At 03:26 PM 5/10/2005 -0500, Dan M. wrote:
>> >Presumably, over an 80 year time period, economic cycles would get out
>> >of sync with political cycles, which come every four years *exactly*.
>>
>> False.   The Presidency does not change Party every four years.  The
>> political cycle is thus irregular.
>>
>> Moreover, since 1953 there have been nine recessions.   Yours and Dan's
>> analysis would ascribe 8 of those recessions to Republican Presidencies.
>
>Would you then, be happy with a comparison from 1920 to 1952?  the number
>of recessions slightly favor the Republicans over that time.

I think that you are responding to the point rather than the big picture.   

Honestly, I don't find any analyses of the type you are proposing to be at
all particularly interesting.   Indeed, I personally think that it is
fundamentally unsound.   In the above case, I was simply using a particular
example to demonstrate this unsoundness.   Addressing the single point,
however, does not change the fundamental unsoundness.

I've debated with myself as to whether or not you would take the above
analysis seriously, or if you are just yanking my chain by being
intentionally absurd.   I'm going to presume that you are being serious,
and so if I make a fool of myself by arguing against intentional absurdity
then so be it.....
Maybe I should try to enumerate my objections in no particular order:

1) the analysis relies upon inherently small sample sizes.   The revised
analysis you propose would have a sample size of a single political cycle

2) economic growth is much more strongly determined by exogenous factors
than the Party in the Presidency

3) the analysis makes no reference to actual policies, but instead only
refers to arbitrary Party labels

4) the economic growth in one year is correlated to the economic growth in
the previous year
Moreover, even if economic growth were completely independent of the Party
in the Presidency, if voters perceived the two to be correlated, this could
create a de facto correlation.   For example, a President who happened to
be up for re-election in times of economic growth would be more likely to
be re-elected

>Also, I can do a rigorous stochastic analysis of the year to year, two year
>to two year correlations, (and others you suggest) in order to see if your
>idea that one year's growth is strongly correlated to the previous years is
>valid.  But, I don't want to take the time to do it, if  you know you will
>dismiss  results that contradict your viewpoint out of hand

I'm very open to whatever data you care to submit.   I'll admit that since
the analsyis is being on %change, I might be surprised, and you might
conclude that they data series is independent.   Again, however, I think
that you would be missing the big picture in your zeal to respond to the
individual points.

JDG

_______________________________________________
http://www.mccmedia.com/mailman/listinfo/brin-l

Reply via email to