Wednesday, May 26, 2010
Confessions of a market skeptic: endowment effect edition
Posted by Derek Ambrosino at 5:05amJonathan Halket recently wrote about a variation of the endowment effect and I just wanted to throw my two cents in on the original endowment effect topic. I am not particularly well researched on the endowment effect theory and I don’t expect any of these thoughts to be groundbreaking, but I think they can be relevant to the pursuit of an improved team.
In a nutshell, the endowment effect posits that once somebody is given a commodity, he/she immediately values it more than he/she would have before acquiring it. That is to say that we’d never be able to make a deal with ourselves because we wouldn’t be willing to pay our own asking price, and we’d never accept our own offers.
First, I reject this theory on, perhaps, its most fundamental grounds. My feeling is that one’s unwillingness to sell a commodity for its market value is not compelling evidence that the commodity’s owner has any delusions about the value of the commodity. Let me support this assertion by stating two fairly incontrovertible principles.
1. In any chain of transactions from producer through ultimate consumer, profit is derived from the gap between what an item is worth to the seller and the buyer’s interpretation of that commodity’s value, either intrinsically or on a secondary market.
2. The value of all commodities is fluid and subject to context, to varying degrees.
The reluctance to swap a commodity for what one perceives to be its fair value is entirely rational in respect to the actual workings of a market, especially for an individual commodity. Remember, nobody is producing a factory of Ryan Brauns and a fantasy team does not sustain itself by commerce, but by the return on investment it makes in its players. As owners, we are not mandated to move any units, ever. The question is not what Ryan Braun is worth to me, or even by the general market consensus. The question is what any individual in your league may be willing to pay for him.
I’ll reiterate more simply. It is not your goal to determine a price point at which you can sell an inventory of Ryan Brauns. Your goal is to extract the greatest disparity of value in your favor from a single transaction.
As a part of the sneaker enthusiast community, this is a dynamic I’m very familiar with. Individuals who sell extremely rare sneakers at consignment boutiques often price them astronomically. Recently, I had read a discussion on a message board where some of the more savvy collectors were complaining that this phenomenon sets entirely false valuations of these commodities. They were talking about the endowment effect without referring to it as such. But, the fact is that without the endowment effect, there’d be almost no secondary market for collectible or limited edition goods, a category whose general qualities apply to a player like Braun. I might only be willing to pay $500 for a pair of Entourage edition Air Force Ones, but if I had a pair I wouldn’t sell them for anything below two and half times that price. Why?
One, for $500, I’d simply rather keep the kicks. (Yeah, I know this sounds pathological, but just substitute sneakers for whatever equally inane and frivolous material vice you indulge in and delude yourself into thinking is culturally superior to rubber and leather.)
Two, I don’t care about the fair market price. I only care about the “one dumb schmuck” price.
Three, the $500 doesn’t replace the sneakers it just gives me a different commodity that may be valued differently by an entirely different demographic of people. And, further, if I do want to replace the sneakers, I must combat the endowment effect again and therefore I know that I will not be able to acquire that commodity for what I perceive its actual value to be.
The lesson here is that when it comes to elective transactions, the general laws of the market aren’t necessarily material.
The first two points above relate to the first of my original two premises in fairly obvious ways. The third point relates more to the second premise.
Quite simply, there is no objective value for any player. Well, perhaps there is, in a vacuum, but those are not the values the prospective seller and buyer are considering when making the transaction. All players’ values are relative to their own teams. The best way to get a trade done is to trade surplus of one asset class to somebody who is deficient in asset class for a largely distinct asset class, of which the other owner has a surplus. Simply, you trade a player that is worth less to you, and in the context of your team, than he is to another owner for a player who is worth less to that owner than he is to you. This is hardly rocket science.
In fantasy baseball, a commodity’s value is very much contextual. When Ariely and Carmon ran their experiment to establish the gap between what a Duke student would pay to go to the Final Four versus the price at which he/she would be willing to sell that ticket, the assumption was that the students didn’t have tickets prior to being presented with the opportunity to acquire one and that they would no longer have one once they sold it. Well… that’s not exactly how the trade market works in fantasy baseball, is it?
