The value of endowment

Well, we don’t run a requests service, but never let it be said that we don’t try to give our readers what they want (or what we want them to want, anyway). The Standard asks us to have a chat about this piece in The Economist which discusses endowment effects, so here goes.

The first question to answer is probably, “Endowment effects?! Wot’s that then?” Well, endowment effects were noticed by Richard Thaler back in the day (which just means sometime before we were born). He saw that people value things more when they own them. So, essentially, you’ll want more money to sell something on Trademe than you’d have paid for it yourself. Our favourite behavioural econ-psychologists, Daniel Kahneman and Amos Tversky, incorporated this observation into their Nobel prize winning prospect theory. They called it loss aversion with reference dependent preferences; you can see it represented by the ‘kink’ in their prospect theory value function at the reference point.

The Economist’s article suggests some nice evolutionary reasons why this effect might exist but, if you’re like me, you’ll be thinking “hang on a tick; doesn’t that mean people constantly pass up opportunities to make a monetary profit?” And, yes, it does. John List thought the same thing so, being a genius experimental economist, he went out into the real world aandanalysed some real markets. He found strong evidence that the endowment effect exists, but also that it diminishes as people gain more trading experience. So, we do probably have a silly, inefficient evolutionary hangover of a heuristic for valuing things but we can learn to overcome it with practise.

Finally, The Standard asks why neo-classical economists ignore all of this stuff all the time. The simple answer is that they’re aware of it, but you can’t save the world in one easy step. Empirical results like this one give us individual heuristics but they don’t give us a decision making framework that can be generally applied. Plenty of people study these heuristics but, until they are combined into a cohesive framework that theoreticians and econometricians can easily work with, they won’t have a broader impact in the profession.

  • Post number 100 for rauparaha – congrats mate.

    I’ll read it tomorrow when I’m not trying to clean up forecasts 😉

  • When its comes to the behavioural approach to economics I feel that Steven D. Levitt and John List made a good point in a short commentary they wrote for a recent issue of the Science journal. They see positives, as most people do, in the behavioural approach but they still remain somewhat skeptical. They write,

    Perhaps the greatest challenge facing behavioral economics is demonstrating its applicability in the real world. In nearly every instance, the strongest empirical evidence in favor of behavioral anomalies emerges from the lab. Yet, there are many reasons to suspect that these laboratory findings might fail to generalize to real markets. We have recently discussed several factors, ranging from the properties of the situation — such as the nature and extent of scrutiny — to individual expectations and the type of actor involved. For example, the competitive nature of markets encourages individualistic behavior and selects for participants with those tendencies. Compared to lab behavior, therefore, the combination of market forces and experience might lessen the importance of these qualities in everyday markets.

    I think this is the main reason most economists have yet to get too excited about it.

  • I have certainly heard that concern raised before, Paul, and I find it a little baffling. Of course, it’s likely that ‘real’ behaviour doesn’t replicate lab behaviour; however, we’re not comparing models of real behaviour with models of lab behaviour in economic theory. We’re comparing models based only on normative assumptions and mathematical conveniences to lab behaviour. I imagine that lab behaviour is a far better approximation of real behaviour than rational (in the textbook sense) utility maximisation.

    That the strongest empirical evidence emerges from controlled lab experiments shouldn’t really be a surprise either. It’s obviously easier to find evidence of something when you hold other things constant than when you look at the real world. I don’t see why these factors should hold us back from trying to incorporate behavioural results into our analysis.

  • Rauparaha: the Levitt/List results suggest that what we’re capturing in the lab experiments isn’t what we’re trying to capture.

    Friedman talked about “as if” rationality. People behave as if they’re rational on average in aggregate outcomes. So, our math models work. But expecting any individual to conform identically to the math model when we put him in the lab and abstract away from all of the context and heuristics that allow him to behave the way he normally does is too much.

    So, for example, our models of perfect competition have firms picking the point where MC cuts AC from below. I’d be hard-pressed to think of a firm that actually does that on purpose. Instead, they respond to all of the cues in a competitive marketplace that force them to come pretty close to behaving that way, or to go out of business. You can’t then put a businessman in the lab, give him a math problem, find that he can’t solve the optimization problem and then declare that therefore our models of competition are wrong.

    The Austrians talk a lot about process and about the path to equilibrium. They’re probably there onto a decent explanation of why the lab results don’t always conform to what we’d expect, but why things work out once we move to the field experiments where folks can rely on the contextual knowledge they’ve built up over time.

  • “Compared to lab behavior, therefore, the combination of market forces and experience might lessen the importance of these qualities in everyday markets.”

    While that might be fine for ignoring lab results, if we move back to the endowment effect, wasn’t it shown to exist in actual markets – as a result, this shouldn’t really hold as a critique of ignoring it.

    I think that fundamentally everyone is right here. The endowment effect is an explaination for why we may have slow clearing price in markets where the trade occur more rarely (eg the housing market).

    It is an example of bounded rationality that still fits well within our “neo-classical” methodology, and as a result is often “implicitly” appealed to be economists.

    “People behave as if they’re rational on average in aggregate outcomes”

    Moving forward to the importance of labs discussion, I would posit that many economists actually treat the rationality of economic agents as a lot more specific than an “average in aggregate outcomes”. Industrial and micro economists look at far smaller sets of markets and make use of just as many (if not more) rationality assumptions than macro-economists. As a result, labs do give us information on this individual behaviour.

    However, my issue with labs is that the results we find there should not be directly compared to real world results, as questions of the stake involved and the “artificial nature” of the decision process will both bias results.

  • CPW

    This was an interesting challenge to the experimental results, I’m not certain how or if it has held up though.

    Has anyone attempted to explain the mechanism behind the endowment effect – do people underestimate the value of goods they don’t own, or overestimate the value of goods they do own? Introspection suggests the latter to me. Does that List paper suggest which way the prices converge over time (kinda busy sorry)?

  • Steve

    Here’s another interesting article on endowment effects, and comes with a link to a great you tube video.

  • Apologies about the request. But I thought it looked like the type of question you’d be interesting in exploring here. Certainly more in your line than The Standard.

    It is good to know that this has been considered in the world of economics, even if it hasn’t made it into a operational framework yet.


  • “Apologies about the request.”

    We were just playing – we love getting requests 🙂