Rich Parkin writes,
> My (non-rhetorical) question is "what does [Wage Dependence] pick
> up that a properly specified "cost of job loss" number doesn't?"
The MAJOR difference between how Gordon, Bowles, and Schor
and I measure CJL is the calculation of the value of government
goods, services, and transfers that benefit employed workers.
*GBS (implicitly) have WD fall dramatically after the 1960s.
*In my most recent calculation of CJL, WD stays about constant
over the postwar period.
This difference in WD is key to the difference in GBS's and my
CJL series after 1970:
*they have CJL fall and stay very low after 1965.
*I have CJL fall between 1965 and 1970, but then rise
BACK UP to the high levels of CJL that existed in the
1950s/'early 1960s by the 1980s.
I recently presented at a New School seminar my most recent draft of
my CJL paper. Comments by David Gordon during the seminar
and afterwards helped me to better understand why WD, in their
approach, falls so much after 1965. I still think they have WD fall
far too much. However, I also see that my measure of WD might
also be imperfect.
Hence my paper on WD: it will help me focus attention on the narrow
question, "what government programs benefit workers and how
important were they to employed workers"? This is an interesting
question in itself, but it also is directly related to the calculation of CJL.
Eric
..
Eric Nilsson
Department of Economics
California State University
San Bernardino, CA 92407
[EMAIL PROTECTED]
