[OPE-L:3340] RE: accumulation of capital revisited

andrew kliman (Andrew_Kliman@msn.com)
Thu, 10 Oct 1996 13:01:11 -0700 (PDT)

[ show plain text ]

A response to Jerry's ope-l 3330 and Paul C's ope-l 3334.

Paul had originally written: "If v were 0 then there would be no incentive
for capitalists to accumulate constant capital. It would obviously be cheaper
to throw away their machines and have all the work done by hand. The
assumption of free labour power undermines the premise of accumulation."

I responded: "None of this is obvious to me. What about sabotage, slowdowns,
strikes, sitdowns, and other S'es? Machines are used to bring an ahuman,
'objective' discipline to the labor process, and the subjection of the worker
to this
discipline. This incentive to substitute dead labor for living is *also*
present when v = 0, but since it (this particular incentive) is independent of
wages, it is most clearly seen when v is set equal to 0."

Jerry responded by asking: "How do capitalists maintain control over
wage-workers? What is it that allows capitalists to increase the intensity of
work, substitute dead labor for living labor, etc.?" He answered these
questions, in essence, with a "cost of job loss" type argument, saying that
what disciplines workers is the threat of unemployment, but that unemployment
is only threatening because they lose their wages when they become unemployed.
Hence, if v = 0, unemployment would be no threat, and presumably firms
couldn't mechanize, automate, or robotize, because workers would walk off the
job. At least I think this was his argument.

I do not deny the disciplining effect of unemployment, although it is a
dangerous and costly strategy for capital, and not the only thing that
disciplines workers. I'll return to that point below. In any case, Jerry's
argument really reduces to the neoclassical theory of why workers receive
wages: if they are free to walk off the job, then, under any conditions in
which work involves disutility, compensation is required in order to induce
them to work. Note that this really has nothing to do with the substitution
of dead for living labor, per se, but to *any and all* working conditions
involving disutility.

Paul replied: "With v=0 you are talking about concentration camp labour being
worked to death without even being fed. The response to slowdowns and strikes
would be to shoot the wretches a day earlier. How many strikes did von Braun
have to contend with at Dora[?]"

Since both Paul and Jerry want to be very concrete and realistic, let me be so
as well: the cost of outfitting the concentration camp with guns, bullets,
fences, watchtowers, and electricity to juice up the fences; the
administrative costs of employing slave labor; and the wages, clothing,
housing, and feeding of prison guards, etc., etc. do not = 0. Nor is any of
this, even the guards' wages, variable capital. It is all constant capital,
and some of it is machines (perhaps machine guns), so my point remains true:
"Machines are used to bring an ahuman, 'objective' discipline to the labor
process, and the subjection of the worker to this discipline."

If von Braun or whomever is not willing to bear these constant capital
expenditures, then I would predict that not only will no work be done by the
slave laborers, but that both the concentration camp and the oppressors will
cease to exist. (The "day earlier" notion is irrelevant, because, to keep
producing, new wretches are needed the next day, and the next, etc. So the
constant capital laid out on means of repression must be there continually.)

Hence, the amount of labor extracted from the slaves is an increasing function
of the amount of repression, and the cost of this repression increases
monotonically with the amount, so the amount of labor extracted is an
increasing function of repression costs: L = L(R).

With this in mind, we can now analyze Paul's original claim that if v = 0,
there's no incentive to accumulate constant capital, because it would be
cheaper to throw away the machines and have all the work done by hand. Take a
simple example. To focus on technical change without having to deal also with
input substitution, I'll assume a Leontief production function for the product
of this prison --- barbed wire. Take a simple example. Assume that to
produce a given amount of barbed wire each day, the optimal amount of machines
to use is 38, each renting for $50/day, and the optimal amount of prison labor
is 500 labor-days (additional work produces no more barbed wire without extra
machines). Assume the specific form of the labor extraction function is L =
10R^(.5), implying that labor extraction increases with repression costs, but
at a decreasing rate. Hence, 500 = 10R^(.5), so that the cost of extracting
500 labor-days is R = $2500. Minimum total cost is therefore $4400, $1900 for
the machines, plus $2500 for repression, if we assume v = 0.

Now imagine, instead, that a pure labor-using process is also available. If
it requires more than 663.325 labor-days to produce the same amount of barbed
wire, it will be MORE costly than the technique above, since it will raise R
above $4400.

