[OPE-L:3804] [DAVID LAIBMAN] Capital-output Ratio & OCC

Gerald Lev (glevy@pratt.edu)
Mon, 9 Dec 1996 14:13:39 -0800 (PST)

[ show plain text ]

I talked to David L about OPE-L some months ago. He declined an invitation
to participate on the list because of other responsibilities that he has,
but asked to be kept informed about our progress. Since that time I have
been forwarding the "monthly reviews" to him. When I sent him the monthly
review for November, I called his attention to the thread on "Laibman on
the output ratio and the occ." His curiosity perked, he asked for copies
of the posts from that thread which I then forwarded to him. Here is his
response./In solidarity, Jerry

---------- Forwarded message ----------
Date: Mon, 09 Dec 96 16:29:00 EST
From: David Laibman <DLaibman@brooklyn.cuny.edu>
To: glevy@pratt.edu
Subject: occ, k/y


Following is my comment on the occ/k/y discussion, for posting to
OPE-L. Thanx for calling it to my attention; let me know if any further
responses emerge!


o/^^^^^) o !
/ / /^^) /\ /^^! /^^)
o(_____/_(_ /(/ \/ !_(_ /!_
David Laibman dlaibman@brooklyn.cuny.edu

Professor Editor, Science & Society
Department of Economics John Jay College, Rm. 4331
Brooklyn College 445 West 59th Street
Brooklyn, 11210 New York, NY 10019
USA Voice/FAX: 212/246-4932
718/951-5219; -5317 scsjj@cunyvm.cuny.edu
FAX: 718/951-4867
Executive Committee/Delegate 50 Plaza Street E., #2C
Professional Staff Congress Brooklyn, NY 11238
AFT Local 2334 Voice/FAX: 718/789-9565


To: Jerry Levy, OPE-L
Fr: David Laibman
Re: Capital-output ratio and OCC


Jerry has shared with me the posts concerning my formulations
concerning K/Y and the OCC, beginning with his summary of the ar-
gument taken from chapter 5 of my Value, Technical Change and Cri-
sis (M.E. Sharpe, 1992). Following are some comments on the com-

First, there is some confusion about my "output ratio" Y/K;
this was my (possibly misguided) attempt to find a simpler termi-
nology for the chapter 5 piece, which originated in the URPE read-
er, The Imperiled Economy (we were all instructed to try to write
in simple terms). For OPE level discussion, we should use K/Y,
which I propose as quantitatively equivalent to (the most useful
definition of) the OCC.

Another general point. Jerry asks (his point (b)): "How
would you evaluate David's position in the last sentence. . .",
which reads: "There seems to be no reason why the denominator
[K/L] should necessarily rise more rapidly than the numerator
[Y/L]; the whole trend is therefore called into question." Now,
at this stage in the chapter, I am formulating the well-known ob-
jection to the view that a rising OCC emerges unproblematically
from Marx's discussion. Jerry should have made it clear that I
then go on to formulate MY OWN answer to the question, which (I
believe) demonstrates what Marx was only able to assert, namely
that -- under mostly reasonable and general conditions -- the
technical composition, k, does indeed rise faster than productivi-
ty, y. When Jerry leaves it as quoted, from the part of the chap-
ter that merely sets the stage, he gives the impression that I
converge with, say, Paul Sweezy's indeterminism concerning the
trend in the OCC. I don't.

Now, from the posts by Andrew and Allin, I pick out three
technical issues: 1) the relation of my definition to the classi-
cal Marxist categories, and the argument for its accep-
tance/rejection; 2) the problem of multiple goods and heterogene-
ity; 3) the question of the implications of my k/y argument for
the historical vs. simultaneous value determination issue. (I
leave to one side, for the moment, the empirical measurement is-
sues raised by Paul; I also do not understand Paul's point about
the level of accumulation.)

1. My argument proceeds initially in a one-good, aggregative
world. K is a stock, and Y (net output) a flow, of the same com-
modity. It is pure fixed capital: no depreciation or flow inputs
in production. Total embodied labor is therefore L, and unit val-
ue is L/Y (as Jerry's condensation of my original text makes
clear). It is then a simple matter of algebra that K/Y = C/L, or
C/(v + s). I am defining the OCC as the ratio of the stock of
constant capital (measured in embodied labor) to the flow of cur-
rent labor. These elements are intuitively (conceptually) equiva-
lent to their physical counterparts K and Y, respectively. The
definition of Q (notation for OCC) carries over easily to more
general conditions: multiple goods; capital flows as well as
stocks. In the austere, one-good, pure-stock world, the technical
composition K/L and productivity Y/L are well defined; the issues
in the theory of the trend in Q therefore stand out clearly.

Andrew cites Marx's ch. 25 distinctions among OCC, VCC, and
TCC. I, too, have read that paragraph (in fact, I probably read it
first). I think Marx is concerned to define value changes as "or-
ganic" "insofar as they mirror changes in the TCC" (quoting inex-
actly, from memory). Marx wants, I think, to isolate trends in
the VCC that are separate from changes in capacity utilization
(i.e., cycles), and changes in the rate of surplus value. What he
clearly seeks is a measure of the ratio of embodied to current
labor as determined by the normal state of production: a key indi-
cator of the level of development of the productive forces as they
evolve within capitalist production relations. That is what my Q
supplies. It is BOTH the OCC AND the VCC, because it is a) in-
variant to changes in the relative magnitudes of v and s (that is
why it has v + s, not v, in the denominator), and b) defined for
the normal flow of current labor associated with the entire capi-
tal stock (i.e., no idle machines).

