[OPE-L:5878] Re: BEA and Depreciation (Jurriaan)

jurriaan bendien (Jbendien@globalxs.nl)
Sun, 21 Dec 1997 02:37:23 +0100

[ show plain text ]

Thank you for your comments, John.
> 1. If capitalists are depreciating their fixed capitals according to
> an "economic rate of depreciation", then it is not clear to me how
> far off estimates based on accounting rates such as straight-line
> or declining-balance will be.

Surely you should be able to test that by comparing the actual book values
of fixed capital stocks cited in official statistics, and the result you
obtain by using the perpetual inventory method. If I may suggest this, you
could also try checking out the appendix to Glyn, Amstrong et al.,
Capitalism Since World War 2 (1st edition) and talk with the authors about
it. They have some ideas on this problem.
> 2. From what I've looked at concerning the BEA stuff, economic and
> accounting methods are used. To get the aggregate they have to
> add apples and oranges.
I don't know what BEA is at this stage, pardon my ignorance. BEA to me at
the moment means (1) the name of an airline, (2) an abbreviation of the
name of the Queen of Holland. I assume you are creating a Marxian
econometric model. What I would say is that often capitalists do not
really know themselves how to value their fixed stocks, and they use a
mixture of valuation methods themselves. And, of course, "creative"
accounting goes on, just as creative accounting goes on with inventories to
understate profits for tax purposes. But what is the margin of error ? You
can, I would say, only find that out by comparing various different
measures. Furthermore what is the magnitude of the problem in relation to
other problems, such as inflation, or inventory valuations ? In other
words, what quantitative difference does it actually make to the trend ?
It seems to me is that the objective is to get as close as you can to "real
profits in the bank", "real taxes paid", "real depreciation written off",
and not get too much into "theoretical profits", "theoretical taxes",
"theoretical depreciation" and so forth. To get to that, you have to go
backwards and forwards between theory and the data. After all, one aim
surely is to bring the theory and the data closer together.

> 3. All of this may be no big deal if
> a. The economy grows in such a way that the rates of investment
> in the various sectors are about the same from year to year.
They are not. Theoretically, capitalist development is always uneven
development in space and time. At the simplest level (for an example) in a
sustained boom phase, investment in Department 1 (capital goods) is
initially likely to be stronger, to the point of "overheating". In a
recession it is likely to be much weaker. And, looked at historically, as
the recession bites deeper and income disparities grow, investment shifts
more to those who have the income to spend, to the "upmarket" sector, i.e.
to luxury consumption.

> b. The overall rate of growth of fixed capital is nearly the same
> as the growth of the various sectors in the economy.
This is not true either for the same reason, and you can see that easily
from the data. (A side issue here is that sometimes, through lack of
adequate data, or because of estimation procedures, the statisticians will
present figures which "level out" big fluctuations a bit. Statisticians
generally don't like big fluctuations in the figures, they distrust them
and will in many cases seek to even them out).

> However, making assumptions such that (a) and (b) hold may preclude any
> notion of the economy "coming apart" or of crises due to an over- or
> under- accumulation of capital.
Well, quite. But that "overaccumulation" (overheating) occurs can be
empirically shown. I don't think in the United States or countries like
that there is a big "problem" with "underaccumulation" except in some
"moral" sense (that they "ought" to be investing more in some area). I
think we can trust capitalist to invest where profits are to be made, and
to research the opportunities better than you and I can. Underaccumulation
would imply a lack of capital available for investment, but that isn't
really the case in the "advanced" countries. It's more likely to be a
problem in some "Third World" countries, because of low profitability,
social instability, cultural factors, social structures etc. Of course you
could also view underaccumulation as too much investment in the "wrong"
places such as luxury consumption, but again that's a moral argument.

> 4. We still face the problem of describing the behavior of the "typical"
> capitalist on the general level. That is, is he is using economic
> depreciation rather than accounting depreciation?

He will use whichever depreciation technique results in the biggest
realised profits and the lowest tax liability, while trying to present a
rosy picture to shareholders. I am not really sure what exactly you refer
to here. Statements of business activity must be related to the purpose
for which they are made, and the questions asked. Capitalists may tell the
taxman one story, the shareholders another, and the statistician yet
another story. Of course if the stories get out of kilter too much then
they lose credibility so there has to be some relationship between all the

Further, no matter
> which he uses we see that the rate of surplus value changes as
> capital ages in all cases save perhaps straight-line.

What sort of measurement of the rate of surplus-value do you have in mind
here ?

> Marxists seem to adopt the straight-line method and hope that the
> data supports this assumption. Given that few if any capitalists
> take this method seriously, the Marxian description of the accumulation
> process becomes suspect.
I think you can trust capitalists to write off as much as is in their
interests to do, and what they can get away with, operating within the laws
of the land (complex international schemes may be devised to dodge taxes,
as known). The point is to concentrate on profits actually realised, and
assets value actually written off, getting as close to that as possible.
Another thing you can do is compare the figures using the perpetual
inventory method with balance sheet data. I mean, if you get a rate of
return on the balance sheet of 12% and the result from using the perpetual
inventory method is 23% then obviously some adjustment is desirable. But I
would say it is probably impossible to get to the deepest and darkest
secrets of the "true" profit rate, you can at best identify margins of
error and show what quantitative difference, it makes if different
assumptions are adopted.

