RE: Some Econ Pessimism?

Crosby_M (CrosbyM@po1.cpi.bls.gov)
Mon, 17 Feb 1997 20:09:45 -0500


On Friday, February 14, 1997 6:36 PM, Michael Lorrey wrote:
<Alexander Chislenko wrote:
[Snip] 1.Let's try to compute inflation. [Snip]>

The questions and examples mix producer prices and consumer prices.
Also, he was showing the difficulty of calculating any kind of useful
time series under what is, essentially a mild singularity situation.
I don't know of any statistic methods (perhaps some form of knot
theory?) that handle singularities very well. We do have ways of
dealing with these singularities though - restructuring, resampling,
reweighting, various tradeoffs between different index functional
forms - average of ratios, geometric means, ratio of averages - but I
don't think that's what you're interested in.

[SNIP]
<Obviously you need to put a dollar sign on the intangible value of
staying alive.>

Inflation indexes, and other economic indicators, can't very well
include intangible values and still have much usefulness.
Furthermore, once you do admit such subjectivity into these things,
they're immediately subject to political hijacking. That's not to say
you wouldn't want to have some estimate of the value of staying alive
to incorporate into whatever process you use to make some decision,
you just wouldn't want to embed it in what are supposed to be fairly
objective measures of marketplace trends.

<Once again we are dealing with material goods rather than
informational goods, so its still out of the issue. Back to megs and
bauds, you use Moore's laws [Snip] you see that there is significant
deflation in high technolgies due to technological advancement, and
the synergies of feeding every new gain into attaining greater gains,
maintaining a 100% rate increase in productivity every 18 months.>

Mike, you're really talking about productivity here. I agree, the
national productivity measures we have are ridiculous, totally
unsuited for measuring service-sector productivity, let alone the even
more abstract knowledge industries. The following is from a Wall
Street Journal article of August 12, 1992 called 'Dubious Figures:
Productivity Statistics For the Service Sector May Understate gains":

"Explaining why productivity in many services is 'simply impossible to
measure', Michael Harper, chief of productivity research at the Bureau
of Labor Statistics, cites the insurance industry. 'Do higher
premiums, which tend to boost an insurance company's revenues, also
serve to boost its productivity,' he asks. 'Or are higher premiums
merely a signal of higher policy risks.'"

Note how important insurance services are to the type of libertarian
society we hope to create. The WSJ article also goes on to note:

"Other services whose productivity the BLS finds impossible to measure
include such major fields as health care, real estate, and stock and
bond brokerage. In all, services for which no productivity figures are
available employ nearly 70% of the people with service jobs - far
higher than the comparable percentage in manufacturing."

Other good example of productivity paradoxes include schools and
day-care centers where an increase in students per teacher looks like
a productivity increase but often results in a quality decrease.

<Given this unaccounted for deflation [in cost per baud], it is easy
to see why the only thing contributing to inflation is COLA payments
by the government union and pension system that are based on false
inflation data from the government.>

COLA payments for government-employee pensions are miniscule compared
to the COLA payments to Social Security recipients in the general
public.

Also, maybe you didn't check out the cartoon URL I provided but the
point is that greater baud rates don't yet have much effect on the
overall cost of living for most people. According to the BLS Consumer
Expenditure Survey (for 1991 - the quickest numbers I can get my hands
on), average expenditures (based on a weighted sample by age-group,
income-group and area) for the typical consumer are:

14.1% food
31.2% housing
5.9% apparel and services
17.4% transportation
5.2% health care
5.0% entertainment
9.4% personal insurances and pensions
11.5% other expenditures

<Are your superiors so incredulous or adamant against acknowledging
the validity of this deflation due to the fact that they are trying to
protect their pensions?>

The inflation you identified has to do with producer prices, which
don't always translate into changes in consumer prices, nor do they
have anything to do with the bulk of the average person's monthly
bill. As I noted, the biggest constituency against adjusting
inflation measures downward are the elderly in general who want their
Social Security COLAs.

<If it is acknowledged that there really isnt as much inflation as
they claim, then the government will not be able to jack up taxes>

The second biggest constituency (in terms of how important it is to
them), is the middle class, because the CPI is used to adjust their
tax brackets - the higher the rate of inflation, the lower the tax
bracket you fall into if your income is unchanged.

You are right that technological advances have had huge unmeasured
impacts on manufacturing and service productivity. So why are costs
still increasing, even if at a lower rate than the official figures?
Well, I suspect that businesses like to keep their productivity
improvements as proprietary information and, in fact, the currently
'low' rate of inflation is a proxy for the fact that businesses don't
have to raise their prices so much because of the productivity
improvements they've been able to make in recent years.

Mark Crosby
(not a government employee; a government contractor)