When Rationality Assumptions Break Down: Behavioral Economics and the Limits of Homo Economicus

You wake at 5:47 a.m. on Friday, September 12, 2008, in a midtown Manhattan apartment. Your phone has already received eleven messages. You silence it and make coffee before you read them, because you know what they say.

You are a mid-level fixed income manager at Lehman Brothers. Your desk holds a $47 million position in a pool of residential mortgage-backed securities, purchased eighteen months ago at par. This morning, that position is marked at roughly $28 million. You have lost, on paper, nearly $19 million.

The envelope on your actual desk, the physical one, contains a single-page summary your assistant printed last night before leaving. It lists two options.

Option A: Sell. Liquidate the position today, before the weekend. Take the $19 million loss, crystallize it in the books, and move on. The market is open. There are buyers, thin ones, but they exist. You will have to explain the loss to your portfolio committee on Monday. Your year-end bonus, which was already going to be uncomfortable, will reflect this decision permanently.

Option B: Hold. These securities are backed by real assets, real houses, real mortgages. Markets overshoot. The Federal Reserve has intervened in Bear Stearns six months ago and signaled it would not allow systemic collapse. Treasury Secretary Hank Paulson is in Washington this weekend. If the government steps in, if the panic subsides, these assets could recover to 70 or 80 cents on the dollar within twelve months. Selling now locks in a loss that might be temporary.

You sit with your coffee. The logic for holding feels almost airtight. The logic for selling also feels almost airtight. Both cannot be right.

You have until 9:30 a.m. to decide.