Finally, imagine a third Leontief technique is . To produce the same amount
of barbed wire, 80 machines are needed. If again v = 0, the total cost is
$50*80 = $4000 for the machines, plus R. This technique will be more cheaper
iff R < $400, and plugging this into the labor extraction function, we find
that the new technique cheaper iff fewer than 200 slave labor-days are needed
under the new technique to work the 80 machines. If this is the case, then
it is NOT cheaper to throw away the machines and have all the work done by
hand. On the contrary. It is cheaper to invest in MORE barbed-wire-producing
machines, and throw away or sell off some guns, guards, and slaves.

The point transfers to the case where so-called "free" labor is being
exploited and where v > 0. In essence, it is this: to get work out of
workers, costs of repression must be borne (e.g., supervisors, cameras, spies,
etc.). Hence, there are "labor costs" in addition to wages. The more work is
done by hand, the greater these costs, in the aggregate, but also per-worker,
because the machines themselves lower these costs by serving as disciplining
devices (they set the pace of work, relentlessly) and monitoring devices (if a
worker can't keep up, it soon becomes clear). Hence, it is rational for
capitalists to substitute machines for workers FOR REASONS THAT HAVE NOTHING
TO DO WITH WAGES. The reasons are fundamental indeed: there's a difference
between labor and labor-power. You can buy labor-POWER in the market with
wages, but you can't buy LABOR. It must be pumped out of the living laborer
in the production process. Some ways of doing so are more costly, some less.
The substitution of dead labor for living labor has proven to be less costly.

I mentioned above, in connection with Jerry's critique, that although
unemployment is a disciplining mechanism, it is costly and dangerous to
capital. That is because capital must also try to repress the unemployed,
keep them from rising up. Remember the Los Angeles rebellion? And what has
been the aftermath? An intensification of the efforts to build more jails and
prisons and put more cops on the street. The prison population in the U.S.
has, if I remember, tripled in the last 10 years. Prisons seem to be the one
growth industry in the U.S. And it is an industry. Those who think so-called
"free" labor is an important "feature" of capitalism should note how much
stuff is being produced by prison labor. And it isn't just food and laundry
in the prisons anymore, or license plates, but stuff like blue jeans that get
sold on the market. Plus, of course, U.S. capital is now turning welfare
mothers into slave laborers, and in New York we've got the municipal workers'
union giving into an agreement that will, in effect, turn some of its members
into overseers. Let's also not forget that this is not just slavery in
general --- both the prison population and the welfare population are very
disproportionately *Black.* So we've got urban plantations in the making.
Nationwide, they're going to start putting together a massive information
bureaucracy that will keep tabs on welfare recipients to make sure they can't
get benefits when they move to another state.

And Clinton says the era of big government is over? The role of the State as
what Marx called the "national power of capital over labor" ("Civil War in
France") has never been greater.

The prisons, the courts, the cops, the national guard, etc., are constant
capital. They are part of what the French economists (and Marx) called the
"incidental expenses" of production (_faux frais de production_), only now
they aren't so incidental.

I would have a LOT more sympathy with those who want realistic and possible
models if I heard them talking about such things. I still wouldn't agree, but
I'd be a lot more sympathetic. This is not a "dig" at anyone on the list. As
Alan, Alejandro, and Alfredo might remember (is Alejandro back on the list?
is Alfredo still on), when my one-sector, single-capital refutation of the
Okishio theorem was subjected to criticism at a conference in Amsterdam in
1994, I responded that yes, capitalism requires money and it requires
competition, but it also requires State repressive power, an
armaments-producing sector, etc., and that if they can abstract from those
things, then I can abstract from their "favorite" features. Why do the
proponents of "realism" never criticize "models" for lacking a military or
secret police, etc.? Why do we hear a lot about models needing to have at
least 2 or 3 sectors, so that exchange and money are incorporated, but not
about the need to have a sector that produces weapons? Why do we hear a lot
of talk about v needing to be greater than 0, but not about costs of
supervision and repression needing to be greater than 0.

This is not only a matter of realism. Capitalism would be *impossible*
without these things, because capitalism is not just a system of markets. It
just seems that way. But the purchase of labor-power in the market produces
nothing, no use-values, no value. Capitalism ABSOLUTELY requires the
extraction of labor from labor-power by repressive means. Despite what folks
like Roemer might want us to believe, it is absolutely impossible for markets
to substitute for this power of capital over labor "at the point of
production," as Marxists used to say. Even if workers were to contract
"voluntarily" to do more work than the equivalent they receive in wages, the
power of the State would be needed to ENFORCE those contracts, and there's
nothing voluntary about that enforcement from the workers' point of view,
absolutely nothing.

This is how capital accumulates.

Andrew Kliman