The behavioral economics of pain

The two main lessons, as I read this paper, are a) pain is less bad when the sufferer can see the endpoint, and b) pain is less bad when the sufferer feels in control to some measure.

The concluding discussion of "happiness economics" is on the mark:

…my personal reflections are only in partial agreement with the literature on well being (see also Levav 2002).  In terms of agreement with adaptation, I find myself to be relatively happy in day-to-day life – beyond the level predicted (by others as well as by myself) for someone with this type of injury.  Mostly, this relative happiness can be attributed to the human flexibility of finding activities and outlets that can be experienced and finding in these, fulfillment, interest, and satisfaction.  For example, I found a profession that provides me with a wide-ranging flexibility in my daily life, reducing the adverse effects of my limitations on my ability.  Being able to find happiness in new ways and to adjust one’s dreams and aspirations to a new direction is clearly an important human ability that muffles the hardship of wrong turns in life circumstances.  It is possible that individuals who are injured at later stages of their lives, when they are more set in terms of their goals, have a more difficult time adjusting to such life-changing events.

However, these reflections also point to substantial disagreements with the current literature on well-being.  For example, there is no way that I can convince myself that I am as happy as I would have been without the injury.  There is not a day in which I do not feel pain, or realize the disadvantages in my situation.  Despite this daily awareness, if I had participated in a study on well-being and had been asked to rate my daily happiness on a scale from 0 (not at all happy) to 100 (extremely happy), I would have probably provided a high number, probably as high as I would have given if I had not had this injury.  Yet, such high ratings of daily happiness would have been high only relative to the top of my privately defined scale, which has been adjusted downward to accommodate the new circumstances and possibilities (Grice 1975).  Thus, while it is possible to show that ratings of happiness are not influenced much based on large life events, it is not clear that this measure reflects similar affective states.

As a mental experiment, imagine yourself in the following situation.  How you would rate your overall life satisfaction a few years after you had sustained a serious injury.  How would your ratings reflect the impact of these new circumstances?  Now imagine that you had a choice to make whether you would want this injury.  Imagine further that you were asked how much you would have paid not to have this injury.  I propose that in such cases, the ratings of overall satisfaction would not be substantially influenced by the injury, while the choice and willingness to pay would be – and to a very large degree.  Thus, while I believe that there is some adaptation and adjustment to new life circumstances, I also believe that the extent to which such adjustments can be seen as reflecting true adaptation (such as in the physiological sense of adaptation to light for example) is overstated.  Happiness can be found in many places, and individuals cannot always predict their ability to do so.  Yet, this should not undermine our understanding of horrific life events, or reduce our effort to eliminate them.

Here are Dan’s papers, and here.  Here are Dan’s riddles.


I am struck by the abstract "A larger and extremely central portion of individuals' experiences involve those that result from injury, illness, hospitalization, and pain." This statement feels almost disconnected and trite next to the deep suffering described in the paper, but I think it is true and important. In particular, it suggests that not only pain management but comfort, patient rights, and bedside manner make a big difference in a person's lifetime utility. This is a really good reason to favor spending a large share of GNP on health care and to oppose bureaucratic/single-payer forms of rationing (as opposed to free market, price rationing).

Did I mention that the hospital where my next child is scheduled to be born has high speed internet in the maternity ward? I just hope that my nursing home ends up being as nice as a 2006-era delivery room.

a) pain is less bad when the sufferer can see the endpoint, and b) pain is less bad when the sufferer feels in control to some measure

Speaking anecdotally from my recent labor experience, I'd agree. I was induced; the drug to induce contractions did so only after a large does; the pain was the worst pain I've ever had in my life. On each contraction, I could see no end to even it--they seemed to last soooo much longer than I'd ever been told in birth class (longer that I'd ever practiced breathing). After 2 hours and being told I hadn't dilated at all, it was even worse--presumably because I couldn't imagine how many more hours of suffering I had.

I ended up with a C section, after which I declined any and all pain killers repeatedly. They made me stay hooked up to a machine to dispense delivery using the button, but they did so, in turned out, incorrectly, so the machine delivered no drugs. I didn't know that at the time. During those hours, I never hit the button. I believe that just believing (even mistakenly) that I could get the medicine was enough relief that I never experienced any pain from the C section whatsoever.

I was wondering if this post got so few comments because people don't like to dwell on imagining what it's like to live with pain, but I think actually the link to Dan's riddles ended up making me forget about the content of the post the first time around.

Efficacy of torture? As a means of producing information? Or efficient in the utility-producing economic sense? It isn't. Because in the case of the former, there is no greater incentive to tell the truth as opposed to
lying. The only incentive is to tell the torturer what you think is most likely to lead the torturer to stop the torture. Or, more probably, since the torturer is likely to be coming from a place of information-asymmetry (or else why torture in the first place?) the only rational choice for the torturee is to lie since that is likely (if we imagine the common frame of torture to include a threat of death) to prolong the life of the torturee for longer than if they give the torturer what they want (i.e. the truth). Unless the torture is so bad, that the torturee wishes to die. In which case, the quality of information produced from a subject who is in the midst of that level of pain is unlikely to be of much reliable value.

But of course, if you're talking about torture in the form of producing utility for the torturer--it could be efficient so long as we imagine that the torturer is taking away a great degree of utility than the torturee. Thus, I guess we can conclude that so long as the torturer is truly, truly sadistic--torture is efficient in the pareto/kaldor-hicks sense.

Or maybe we should just conclude that arm-chair quarterbacks should keep their mouths shut when it comes to torture.

Comments for this post are closed