Re: Theory vs. Data

From: Robin Hanson (
Date: Thu Jun 29 2000 - 13:59:53 MDT

Damien Sullivan wrote:
> > An earthquake or storm large enough to break water systems is no small
> thing.
> > *Lots* of bad things happen because of such an event, so the question is
> > which of those many things is as fault for disease.
>When the disease is cholera and other diseases spread by contaminated water,
>the culprit seems pretty clear.

Do you know of any study of such events which indicates that the extra deaths
that they induced were relatively concentrated in water-borne diseases?
Without such systematic evidence, all we know is that deaths go up when big
bad things happen.

(Btw, in the short term the US mortality rate goes *up* .5% for every 1%
  the unemployment rate falls. See May 2000 QJE.)

>The world had lots of diseases as bad as AIDS:
>smallpox, tuberculosis, cholera... gone now, and we live longer.
>... I see major medical events, and I see a falling death rate. It seems
>rather plausible that the stream of advances allowed the death rate to
>continue to fall,

If the only data we had was steadily falling death rates and steadily
increasing medicine, it wouldn't be inconsistent to attribute the former to
the later. But you couldn't really call that evidence for the position.

And we do have other evidence. We see that variations in medical spending
don't explain variations in mortality, and we even have a controlled
experiment on the aggregate value of medicine (the Rand HIE).

>[Regarding ]
>I also see, in the _next_ figure, causes of deaths in the US in 1900.
>... So eliminating deaths from pneumonia, tuberculosis, the water diseases,
>and diptheria, would reduce the death rate by 31%. ...
>Now, the CDC does have an interest in all this, but unless Robin wants to
>accuse them of cooking the data, this seems good evidence of the utility of
>antibiotics and sanitation, and their connection with lower mortality rates.

Yes of course, death rates from infectious diseases have come way down
since 1900. I don't have to accuse the CDC of cooking those stats to question
how much credit antibiotics and sanitation deserve for all this. In fact,
antibiotics didn't appear until after most of the drop in these diseases had
already happened, it is pretty clear that antibiotics can't take most of the

>But the claim that the increase in First and Second World life expectancy has
>no known explanation, which Robin seems to be making, definitely hits the
>bogosity filter.

Well now sonny boy, folks around these parts tend to take offense at
being called "bogus". :-)

I'm not saying that we know nothing about why life is longer. Just that
medicine, sanitation, and nutrition can't seems to take credit for the bulk
of the increase.

And if you want to stick with calling this "bogus", I think you should feel
some obligation to either look at some of the sources I've cited, or come
up with some cites of your own. I teach health economics here at GMU,
and spent two years as a health policy postdoc at UC Berkeley, so I would
hope you might give me enough benefit of the doubt to at least read a few
things about this.

Robin Hanson
Asst. Prof. Economics, George Mason University
MSN 1D3, Carow Hall, Fairfax VA 22030
703-993-2326 FAX: 703-993-2323

This archive was generated by hypermail 2b29 : Thu Jul 27 2000 - 14:14:46 MDT