talks were given. Politicians began to listen. The cookie jars of government started to open up to the behavioral economists. Money and power! Power and money!
And naturally, with the money and the power comes the arrogance. And this is their second mistake. The mistake that since the beginning of time has been made by those smart, but not wise. The mistake that til the end of time will be made by those smart, but not wise. Humans are irrational, but we the enlightened ones have the fix! Nudge! Libertarian Paternalism! Economists to the rescue! Government to the rescue! The people must be saved from themselves!