Thursday, June 14, 2007


Sometimes I am joking when I say this, but usually it is true - Human Factors is the solution to many of the world's biggest challenges. Here are two examples I thought about on my way to work this morning, primed by NPR stories on the radio:

1. The sub-prime mortgage crisis

As usual, the proposals coming out of Congress are way off the mark. Either they will underregulate and not solve anything, or they will overregulate and keep people with mixed credit histories from getting mortgages. What they need to do is apply human factors. The regulation should not say what mortgages are allowed or not allowed (but keeping the basic framework defining what borrowers can afford). It should require lenders to inform applicants exactly what the terms and risks of the loan are and then let them choose. This requires human factors because the current financial and legal gobbledygook is incomprehensible to the average borrower. A good human factors analysis could identify some structure that would solve the problem.

"The payment will start out as $1000 per month. It will stay at this level for 2 years. After that, it can go up by as much as $250 per month every year. So in 6 years, the worst case is that you will be paying $2000 per month. Given the low level of current interest rates, this is likely. If you are unable to pay, it may be possible to renegotiate, but more likely is that the bank will foreclose on your house and you will lose it."

2. The multiple challenges of prescription drugs (costs v benefits, risk communication). Again, the government is going to regulate by determining what can be sold, where, why and for how much. Instead, they should use a market-based solution, but requiring enough information for consumers to make informed decisions.

"Nine out of 10 people who take this drug for heart disease will not die from it for at least ten years. But one person will. One out of ten people will get pancreatic cancer. Three out of ten will have trouble sleeping. If you are willing to take these risks, go ahead."

And then when approving drugs, we can do a cost/benefit analysis to estimate if the aggregate effect on health care will be positive. Then leave the individual decisions up to the people.

Econlog had a post a while ago (sorry - I couldn't find the link) suggesting special (i.e. well marked) stores that sell unapproved drugs. As long as the information is good, people should be allowed to take the risks - knowing that if something goes wrong they can't sue or use public money to pay for the consequences. Not all unapproved drugs would be there, but the ones that are either promising and just not fully tested yet, or the ones that have significant benefits, but also high risks or side effects. For example, if I have cancer and 5 years to live, I should be able to buy a drug that will extend my life by 6 months, even if it increases my risk of developing some other disease ten years down the road. Or maybe I am willing to take a 20% risk of incontinence to eliminate my allergies. It should be my choice.

Any other examples???