fredag 10 juli 2009

Eric Crampton om riskbedömning

Patri Friedman (Milton Friedmans sonson, David Friedmans son), skrev relativt nyligen:

Fatalities on motorcycles happen at about 20x the rate (per mile driven) as in automobiles. Based on the latest stats I could find, the fatality rate per 100M vehicle miles traveled was 40 in 2004 for motorcycles. Another source gives the rate at 1.7 for passenger cars.

I drive about 8K miles/yr in my car, and many more as a passenger. Perhaps 12K miles/yr total? Let's just call it 10K for simplicity. That means the fatality rate, on an annual basis, of a motorcyclist driving 10K miles / year is 10,000 * 40/100,000,000, or 0.004, or 1 in 250. So the average motorcyclist who rides 10K miles/yr has a 1 in 250 chance of dying in an accident. That's huge! (A car driver over the same distance has a 1 in 5880 chance of dying in an accident).

Men trots att argumentet baseras på korrekt statistik så missar han målet - Eric Crampton:

But those numbers don't correct for agent type. What he really needs, and what I don't think exists, is data on relative fatality rates for risk-averse drivers in both types of vehicles. I'm sure motorcycles are still riskier, but twenty-times riskier, correcting for agent type?

Specify that there's an underlying distribution of risk-aversion running from highly risk averse to highly risk/thrill seeking. And, suppose agents sort across vehicle class by underlying risk aversion. So the most risk-averse agents buy a Volvo, the median agent buys a Toyota, and the most risk-preferring agent buys a motorbike. If we then find that motorcycles have higher fatality rates than cars, I don't know what portion of the difference comes from agent heterogeneity and how much comes from motorcycles being more dangerous.

På samma sätt går riskbeteenden hand i hand även på andra fronter, vilket jag nämnt i förbifarten tidigare - statistiskt sett röker feta mer, är sämre utbildade, mindre nöjda med tillvaron, dricker mer alkohol, äter sämre och sover sämre än ickefeta. Det betyder dock inte att det går att plocka ut någon av de riskfaktorerna och applicera på individnivå.

Inga kommentarer: