Are You a Behavioralist or an Institutionalist?

Noah Smith’s response to my previous post on the future of behavioral economics makes an interesting point: behavioral economics is indeed somewhat more established in finance.  I think this is correct and the examples Noah cites are all worthy representatives of the sub-field of behavioral finance.

Is behavioral finance gaining ground in the wake of the financial crisis?  My own reaction to financial crisis is that macroeconomists indeed have a lot of work to do.  I don’t think simple modifications to existing DSGE models are going to do the trick.  Instead, I think that we need to improve our understanding of the architecture of financial markets by gathering data on financial flows and contracts of the key participants in these markets and hopefully getting a better sense of how these markets connect with the “real” economy.  We can then incorporate this institutional detail into quantitative DSGE models so we can be better prepared for the next crash.

An alternative reaction is to say that what matters most is the psychology of the market participants.  Perhaps asset price bubbles and speculative behavior is fundamentally tied to “sentiment” or other factors that have traditionally been in the domain of psychology.

Of course we don’t have to make a choice here.  It’s not as though learning more about the role of psychological or subjective forces in markets is a bad thing.  Even if I think that we would be best served by focusing on financial market institutions, we would certainly be made better off with a greater understanding of the behavioral tendencies of financial market participants.

Nevertheless, I think it’s telling to think introspectively about where more work is needed.  Do you think “originate to distribute” behavior is best understood as rational people reacting to poorly designed market institutions (or poor policies) or do you think that this behavior is caused by irrational over-optimism?  Do you think that when the loan markets “froze” (a terribly imprecise term incidentally) that this was a rational decision on the part of lenders who were justifiably concerned that they either wouldn’t get paid back or that they needed to retain their liquid assets rather than commit them to the markets or do you think that frozen loan markets were caused by irrational panicking?  How you answer these questions might tell you a bit about whether the future of behavioral finance is bright or dim.

Advertisements

11 thoughts on “Are You a Behavioralist or an Institutionalist?

  1. “We can then incorporate this institutional detail into quantitative DSGE models so we can be better prepared for the next crash.”

    Ehm, no. I don’t know the behavioural literature but another type of market failure, asymmetric information, can e.g. lead to credit rationing. You cannot combine a rationed loan market with a market clearing model like DSGE.

    If you assume that the world is perfect and then add some imperfections here and there you will get a biased picture. If you wanna take microfounded macro seriously you gotta think like a microeconomist, throw DSGE out of the window (no micro guys use something similar to it) and start with the respective market failures (as the micro literature does).

    Bob Solow was simply right, DSGE does not stand the smell test. It has the theoretical problems I mentioned above and it does not match the facts, not those of the Volcker disinflation or those of the Great Moderation. First time in macro history that a model which cannot even explain the past has been accepted … and now all mainstream economist think that this utter failure of a model can be amended with financial frictions or whatever.

    If it would not impact the well-being of millions I’d call it a joke.

    • Why would you think that asymmetric information can’t be used in a DSGE framework? It’s already been done.

      • Of course it can and has been used. My point is that this is methodologically dubious as the very micro literature about asymmetric information does not use general equilibrium models.

        No microeconomist uses something akin to DSGE so if you wanna become serious about microfoundations you might wanna start via pushing DSGE out of the airlock.

      • Let me state it more bluntly, if you show microeconomist mainstream macro models they laugh about them. Once macroeconomists start reading the micro literature about general equilibrium, start thinking about stuff like Sonnenschein–Mantel–Debreu and incorporate it in their GE models I will take them seriously. Until then their basic model is worthless.

        And no, relaxing one or two of the nonsense assumptions of DSGE per paper is not a worthwhile path of research. You’d have to change all the assumptions that make DSGE what it is to get something even remotely close to reality … so why start with DSGE in the first place?

      • I guess I don’t see your point. There is no substantive difference between the micro models and the macro models. Perhaps the difference is the macro guys tend to quantitatively compare the models with the data or choose realistic values for parameters that enter the models? Is that what you mean? If so, this hardly sounds like a sin.

      • “I guess I don’t see your point.”

        That much is obvious. As I already wrote, point out one microeconomist who uses something similar to DSGE and I take back my argument. Until then STARTING with a model that ignores aggregation problems and the resulting multiple equilibria problem ( Sonnenschein–Mantel–Debreu ), ignores heterogenity, uses the ridiculous notion of rational expectations (how about listening to the critique of a mainstream guy like Phelps: http://www.bloomberg.com/news/print/2013-02-11/expecting-the-unexpected-an-interview-with-edmund-phelps.html) and assumes perfect information is pointless even if you amend it with one or two imperfections.

        By the way, I am for general equilibrium analysis. Keynes wrote an imprecise but good passage about how wage flexibility would not end the underemployment equilibrium because wages are not just the price on the labour market but also influence income distribution which again influences consumption and if we assume that consumption is convex in income lower wages will indeed reduce consumption.
        I am not saying I totally buy this story (not even the modern formal version of some Post Keynesian authors) but despite being 80 years old this little vague passage is far closer to the real world than DSGE + one or two “frictions”.

      • Let’s take a specific example. The CBO recently made economic projections for two policies (the ACA and the proposed increase in the minimum wage). For those projections, the CBO used essentially supply and demand models parameterized by “reasonable” elasticities (where reasonable means roughly average elasticities from the empirical literature). This is where the employment reduction estimates come from. Do you object to this approach?

        On Phelps, it sounds like he is talking about the strong-form of rational expectations (the one used for policy changes). The weak form is the one typically used in DSGE models (with fixed policies).

  2. In both of your questions, the actors were very rational. We had a very poorly designed system, and it was poorly designed largely to a poor understanding of human behavior. Farmer has done some studies that I’ve been meaning to read regarding rational actors creating irrational outcomes, and at face value that is what happened. The exception is with the people who designed the system — there doesn’t seem to be anything rational about the design, perhaps because it is closer to anarchy than a designed system.

  3. I think behavioral economics has more to say about consumer protection and disclosure than it does about financial panics and bubbles. However it is pretty obvious that understanding consumer behavior becomes important when securities are based on the supposedly rational behavior of consumers.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s