NetNews Usenet Archive 1992 #27/NN_1992_27.iso/spool/ont/general/2083

Internet Message Format | 1992-11-17 | 5.6 KB

Xref: sparky ont.general:2083 tor.general:1562 mi.misc:755
Newsgroups: ont.general,tor.general,mi.misc
Path: sparky!uunet!utcsri!geac!torsqnt!tmsoft!telly!moore!bkj386!brian
From: [e-mail] (Brian Jenkins)
Subject: Re: Computers and Productivity (was free trade, memory prices, et al)
Organization: Chaos in the Basement
Date: Tue, 17 Nov 92 15:50:54 GMT
Message-ID: <[e-mail]>
References: <[e-mail]> <[e-mail]> <[e-mail]>
Lines: 102

In article <[e-mail]> [e-mail] (Marc Roussel) writes:
>In article <[e-mail]> [e-mail]
>(Brian Jenkins) writes:
>>The work we do now is exceedingly more
>>sophisticated -- and the law requires it -- than could have been done
>>40 years ago.
>
>     This is part of what I was trying to say, albeit clumsily.  I doubt
>that there are fewer actuaries around now than there were before
>computers became common.  While computers have taken over some mundane
>tasks, they have allowed the world to become more complicated in such a way that
>overall productivity has not increased as much as we might have expected
>it to.  (In many fields, productivity has not increased at all.)  The
>law requires you to do more sophisticated actuarial work because it is now
>possible for you to carry out your other duties in addition to the new
>ones.  Does this extra work add materially to the economic output of
>your firm?  Maybe, maybe not.  (I simply don't know enough about
>actuarial work to comment.  Perhaps Brian can post a rejoinder.)

Yes it has.

However, if the government decides you have to do X, and
doing X takes a lot of time, you still have to do it ---
efficiently or otherwise.  It is hard to be productive if
shut down.  I think your point is that expectations have
increased over the last 40 years but this is part of
"productivity" (it is at least when the expectation can be
picked up).

Let me try this from a different angle.  In 1973 I helped a
friend do a valve-job on his 196x Austin Mini.  The valves
were "reground" using an axe file.  The engine worked fine
(the car died when both the metal and the plywood floor
fell out).  You think you can do this on one of these
pollution controlled, full efficient, high compression
engines in car these days?  No, you use computer controlled
equipment.  Yes, the valve job is less productive, but you
may have picked it up in cost savings from the fuel (once
you factor out the job loss in the fuel industry) or in
"social" benefits (health, building maintenance, "quality of
life") from the reduction in sulphurous acid, nitrous acid,
carbonous acid (and the related -ic acids) and such --- all
of which have a non-attributable social cost (e.g. hospital
costs).

"Classical" economics was not this evolved to reflect this
stuff (maybe they needed a computer :-)).

>>From a macroeconomic point of view, productivity as defined, is masked
>>to a large degree by the amount of structural unemployment being
>>created and by other economic problems.  I also think that the shift
>>towards an information/service economy (non-"productive") from an
>>industrial ("productive")market and government manipulation via
>>monetarist policies from Keynesian have screwed up the classic
>>measurement system -- but the economists disagree.  But take compounded
>>productivity gains of about 2% p.a. in the 1960's and 1970's, and maybe
>>0.25% since 1980 and you are still talking over a 50% productivity
>>increase (about 1.4% per year).  Remove the unemployment and the other
>>"non-productive" influences and I am sure that the result is much
>>higher.
>
>     I more or less agree with this.  My concern is that we keep
>equating computer use with productivity when classical measures of
>productivity often show no (or only modest) change with increased
>computerization.  There is a danger that the credibility of the computer
>industry will be damaged with the passage of time.

Actually, the concern is more that the measurements are too
old fashioned. Economic management has changes drastically
in the recent past (thanks Tricky Dick, the ........).  We
moved from supply/demand management of Keynesian economics
to the manipulation of the money supply (usually by messing
up interest rates arbitrarily).  Classical productivity is
a Keynesian measure for a non-Keynesian world.

I think that the econometrics need revamping --- hopefully
by someone who is an economist --- which is being done
slowly.  However, newspapers still pick up the old numbers
that they are used to publishing.

Having done some of the inital OA-type work at several
companies, the classical "productivity" measures were never
used.  We looked at man-days saved and the time to pay back
the equipment costs.  There was no assumed expansion of
work or price changes, but lower costs, so there was a
productivity gain.

These days, automation results in higher quality product
for the  same price (no productivity gain) or less
expensive product of "similar" quality (as I recall a
negative productivity gain).  However the current measure
is that you are still in business the next day.

If you are going to measure what computers or anything else
do to benefit companies/society you have to choose an
appropriate measure or your credibility goes down.

--
----------------------------------------------------------------------------
from the basement of               [e-mail]
Brian Jenkins
----------------------------------------------------------------------------