Geithner plan arithmetic - Paul Krugman Blog - NYTimes.com

Geithner plan arithmetic - Paul Krugman Blog - NYTimes.com

I agree with the subsidy argument made here to reject the Geithner plan but the example while at first very neat, nonetheless misses an element of investors preferences: many investors have already lost a lot and it is not unreasonable for them to demand more to risk what they've got left. Here is how it works:

Let's suppose Dr. Krugman's net worth is $100 million (including job security and reputation) and I say we flip a coin with equal probability of head and tail.

If it falls on head, Dr. Krugman wins and receives $1 billion; If it falls on tail, he loses everything he has got and pays out his net worth of $100 million. Will he take the bait?

I do not know the structure of Dr. Krugman's risk preferences but I am sure most people will not take it, even though their expected gain here would be $450 million.

This is what I call the fallacy of expectations based risk management. Many of the firms that go under in every financial crisis make the same mistake, often advised by very, very smart people. They make their investment decisions based on expectations profiles without full appraisal of the sustainability of downside scenarios that higher order metrics such as variance/volatility or above fail to capture. It seems a lot of very, very smart people out there are still making the same kind of intellectually cute but intrinsically flawed argument.


For this reason, I think the more simplistic example of my earlier post makes a more robust argument explaining the subsidy part of the plan.