Too Few To Mention

Chatting with one of my students yesterday, I found myself at loggerheads over an appropriate description of EE-relevant regret. It seemed to me that regret happens when outcomes don't match expectations. Moreover, it's when outcomes fall short of expectations. A recent example for me was the inexplicably popular film The Hunger Games. Many of my friends that I hold in quite high esteem assured me that the books were pretty good and the press coverage of the film suggested it would be good. I was bitterly disappointed.


But did I have a legitimate complaint? I suspect the answer pivots on information asymmetries and search theory. Before investigating further, let's make one of those cute little 2x2 matrices that economists seem to love so much. The rows are for informational disadvantage (from low to high) and the columns are the costs of obtaining more information (from cheap to dear). Please excuse the typeface shift--a fixed font width is more convenient for typesetting.

                           Cost of information  
                                                 
                        |   Cheap   | Expensive |
Information |    |  Low |     1     |     2     |
Disadvantage|    | High |     3     |     4     |

Please excuse the shoddy appearance of my ad hoc illustration. I think the idea is more or less there, however. Let's consider each cell in order.

1. When there is little in the way of asymmetrical information, any mismatch between expectations and outcomes will be due to random error. If it's easy and cheap to obtain the relevant information to reduce this error, I don't reckon that most folks would feel all that much sympathy for people who feel regret. An example of this might be someone who buys a gallon of milk, takes it home and find that it's gone bad when they try it our on their Count Chocula the next morning. If they would have taken the small step of glancing at the expiration date (actually the sell-by date, but for most folks, this is interchangeable), this error might have been avoided. I would argue that many of the information-sharing advances of the past century or so have been to reduce the cost of information so as to fit more residual consumer errors in this cell.
2. If there's little to no information asymmetry and high costs of information, we're in a world of Acts of God. Bad things happen from time to time, and unfortunately, there's blessedly little anyone could have easily done about it. Random errors do happen. That's just life, people. The world, as Douglass North puts it, is non-ergodic. If you don't like it, tough beans.
3. Now we enter the realm of exploitation. If the seller knows something the buyer doesn't know, and the buyer could find out with a little extra elbow grease, then it's pretty hard to feel sorry for buyers who express regret. This is the gullible goofball who actually buys the Brooklyn Bridge. Caveat emptor, chump.
4. This cell lands firmly in non-EE territory. One party to the transaction has relevant information the other cannot easily obtain. Lemon laws exist to choke trades out of this cell.

I think it's perfectly reasonable to worry about Cell (4), but that the other sources of regret need not influence our moral intuitions over the nature of exchange. I worry that they often do, however. Every so often, we'll have a Cell (2) event, like a rash of tainted beef or some bad kush or something, and it becomes a rallying cry for intervention. A bunch of customers expressed regret... someone has to do something! Hand-wringing ensues. Consider carefully first the joint production characteristics of the regret before applying a remedy, says I.