> 5. Generally, in _Capital_ all we are told by Marx is that the
> depreciation rate for the first year of an investment is, by
> assumption, 10%. Thus, he is clearly using an accounting rate
> of depreciation. In the answers Engels provides Marx, he at
> first suggests that he uses a declining-balance method and then
> follows up with an example that is straight-line. Note that
> Marx himself knew nothing of economic rate using present values
> since the method of depreciating in this fashion was unknown
> during his lifetime.
Agreed. Marx also makes some remarks on depreciation in his correspondence
but I haven't got the texts handy here. I think one should go by the
depreciation "tactics" which capitalists actually use, real capitalist
behaviour, but that is something that can be settled through investigation,
even if maybe not perfectly satisfactorily.

> I think we need to know what we are estimating before we estimate. That
> is, we need a basic model which shows the accumulation process with
> economic depreciation and that allows us to consider the cases where
> capital is over- or under- accumulating. Indeed, without this we
> may be observing a falling rate of profit simply because profits
> are treated as part of depreciation or a rising rate of profit should
> the opposite be the case. Marx himself seems to furnish few clues
> here since, as I recall in Vol. 2, he indicates that it all works
> out in the crisis.

I don't deny that you need to have a good concept of what it is that you
are trying to measure. All I am saying is (1) you will never get a perfect
measure, (2) you will never be able to test the "pure concept" adequately
because of deficiencies in the data. But I believe you can demonstrate a
falling rate of profit IRRESPECTIVE of whether you include depreciation in
profits OR NOT, simply because depreciation rates remain fairly stable from
year to year whereas observed profit volumes fluctuate much more strongly
from year to year, and from decade to decade. Just look at the data, and
you will see what I mean. What I would do is put all the
methodological/measurement problems in a list, and then look at the
official statistics, accounting practices, balance sheet data and the
taxation regime in order to arrive at a solution which approximates the
real behaviour of capitalists. You cannot do better than that !
> I have yet to see the BEA stats in terms of historic costs. I do
> understand they exist. The "real value" of fixed capital is most
> often presented in terms of current costs.

Capitalists are generally interested in what assets are worth today, the
value of assets today, not some historic value. Statistical surveys tend to
reflect this. Capitalist can be relied on to understate or overstate the
value of assets depending on the resulting tax liability, and what they can
get away with in terms of their shareholders. To find out what is more
likely to be the case you have to know more about the actual taxation
regime applied among other things. I presume valuation at current cost
refers to "what it would cost to buy the asset new today", which is not the
same thing as "todays market (sale) value of used fixed equipment".

Thus, as we attempt to describe the traverse from period to period or from
year to year,
> we are generally left without a clue as to what took place.

Why is that ? And why is that a problem ? Because you want to predict
capital stock-values in the future ?

> the meantime, capitalists themselves are not so foolish as to ignore
> the actual changes in valuation that take place as their fixed
> tangible assets age when they set up their depreciation schedules.

Agreed. But they cannot do any old thing. It has to be credible to the
taxman among other things. (Well, in New Zealand the Fletcher Challenge
corporation some years didn't pay hardly any tax, but I don't think that
was because they fiddled with depreciation figures !).
> In general, empirical studies concerning Marx's falling rate of
> profit attempt to show that it falls when inputs and outputs are
> priced according to current costs.

True (although some studies get a bit more sophisticated, for instance
Shaikh adjusts for capacity utilisation, price inflation and so on). That
will usually suffice to get a rough idea of the longer term trend, of the
real results of the accumulation process over let's say a decade. But that
is not the same as getting a good view of real profitability from year to

Given they can do so, they
> then face the problem of describing the how's and why's of
> capitalist investment decisions and are forced to argue that
> rising real wages play a role in the process of accumulation.

There is nothin unMarxist about that, surely ! To paraphrase Shaikh,
capitalists are continually engaged in a dual battle: a battle to cut costs
(including labour costs) and to increase sales. Where the Keynesians go
wrong, when they argue that "higher wages would produce more economic
growth", is (1) that you cannot force capitalists to invest where they
don't want to (such being the nature of private property) (2) capitalists
may not invest in additional production capacity, they may simply put their
prices up, (3) additional demand may not in fact substantively restore the
rate of return (although increasing profit volumes) or international
competitiveness. And, to paraphrase Mandel, the real problem for
capitalists is that an end to the recession requires both an increase in
the rate of return AND a sustained increase in aggregate demand AT THE SAME

> Of course, if real wages were then to fall, we would anticipate
> that capitalists would switch back to the old technique.

Why ? They might want to switch back to the old technique if they could,
but maybe they cannot because of competition.

It seems to me that this is an area to which Marxists ought not
to travel.

Well it depends what you mean there. And we could discuss that if you wish.