Valuing Life or Valuing Regulatory Budgets?

This piece in the NYT on valuing life has many interesting angles.  The writer discusses the origins of cost-benefit analysis in regulations, as well as the variation (dare I say manipulation?) of the value of life implicit in calculations by the EPA, FDA, etc.

My first reaction: These numbers are political jockeying.  If I run a regulatory agency, I want the highest number possible. 

Why?  First, I look most attractive to the public.  Who do you support, the incompetent hack who says you're worth $150,000 or me, the sophisticated thinker, who values your life at $2,000,000?  What about $10,000,000?

Second, I get the biggest budget.  The marginal cost of saving a single life goes up very quickly as the number of lives lost falls.  However, I know I cannot run a "zero risk" regulatory organization.  Everyone will see through that quickly.  (Imagine the head of the NHTSA saying "we cannot tolerate a single death on US highways."  Everyone knows that's too costly.  Traffic and the economy would grind to a halt.  Heck, you'd have the TSA running the whole country!!!)  However, if you hike up the value of the lives as you slowly nudge down the number of tolerable deaths, you will impact your agency budget very nicely.

What if The People vote with their wallets?  According to the American Council of Life Insurers, the total life insurance in force in the United States is about $18.1 trillion at year end 2009.  There are 308 million people in the United States.  So, collectively, we only insure ourselves to the tune of $58,766 per person.  Not very much. 

You could say that number is no good because not everyone buys life insurance.  Surprise:  That $18.1 trillion covers 291 million policies!  (Yes, many people have more than one policy.)

So, let's bump that number way up:

• Maybe poor people don't buy life insurance.  (Not exactly true, especially since this includes group policies, but what the heck...multiply the number by 10 because only the top 10% of wealthy people buy life insurance) Make it $587,660.
• We don't insure children.  (I certainly think children are not a positive cash value asset, and cash doesn't compensate for losses, etc.  Oh, and we already accounted for them in the "poor" adjustedment.)  Double the number again: $1,175,320.

We're still covering only a fraction of what politicians think we're worth!

This leads to my third reason the government numbers are useless: Behavioral finance tells us that we, as individuals, over estimate our risk of remote events, and under estimate our risk of the mundane.  In other words, we worry more about plane crashes than crossing against the light in New York City.  Government "value of life" calculations feed that irrationality.