[OPE] Jim Devine's model of the recent economic collapse.

From: Jurriaan Bendien <adsl675281@telfort.nl>
Date: Thu Oct 15 2009 - 15:06:10 EDT

Jim Devine likes math. He takes a handful of variables which the theory or
assumption tells him are the important ones, and then he tries to show there
are some quantitative implications for the configuration of variables if
particular variables go up or down - that particular effects will
necessarily happen or be ruled out, if particular variables go up or down.
If for example you had a trillion dollars worth of credit facility and
within a few years all that disappears, what is the effect? That sort of

I like math too but my mathematical ability got frustrated at a certain
point in my life, I have to brush it up more when I get some peace of mind.
The other thing is that I usually prefer not to use math if ordinary
language will do fine to make the argument. Math can add logical rigour to
the argument, but if in fact what we are trying to measure simply cannot be
measured very accurately and precisely, then, by adding very sophisticated
mathematical pyrotechnics, we do not achieve a great deal, other than that
somebody might think "wow, these guys must be really intelligent". I do not
really see it as my job to show off whatever intelligence I have, but
instead to fathom what is true and real. For this purpose math is at best an
aid, but not some kind of miracle cure. If the conceptualization is bad,
which it often is, you can do a lot of mathematical huffing and puffing, but
the results are tenuous anyway or simply lack validity.

The main value of math and statistics is to achieve a better sense of
proportion about a problem you are facing. How big or small is the problem?
What happens if I do this or don't do that? What are the implications of
this, and what are the realistic alternatives? Then you have to relate
quantity and quality. I don't get it right all the time, I make mistakes,
but I realize the importance. I worked in the past with some topnotch math
guys, and they showed me, that math is a tool to solve certain problems, but
it doesn't solve every problem, and if you try to use math indiscriminately
to solve any problem, you just end up with a mystical consciousness, a sort
of reified rationality. A weebit of mysticism is part of everyday life,
okay, but I do not aim for a mystical consciousness, and so therefore I
don't like to overextend any intellectual tool beyond its appropriate

To arrive at a rule-of-thumb medium term forecast "approximately double the
unemployment and half the GDP growth rate" for the US, Europe and perhaps
Japan I just looked at the economic stats, the proportions, how they are
related, and compared this to what I know from past experience grappling
with a lot of time series data and the theories I am familiar with. Then I
looked at what forecasters are extrapolating using past data, and tried to
sort out the valid insight from the silly dogma. And I conclude that the
unemployment rate of the labour force is going to be durably higher,
approximately twice as high, as it was before. As the economy recovers from
recession, unemployment will decline to about that level, so if on average
you had about 4% unemployment you're going to have about an 8% unemployment
level in the future. It's just going to make things tougher for the youth
and for people who are for one reason or another in a fairly marginal
position in the labour market.

On November 23, 2008 I estimated on OPE-L that "it is not unreasonable to
suppose that US unemployment could even reach 10% of the labour force at the
end of 2009 or in 2010." That turns out to be pretty much correct. Doug
Henwood in his punditry conversation mooted the possibility of a classic
depression scenario of 25% unemployment, a sort of horrific spectre of
meltdown, but I said on his blog page no, that is way off, it is not
realistic in the immediate future. "Joblessness" on the broad measure in the
US is supposed to be 17% or about one in six working-age adults, but I was
talking the official rate of unemployment in the labour force as I said.
This is not rocket science, and in fact I wasn't the only one making that
type of forecast, all it takes is a bit of experience with time series data
and a bit of understanding of theory and economic policy. The Fed knows very
well that unemployment is a highly political matter, you have to keep it
within bounds, if you get a quarter or a third of the workforce out of work,
a lot of people are going to be thinking revolution, things become very
unstable, and an economy highly dependent on credit simply cannot tolerate a
lot of instability


ope mailing list
Received on Thu Oct 15 15:12:10 2009

This archive was generated by hypermail 2.1.8 : Sat Oct 31 2009 - 00:00:02 EDT