Huge shout out to our listener Fiachra. He’s from Ireland and wrote me a great email about his change of majors from public policy to economics, and has been reading a lot about ‘behavioral economics.’ If you are a die hard GG2K fan, you may have heard our very first episode ever from a few years ago on the efficient market hypothesis. Just as a reminder, the EMF essentially says that all market information is immediately absorbed and reflected by prices. The big assumption that we make in that framework is that people and by extension, markets, are efficient. This assumption is where behavioral economics tries to wedge itself in to explain some of the anomalies that happen in markets.
So we’ve talked about economics as being a study of decisions. When we talked about utility and opportunity cost, we found out that we are constantly making trade-offs about how we spend our resources; money, free-time, etc. Behavioral economics explores the social, cognitive, and emotional factors of these decisions. And the more I read about it, it’s not necessarily about having its own methodology, but it gives examples of where the traditional model breaks down. So rather than talk about theory for the whole podcast, Fiachra sent me several studies that illustrate what we are talking about. I’m going to use two from him and one that I found on my own.
So will start with the one that I ran across when researching behavioral economics. The study was published in 1997 by Colin Camerer of Caltech and looks at how New York Taxi drivers structure their day. The cool part about looking at taxi cab drivers is that we had a bunch of drivers that all get to make the same decisions about how many hours they work each day. Their wages are ‘elastic’ in that their effective hourly wage fluctuates from day to day, depending on how many fares they get compared to how many hours they spend driving around looking for fares.
If we took the traditional view of rational decision making, in order to make the most money overall, the rational thing to do would be to work more hours on the days that were busy and they didn’t spend a lot of time looking for fares. Then on days that were slow, the rational decision would be to quit early and gain back some leisure time for themselves because it wouldn’t be worth it to sit and grind it out.
As you may have guessed, the exact opposite happens. The study found that drivers worked longer hours on the slow days and quit early on the good days. The study writers had some theories about this, but decided to ask the fleet managers, ‘Which sentence best describes how many hours cab drivers drive each day?
– Drive until they make a certain amount of money – 6
– Drive a fixed amount of hours – 5
– Drive a lot when doing well, quit early on a bad day – 1
So this appears to fly right in the face of traditional economic theory. But bringing in a well established concept from psychology that I think we’ve mentioned before helps explain. Loss aversion. Humans feel the pain of loss much more than they feel the pleasure of gain. So setting a daily earnings target ensures that they will never feel like they ‘missed out’ on income for the day. Similarly, once they hit their target on the busy days, they feel like they’ve done their jobs and can quit early, even though it doesn’t maximize their earnings. I can’t wait to have my next drunken discussion with a cabbie about this study…
Ok, I’m going to lean on Fiachra for the next two studies (And this next section is directly from him! Note the Irish flare in his writing like ‘behaviour’ and ‘queue’ :))
So last is a study that pokes a hole in the traditional thinking deals with financial incentive for performance. Being paid for performance is not a controversial concept. Every year I sit down with my manager at the beginning of the year and she tells me what my max bonus I can get is if I complete various objectives. The conventional wisdom is that if my potential bonus is bigger, I’ll work harder throughout the year to achieve it.
So this last study explores what this looks like. Does performance just increase in a straight line, is there such thing as too big of a bonus, etc. I’m not going to go too deeply into the study but will link to it on the website, but basically they gave test groups a variety of tasks that they had to get to a “very good” level of proficiency to get the max bonus. Some of these tasks were purely physical, like stacking washers in some sort of puzzle shape, and some were more cognitive, like playing a simon says game. They ran these games with a variety of ‘bonus’ levels.
Without going into too much detail, they found that there was a sweet spot for the bonuses, and if you went above that sweet spot, performance actually declined in 8 out of the 9 tasks. They actually do a good job in this study of not claiming to know ‘why’ this effect happens, just that it does. But I can think of a bunch of reasons, one in particular being that I can’t putt in golf when any money is on the line.
I certainly have some concerns with this study. Even though they come right out and address that it would probably be wrong to extrapolate the results of these insignificant tasks to the real incentives provided by institutions, I think the implication is there. But I also think that this is one of the studies that at very least makes us think a bit harder about accepting the efficiency of our financial markets. The US stock market and financial institutions are built on big bonuses, and we kind of lull ourselves into thinking that by paying big performance bonuses, our hedge fund and mutual fund managers are going to perform better.
I can’t stress enough that it’s quite a stretch to extrapolate stacking quarters and remembering numbers to performance in the financial markets, studies like this at least make us ask the question if we should really be expecting better performance the higher and higher bonuses that institutions offer.
If you get nothing else from this podcast, it’s that behavioral economics attempts to bring in psychological, social, and emotional factors to questions that traditionally are answered by arguments based on rationality. I think I’ve only scratched the surface, and I’m honestly not sure how I feel about a lot of it. But maybe we can have Fiachra on next year after he’s taken a few classes and he can let us know what behavioral economics can tell us about macroeconomics. That’s when I think it will really grow some teeth and start disrupting the status quo. Again, big thanks to Fiachra for helping me put this one together.