It was found that the proposed selling price of the ticket was inordinately higher than the proposed buying price. But, how would that change if the subjects had two tickets to a game and were only asked to sell one? What if I’m only asked to sell my courtside seat, but I would still be left with a mezzanine level seat. I can only presume the gap between what you think the objective value of that seat it and what you’d be willing to sell it for would shrink considerably.
As a potential trader in a fantasy league, I’m generally only actively in the market for certain types of commodities at different times and I’m naturally more reluctant to make certain types of transactions. I feel that the endowment effect is profoundly at play only in fairly limited types of transactions. For example, if I own Alex Rios, am offered Justin Upton, and refuse, then that’s likely the Endowment Effect. These are two fairly similar commodities in the sense that they are five-tool outfielders and in this case I would be getting the better player, but subject to the endowment effect, which drowns out the fact that Rios is overproducing.
However, if you offered me something like Carl Crawford for Kendry Morales, that may just not be a great fit for my time and I might decline for that reason.
Surplus value in the form of stats beyond what it takes to win a category or quality bench players who can’t crack your starting line-up are materially worthless; I say “materially” because their value exists but in the form of depth, insurance, or flexibility. So, once again we return to the theme of absolute value being less meaningful than optimal distribution of value across asset classes.
Thus, the endowment effect is only necessarily detrimental to the extent that it prevents you from trading a player you feel like you have to trade – if you are too attached to your surplus value. And, once you recognize said asset as surplus value, chances are your attachment to said asset wanes therefore decreasing one’s sense of endowment. In the absence of this scenario, the endowment effect may be either beneficial or detrimental depending on whether you can find somebody willing to pay your price – not a market, but a single buyer.
It does bear repeating that value is only one part of the equation when making a deal; asset class often trumps small differences in absolute value. While I agree that all players have a price and that you should be open to trading any of your players at any given time, for the right offer, I think that point is actually a bit rhetorical. Your team is assembled the way it is for a reason. There’s a balance of asset classes that is important to maintain and there's risk involved in simply making any deal that nets you more value because there is no guarantee that you can swing the next trade that would be needed to restore that balance.
Savvy owners recognize when another team has an imbalance of assets and will refuse to give you fair value when they see you are backed into a corner. When making a deal the key is to determine how valuable the player you are targeting is to his team. I used to play in a league that I dominated by drafting a disproportionate number of closers and selling them off after amassing a big lead. If the league would have stonewalled me, it would have been very difficult to win. Many owners, however, focused exclusively on the value of my closers to their teams without considering what would have happened to my team had I been unable to deal those closers.
Now, from a seller’s perspective I only needed to find a few buyers at different points throughout the year, so even if the majority of the market is just waiting while I dig my own hole, all it takes it one guy to say, sure, I’ll trade you Dan Haren for Mariano Rivera for me to solidify my title.
From the buyer’s perspective, he’s getting a valuable commodity and the more the rest of the league avoids my closers, the more valuable it becomes to him to add one. But he’s only offering fair value because he believes that my commodity’s market value reflects its value to my team and that the market for my player is competitive relative to that market value. In reality, that wasn’t the case.
And, with that, I’ll pose this question to the readers: You have a deal on the table that will help your team, but it may help a team that is already ahead of you even more. Should you make this deal? Of course, this is a very general question. So, take it where you wish.
Derek Ambrosino aspires to one day, like Dan Quisenberry, find a delivery in his flaw, you can send him questions, comments, or suggestions at digglahhh AT yahoo DOT com.
This is interesting but I am not sure I see exactly how it gets at the endowment effect. Although there are (increasingly more) empirical studies calling into question the existence of such an effect. However, a significant chunk of the literature shows the effect to be true even if a person doesn’t actually own the asset, but rather just imagines what it would be like to do so. Even then, the buying price/selling price gap persists.
See, e.g., here: http://www.predictablyirrational.com/pdfs/bb.pdf