Tuesday, November 10, 2009

Convenient falsehoods

I have been reading Obama's other book "The Audacity of Hope". It is pretty heavy going and I haven't finished yet. However I have completed chapter 5 "Opportunity" which contains some policy discussion. I disagree with much of it which doesn't mean too much as Obama and I have different priorities. What is more disturbing is that Obama seems to have difficulty critically evaluating proposals which appeal to liberal prejudices but won't actually work.

Obama uncritically supports ethanol on pages 169-170. Corn based ethanol is a good example of something which is superficially appealing but doesn't actually advance environmental priorities.

On page 161 Obama claims "... Recent studies show that the single most important factor in determining a student's achievement isn't the color of his skin or where he comes from, but who the child's teacher is. ...". Obama doesn't cite these studies so I can't specifically address them but this statement is contrary to numerous other studies that show the most important factor achievement in student achievement is the characteristics of the student and the next most important is the characteristics of his classmates. The quality of his teacher (within the range commonly found in American schools) hardly matters.

Finally on page 177 in discussing health insurance Obama claims "... The bigger the pool of insured, the more the risk is spread, the more coverage provided, and the lower the cost. ...". This misunderstands how insurance works. Pooling costs spreads them more equally but it doesn't reduce them. Pooling a bunch of high risks doesn't make them low risk or low cost.

This inability to recognize well intentioned nonsense won't matter too much if Obama appoints good advisers to handle the details. And Obama would hardly be the first President to be a big picture guy. Still Obama is the President we have at the moment and I would prefer a President with a better BS detector.


  1. Re: risk pooling, maybe your unspoken assumptions differ from Obama's? If the mechanism for enlarging the pool tends to increase the proportion of low-claim enrollees (e.g. if healthier people tend to opt out more often, and the chosen mechanism makes opting out more difficult) then enlarging the pool could indeed reduce avg cost per enrollee.

    (OTOH "avg cost per person, enrolled or not" seems likely to be a better metric than "avg cost per enrollee in program x"; but w/o more context I can't be sure).

    Mention of presidential BS detection reminded me of a book I ran across recently. I think the title was "Physics for Future Presidents." I was pretty impressed by a quick skim of part of the chapter on Yucca Mountain in which the author discusses e.g. the appropriateness of the regulatory safety reqts in the context of already-existent dangers (groundwater contamination due to naturally occuring uranium). If you get a chance to look at the book, I'd be interested in your reaction.

  2. It seems that insurance companies can do very well for themselves by slicing and dicing the data on who qualifies for insurance. They do not seem inclined to spread risk around. Instead they boot out "high risk" people from their networks. For example, in Colorado we have: "chubby baby," uninsurable because he was in the top 1 percent in weight (breast fed kid), "skinny toddler", also uninsurable (too thin,) and "tiny woman" (too short, I suppose.)
    Spreading the risk around probably does make insurance more affordable for some people who are high risk but it also increases the cost for those who are at a low risk, but insurance companies would rather cherry pick their policy holders to reduce their total costs instead. This seems unfair in cases such as the above.

  3. Insurers don't always mind high risk customers as long as they can charge them appropriate rates. Consider Progressive which has been in the news recently:

    Progressive originally focused on drivers with a history of accidents and moving violations, a market shunned by many insurers. ...

    But if you require insurers to charge everybody the same rate then they have a big incentive to get rid of high risk customers as they are very unprofitable. This will inevitably cause problems in any insurance based health care system that tries to charge everybody the same rate regardless of risk.

  4. It is true, as you say, that companies can make a lot of money insuring people with problems. A finance company in NY where I once interviewed made much profit from those needing a second mortgage due to medical problems. No doubt they charged a bundle for this service, and they got your house if you couldn't pay.

    Drivers with no judgement do cause accidents and besides they may hit an innocent driver. There are also those who get sick for no apparent known reason, and those who have a high risk life syle. I would guess insurance companies though don't differentiate with a cancer diagnosis. They probably just charge both clients more in premiums.