Tom Walker wrote:
> Okun's Law is an empirical observation that is partly tautological. It is
> "lawful" to the extent that it is tautological and "breaks down" to the
> extent that it isn't. Go figure.

These types of considerations of course explain why I put the word
"Law" in scare quotes.

It really should be called "Okun's empirical generalization that works
pretty well on average over long periods of time."

I don't know why Tom says that this "Law" is "tautological."[*] Okun's
law adds a generalization to the raw empirical data.  It's an
empirical generalization based on the application of statistical
regression to a specific country during a specific period of time. (It
differs between countries and for different eras.) It is "lawful" to
the extent that the economy is on the regression line. It "breaks
down" to the extent that random variation away from that line is
significant (or the structure of the economy changes).

The standard statistical tests are half-decent; in a simple regression
based on yearly data for 1949-2012 in the US, the t-stat on the Okun's
"Law" "constant" is way above 2 (a standard cut-off for statistical
significance). In any event, this "Law" works better than the
statement that "we don't know what the relationship is between the
rate of growth of real GDP and changes in the unemployment rate" or
"changes in unemployment rates have no relationship with real GDP
growth rates."

I'm sorry if this is pedantic. I know that Tom knows statistical
methods, but others might not.
-- 
Jim Devine /  "Reality is that which, when you stop believing in it,
doesn't go away." -- Philip K. Dick

[*]  To the Wordnik web-site, a tautology is an "empty or vacuous
statement composed of simpler statements in a fashion that makes it
logically true whether the simpler statements are factually true or
false; for example, the statement 'Either it will rain tomorrow or it
will not rain tomorrow.'" I don't see how Okun's "Law" fits this
definition.
_______________________________________________
pen-l mailing list
[email protected]
https://lists.csuchico.edu/mailman/listinfo/pen-l

Reply via email to