Marx experimented with c/v, v/c, c/(c + v), etc., in various
parts of his writings. (This, by the way, puts the kabatch (sp?)
on efforts to claim unique "authenticity" for any one particular
formulation of the ratio.) He did not, for reasons that are more
understandable in the 19th century than they would be in the 20th,
use the stocks C and V, as in, for example, C/V. The "stock of
variable capital," however, is a very suspect concept: it either
refers to a financial "wage fund," which would not be relevant at
the level of abstraction of the general argument, or to an actual
stock of labor power, which would, in any case, be "owned" by
workers, not by capitalists. We are left with C, which (under
modern conditions) completely dominates any wage fund V on a per
worker basis in any case. The corresponding definition of the
rate of profit should be s/C (flow to stock); in principle s/(C +
V), but again V is a tricky and quantitatively vanishing concept.

Andrew asks, "How can surplus-value be part of capital?" The
simplicity of the question makes it sound didactic. Of course
surplus value is not part of capital. My definition of Q would be
more precisely called "the organic composition of capitalist pro-
duction." I believe that is what WE NEED. I am content never to
know whether Marx would agree with this formulation, were he still
around. I do not, however, thereby agree to an assumption that he
would not!

2. Heterogeneous capital stock. Of course, once K is a vec-
tor, then so is k; the technical composition of capital can no
longer be defined as a unique and unambiguous scalar magnitude.
To get it into scalar form, some constant aggregator is needed;
the most likely candidate is the labor values. It is now harder
to make the argument based on the relative magnitudes of the
changes in k and y. But, as John Ernst points out, the same ambi-
guity applies to y as to k.

More difficult is the question of valuation. I think that in
the ratio C/L, the relevant concept of C is the capital stock as
actually valued in capitalist practice; i.e., the elements of the
physical capital stocks times their prices of production. For
this purpose, the prices of production have to be actually de-
fined; this takes us (back? forward?) into the value discussion,
and I have sounded off elsewhere on my view of the continuing va-
lidity and importance of benchmark ("equilibrium") positions in
the theory of capitalist reality, vs. the perpetual (and therefore
undefinable) "non-equilibrium" view, which, I think reduces to a
variant of a crude empiricism.

The real question, I think, is this: does my analysis, in
terms of conjunctural profit rate maximization and the mechaniza-
tion function, survive transplantation from the one-good model to
a heterogeneous world? My answer: I certainly hope so! (I have
worked it through in two sectors, with one capital good -- so that
technical compositions remain well defined -- and with technical
change choice decisions being made out of equilibrium, and the
dynamics from the one-good analysis seem to remain in place; Val-
ue, Technical Change and Crisis, ch. 8.) I am reasonably sensi-
tive to the paradoxes of heterogeneity; Ed Nell and I wrote the
last-ever article in the "Cambridge controversies" debate of the
1960s and 70s (AER, 1977). We had the last word, and the
neoclassicals couldn't care less!

3. Should every element of the capital stock be entered at
historical cost rather than replacement cost? Marx certainly did-
n't think so: he was at pains to describe the "cheapening of the
elements of constant capital" by technical change, just as he was
concerned to show that the fall in the value of those elements did
not completely compensate for the rise in their mass (i.e., the
assertion, once again, that k rises faster than y, or -- the same
thing -- that k rises faster than L/Y falls).

My answer is a qualified "no."

What matters in capitalist competition is the dynamic strug-
gle to survive and expand. The capital stock that matters for the
rate of profit that matters (for future accumulation) is one
valued at its expected replacement cost. When productivity cheap-
ens the replacement for an existing machine that was purchased
earlier for more money, that machine is subject to moral deprecia-
tion (Marx's term). The POTENTIAL profit rate has RISEN, and if
you don't get it, your competitors will. This is simply an appli-
cation of the Marxist proposition that it is the social, not the
individual, situation that determines value.

Yes, of course, the capitalist who has borrowed to purchase a
machine that is now depreciating morally must repay the actual
loan, in money. His creditors will not be satisfied with a re-
duced amount of money, now that newer machines can be purchased
for less! But the ability to repay loans depends on success in
the competitive struggle; dynamic collateralization -- the power
to roll over debt, and eventually repay -- is determined by the
creditors' perceptions of the ability of a firm to compete in the
present for profits. And that means keeping up with technical
change. The specter of a firm having to repay old loans out of
earnings from production involving rapidly obsolescing capital
goods is precisely a vision of the situation facing a capitalist
who is not keeping pace: this capitalist of course faces a fall in
the profit rate that is peculiar to it. Successful capitalists do
not face THIS fall, and to base a theory of the falling rate of
profit on it is to miss the track of successful accumulation.
This is precisely what the historical-cost theorists do: they
chart the fall in the rate of profit of the marginal firms that
are heading for bankruptcy. The truly revolutionary analysis ex-
amines the successful firms, which can replace obsolescing capital
stocks with latest-vintage ones, use the power thus afforded them
to roll over or repay old loans (at full value), and still compete
for market shares and accumulation against all comers. These cap-
italists have, of course, higher profit rates than the losers.
The question is: does the process lead to a fall in THEIR (top-
dog) profit rates over time? To find out, we will have to value
their capital stocks at replacement, i.e., incorporating changes
in productivity: that valuation reveals the true expansion poten-
tial, and therefore the real positions of the capitals in the com-
petitive struggle.


Well, I hope this is not more than anyone really wanted to



David Laibman