Quote of the day: Richard Thaler on BP

June 15, 2010

“The spill has reduced BP’s market value by 44 percent, or about $82 billion, so it’s clear that BP had a strong economic incentive to make good contingency plans.”  Does what ‘has’ happened (present tense) really work as evidence of incentives that were ‘had’ (past tense)?

For the full article published June 11, see here.

8 Responses to “Quote of the day: Richard Thaler on BP”

  1. Will Davies Says:

    To be fair to Thaler, isn’t he making a similar point to you? He is attacking the view of agency theorists that incentives, neo-classically understood, are sufficient to manage risks of this nature.

    His argument represents the strange ambivalence of behavioural economics more generally. It won’t ditch the principles of neo-classical economics – Thaler is suggesting that in a perfect world, the fear of losing $82bn in value would have led to better risk management by BP. But it is also empirically aware of how far reality diverges from this norm.

    I guess the classic Nudger’s solution in this case would be analogous to government anti-drink driving adverts: force firms to imagine the most gruesome worst case scenario, get them to calculate what this will do to their share price, and then leave them alone to work out their risk management strategy. Homo economicus lives on, thanks to their worst nightmare…

  2. danielbeunza Says:

    I agree with Martha but disagree with Will.

    As Martha implies, Thaler makes the typical mistake of behavioral finance — assuming that probability distribution was known in advance. Taleb calls this the “ludic” fallacy, as in comparing simple setup such as games of change, i.e., dice, to complex decisions such as BP’s. And I agree with him on that one.

    Will offers a “drunken driver” or smoker-style solution. Take the worst case scenario, then impress it on the decision-maker. But again, this assumes that professional oil executives might not be paying too much attention to the nightmare, just as drivers or smokers. In this, we are back to the Black Swan.

    To me, that’s mistaken. In the case of the merger arbitrageurs that David Stark and I followed, they were keenly aware of the nightmare scenario. But gave it some thought and decided that it was not a problem. They were wrong, of course, but the point is that lack of attention was not the problem. It was something else, which we call “resonance.” They took the lack of concerns by their rivals in the market as strong reassurance, not realizing that their rivals knew as little as they did.

    Ok, so what would I offer instead. One possibility is to promote direct dialogue and conversation between the different actors involved. This is not always possible (i.e. the traders could not do it — it’d be almost a form of collusion). But it is possible for a regulator to sit people in the same table rather than sit on your Fed bank watching spreads on the computer.

  3. mahperi.com Says:

    The main task of financial management include

  4. Will Davies Says:

    Daniel: I think this depends on how you understand the ontology of an ‘incentive’. You can view an incentive as some sort of epi-phenomenal thing (I experience it in my brain), in which case it is true to say that it often only comes into existence once it is too late. In this instance, BP clearly did not ‘experience’ an incentive to prevent the spill, or else it would have done so. But that’s almost a tautology.

    Thaler is not saying this. He is not saying (as you put it) that the probability distribution was known in advance. Clearly if it had been known in advance, people would have behaved differently.

    What he’s saying is that it could have become known in advance, or at least better aproximated in advance. One technique for doing this might have involved the conversations you describe. Scenario mapping might have been another (invented by an oil company, incidentally). And economic modelling would surely have been part of a successful mix. These are all efforts to ‘tame chance’ as Hacking puts it, and I don’t see why the statistician is necessarily more deluded than the futurist and conversationalist.

    It is true that behavioural economics attributes some normative reality to mathematical probability, and then uses this to evaluate and criticise empirical decisions. But this is a bit different from saying that it treats probabilities as known in advance. Maybe this is splitting hairs though.

  5. marthapoon Says:

    They didn’t estimate the risk, they couldn’t estimate the risk, they did and they didn’t care…?

    “The problem isn’t that people have cognitive biases in assessing unlikely events. […] It’s not individual irrationality; it’s power, pure and simple. Free market economics has already whitewashed enough egregious corporate behavior. Let’s not repeat that mistake with behavioral economics.”


  6. Yuval Says:

    Great quote!
    This is a great example of having a 20:20 vision in hindsight…
    That said, the statement focuses on financial markets as another arena where environmental disasters unfold, as the dramatic drop in BP’s market value illustrates.
    And now, back to exam marking (and Martha – back to your thesis)

  7. Michael Bishop Says:

    Daniel, it seems rather odd for you to write:

    As Martha implies, Thaler makes the typical mistake of behavioral finance — assuming that probability distribution was known in advance.

    I thought, “people are bad at estimating probabilities” was one of the pillars of behavioral economics. Indeed, Thaler explicitly says so more than once in his piece.

  8. danielbeunza Says:

    Michael — point well taken. What I should have said is that Thaler assumes that the probability distribution was “knowable” in advance.

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 )

Google photo

You are commenting using your Google 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 )

Connecting to %s

%d bloggers like this: