Commitment devices: keeping temptations at bay

July 10, 2008

Please welcome a new guest blogger on SocFinance: Zsuzsanna Vargha

Economic sociologist Bruce Carruthers just reviewed Avner Offer’s book The Challenge
of Affluence
for the American Journal of Sociology.

The most interesting concept of the book is the “commitment device”. As Carruthers writes, “’Commitment devices’ make decisions tractable and help to protect individuals from various temptations, including those that produce myopic choices and loss of self-control. To give an example, a hard-drinking partygoer who gives her car keys to an abstemious friend is using a third party to precommit to not driving while drunk. At the macro level, a gold-standard system helps governments resist the temptation to inflate their currency. Sexual behavior and savings offer other indications of the extent of self-control. Commitment devices frequently utilize social structures, conventions, norms, and institutions, and they operate at both micro and macro levels. They are costly and hence not evenly distributed throughout society: evidently one of the privileges of the rich is that they can hire some self-control.

The key problem, according to Offer, is that rapidly increasing affluence can eclipse existing commitment devices and lead to irrational or bad outcomes.” This is interesting because Offer is working from a behavioral economics framework, yet the result sounds much more like Bruno Latour. I am thinking of Latour’s example of the hotel manager in his essay “Technology is Society Made Durable” (In: A sociology of monsters: essays on power, technology and domination, edited by John Law). In very simplistic terms, in Latour’s story the hotel manager is trying to get guests to leave their room keys at the hotel, instead of taking them on their outings (and losing them). The manager tries various methods of combining the key and objects, until he finds one that works: attaching a large weight to the key, which makes it uncomfortable to carry the key. This way the manager succeeded in aligning various actors for his cause: the key, the weight, the guests. The technology of the key-with-weight is thus in fact a network of human and non-human actors. Commitment devices, in this light, are technologies for non-action. Offer calls it selfcontrol.

In any case, the concept points to the other side of technology as enabling action—you also need to work hard and configure technologies to restrict action. This is not new but Offer uses it to argue that prudential action is overridden by growing material well-being. This is because commitment devices cannot make their way to all the places where temptation makes its way. Using Offer’s concept of “commitment device” would be a fruitful approach to work out the overlaps between behavioral economics and actor-network methodologies. This is all the more important since Social Studies of Finance has been inspired by not only science and technology studies but also by behavioral finance. There is a lot to be said about affluence, temptation, and commitment devices, and whether Offer is on the right track by concluding that affluence erodes prudence. I also wonder what the substance of the argument means for finance: commitment devices make prudence possible. At a minimum, the concept could be tried out on topics like risk management, accounting (mal)practice, investing versus gambling, to name a few. 

3 Responses to “Commitment devices: keeping temptations at bay”

  1. marthapoon Says:

    I’m curious Zsuzi – Is there really any such thing as a non-action? As you point out, with the keys, the person can either fail to return the keys or return them. Both seem to be actions. So in a sense then, what you’re suggesting is that materiality does not eclipse commitment devices (it does not make them less effective) so much as it can actually configure temptations. That is to say, materiality generates sets of possible actions into systems that Offer considers should really be resisted…

  2. Avner Offer Says:

    I did not invent the term ‘commitment devices’. It is current in monetary economics, in relation to the solution of time-inconsistency problems. See also

    Devices under Self-Control Problems: An Overview’, in Isabelle Brocas and Juan
    D. Carrillo (eds.), The Psychology of Economic Decisions, ii: Reasons and Choices
    (Oxford, 2004), 49–65.

  3. Zsuzsanna Says:

    I agree. Both are actions. The term non-action only makes sense when you are looking to categorize actions in order to evaluate them. So you’re right, not acting on a temptation is action–in the way that silence in a narrative is. If you’re interested in getting people to do certain actions but not others, you will categorize. What I should have said is that these technologies commit you to a certain course of action while making other actions very difficult. The problem is that this is the very definition of social life. So there has to be a more specific meaning of this concept in order to be useful. Offer recognizes that CD’s are a very broad category and they are everywhere (see pp.48-50), and differentiates among them based on their effectiveness. But Offer starts with an observation (growing wealth and declining prudence), to which his answer is that commitment devices are failing. But that doesn’t answer the question what is not a commitment device. What Martha and I would argue is that commitment devices can only be found once we define specific temptations, relative to which the devices are restrictive. Put simply, the major worry of social scientists and states is overconsumption, so that people are not able to distribute their resources over the life course.

    Professor Offer:
    Thank you for the correction and the reference, I am looking into it. From my (sociological) perspective, your use of the concept is intriguing. Economists like Brocas et al. seem to be interested in which commitment devices are used and how well they work. That is important, but the failure of CD’s to adapt cannot be understood without your addition, which is that the existing devices for self-control no longer match the new sets of available actions that we can call “temptations”.

    Does it make a difference to the argument whether people explicitly want to lock in the future and fail at it, or whether it is our social science perspective that present rewards (spending) should not be favored over future? I suspect that it affects what actions will be defined as problematic temptations to cope with. This gets back to the question of what is a commitment device–if it is a purposeful solution invented for an identified problem, there needs to be a problem.

Leave a Reply

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

You are commenting using your 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: