BP and the sociology of worth

June 18, 2010

Over at OrgTheory, my coauthor David Stark has raised a key question for economic sociologists. Based on Marion Fourcade’s research, David asks, “How would the choice of valuation method shape the damages paid by BP?” The method used for the Exxon Valdez spill, itself based on economic research on valuation, ended up being a survey. Of it, David writes:

Survey respondents were given visual and oral information about the spill and its effects, as well as informed about a Coast Guard program to prevent future spills. Basically, (and not at all unlike being asked how much you would pay for a basket of groceries) they were asked how much their household would be willing to pay for the program to be implemented. The result was $31 per median household. The economists aggregated these individual “preferences,” multiplying by the number of US household, to derive the figure denoting American’s total willingness to pay (a measure of utility loss) for the environment of Prince William Sound. That figure, in turn, was the basis for the government’s legal case and, therefore, the key determinant of the price that Exxon paid in the ultimate settlement.

David then goes on to explore how different questions might yield different valuations. He offers alternative questions, in terms of comparable expenditure on defense, or on the basis of percentage earnings for BP.

What I find interesting about David’s post is the “snatching” he accomplishes. It takes a public issue that supposedly falls under economic jurisdiction, and places it squarely within the domain of sociology. Different questions trigger different mental associations, and categorize the phenomenon in a different manner. It is, ultimately, a controversy over frames — not unlike the controversy that Raghu Garud and I studied in the case of valuing Amazon.com in the dot-com years. And, as with any such controversy, there is an element of unpredictability as to how a given question might be responded. The natural question to ask now is, who gets to decide on the question that is ultimately asked? on what grounds? and, should the public (and the company) have a say on it?

4 Responses to “BP and the sociology of worth”


  1. I’m hesitant to critique contingent valuation, knowing little about the economic/philosophical debates that led to their development, but it strikes me as interesting that they base valuation on *average* willingness-to-pay. This seems somewhat at odds with economic theory: in an auction, price (should be, according to theory) determined by the highest value any particular person is willing to pay. It also seems at odds with the traditional ideological alignment of economics toward big business. For instance, pharmaceutical companies might report a higher value for certain environmental goods (like rainforests) than the “average survey respondant” as undiscovered plants are thought to be an excellent source of future drugs.

    It’s been written elsewhere that there is a cuteness bias with contingent valuation — people prefer paying to protecting cute/fuzzy animals over ugly ones, even if the loss of an ugly animal could be ecologically catastrophic. This problem is presuambly non-trivial — it’s well known that predator-prey relationships can be characterized by significant non-linearities, so the decline of certain species can either kill off or allow some other species to over-run an ecosystem, compounding environmental damages in ways that are not obvious to a survey respondant.

    Given these pecularities, my question is: how did the economics establishment come to decide that uninformed non-experts are qualified to determine the value of environmental goods? Why not submit these surveys only to ecologists, who presumably have a greater degree of expertise about the ecological importance of a species, ecosystem, etc, or perhaps representatives from relevant industries (like pharma in the case of rainforests).

    In general, there seems to be a bias against expert opinion embedded in these techniques. This is particularly interesting when you consider that economists often insist that the discount rates used in climate models be calibrated from values used by economists, rather than allowing public input on such a tricky moral and ethical question.

  2. danielbeunza Says:

    Taylor — interesting point. i never thought that willingness to pay studies were a way to reframe the jurisdiction away from experts and open it to the larger population of the affected. The economists’ approach is that everyone always knows his or her willingness to pay. Of course, we now know this not to be be the case. As Callon would say, they need calculative devices to answer the question of willingness to pay. I can see, in the run-up to that question being asked, competing websites offering different ways to address the question of “how much?” Greenpeace, for example, could we sponsoring one calculative frame, whereas BP probably would sponsor a different one.


  3. Daniel — I didn’t mean to suggest that it was an intentional reframing by economists, only that many aspects of contingent valuation seem a bit incompatible with traditional economic orthodoxy. So there’s probably an interesting story about how these techniques came to be accepted within the discipline. I guess this was more of a reaction to the paper David posted at orgtheory — it seems to treat contingent valuation as an obvious and linear development within economics itself. But I realize now that this is quite unrelated to what you are talking about — I probably should have just posted it on the orgtheory post.

    Anyways, onto the alternative framings you and David mention. One problem I see with the specific method he mentions is that by framing the spill in terms of some other *public* expenditure, there’s a good chance that the average person’s valuation of the referenced public good itself would fail to reflect its true cost. Specifically, a space shuttle launch costs $1.5 billion, so presumably the general public should receive at least $1.5 billion in value from it. But I’m willing to bet that if we applied David’s method to space shuttle launches themselves (ask people: how many new public schools would you be willing to give up for a space shuttle launch?), we’d arrive at a value that’s far lower than its cost. This would reflect the log-rolling implicit in public finance — many things the government pays for aren’t actually valuable to the average voter, just a small subset of the public (i.e. defense contractors in Texas and California). Less cynically, I might say that the public often doesn’t know what’s good for them. Either way though, the public’s stated valuation of a space shuttle launch is unlikely to reflect its true cost.

    Therefore, one criteria you’d want this alternative framing to have is for the valuation to remain consistent and transitive when different reference items have been selected. In other words, if you find that people value 15 space shuttle launches = 1 BP oil spill, and they value 15 space shuttle launches = 50,000 new public schools, then they should also value 50,000 new public schools = 1 BP oil spill.

    I suspect consistency and transitivity is a condition that is easier to maintain with smaller, well-known, familiar goods rather than giant national projects; e.g., public schools instead of space shuttle launches. Of course, valuing BP in terms of 50,000 public schools isn’t very meaningful either.

  4. danielbeunza Says:

    I very much agree (and I think that the comment is more relevant here than at OrgTheory). Your requirement for a transitivity condition would make for a wonderful attack on contingent valuation. Transitivity is a property that, as we all know, lies at the core of the assumptions for rational choice. And I am sure that survey responses would not pass that test.

    But the core point, as you rightly point out, is how to make choice meaningful. One interesting example in this arena is the use of “key performance indicators” in the companies that provide data on company performance on environmental, social and governance. In addition to stating that these firms are polluting so many tonnes of CO2 per year, they adjust that figure relative to company sales, and provide a “CO2 intensity” measure that makes comparisons possible across different firms of the same industry.

    Once again, however, different “key performance indicators” frame the problem differently. Should you divide by sales or profits? Different numbers produce different league tables. And so it is fascinating to see how the controversy is shifting towards debating these indicators.


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

%d bloggers like this: