Positive side to anti-LTV case
Steve.Keen at unsw.EDU.AU
Steve.Keen at unsw.EDU.AU
Fri Sep 16 22:00:21 MDT 1994
Louis Proyect recently wrote:
"One of the problems with the debate over LTV is that it is confined to
an exegesis of Marx's writings with barely any reference to history,
politics, society or any other external reality. Therefore, it has more of
the character of a philosophical discussion rather than something that
would serve to help those of us who are interested in transforming
"Steve Keene needs to speak more about the socio-economic reality that
his anti-LTV theory would serve better to explain. He has done this in
only the most cursory fashion."
Believe it or not, I couldn't agree more. The problem has been
that, since my antagonists in that debate haven't accepted that I
could have a valid foundation, I've spent all my time trying to
explain my foundations, and done nothing to explain what can be
built on top of them. That is the task I am currently engaged in--the
work I've been debating with Chris and Juan was completed years ago.
In what follows I'll try to outline what I see as the positive
aspects of my anti-LTV analysis--in other words, the shape that
marxist analysis could take with a theory of value but without the
labor theory of value (this is to distinguish it from Roemer/
Morishima style marxism, which attempts to apply conventional
economic analysis using marxian categories, such as class--which
is a concept that non-marxian economics eschews).
(1) The following is confined almost exclusively to
marxian "economics". There are many issues with respect to
philosophy, the analysis of history, etc., which could be affected
by my approach to Marx's theory of value; but for reasons of
academic specialisation, I'm focusing on the economic ones first.
If this means that I spend too much time for some people's liking
discussing such things as trade cycles, theories of price, etc.,
then I apologise; but I had to start somewhere.
(2) The following is also *very* preliminary. In my actual
research, I am to some extent starting at both ends, and
working towards the middle. That is, I have a developed theory
of value [the case I have been putting--ad nauseam -:) --in
my disputes with Chris, Jean et al]; and I am currently
working on an application of where I believe this theory
of value leads in an economic analysis sense--which is to
a particular set of post-Keynesian economic concepts, and
chaos theory. I intend filling in the middle--the arguments
which link these two extremes--as I go; and that may take
me a couple of years.
(3) There are a lot of preliminaries in the following, which
may bore some of you (especially those with a non-economics
background). There is also no claim to uniqueness in some of
the following--much of it exists in current non-mainstream
economic thought. What is unique is the theory of value, which
provides a unifying force that is lacking from non-mainstream
"Executive Summary" (for those who don't want to wade through
(a) The theory of value is a consistent alternative to the
neoclassical theory of value; something non-mainstream economics
has to date lacked;
(b) Secular and production analyses which I argue have dominated
marxian economics--the transformation problem (TP) and the tendency
for the rate of profit to fall (TRPF)--are to some extent replaced,
and to some extent complemented, by growth and cyclical issues;
(c) Several crucial dialectics become the focus of marxian analysis
of capitalism; the ev/uv dialectic as I've elucidated it earlier
as it applies to value creation is only the first of these. The
three other key ones are the realisation problem, the dialectic
of labor-power, and the dialectic of money;
(d) the monetary aspects of Marx's thought are much more important
than they have been in conventional Marxism, and these are consonant
with work already being done by several post-Keynesian economists
(e) the analysis is critical of capitalism, but supports the mixed
economy rather than socialism, and reform rather than revolution;
(f) there are aspects of the analysis which are very easily
integrated with modern theories of chaos; there is an argument that
what dialectics actually did, as developed by Hegel and Marx,
was provide a philosophical analysis of forces of social
dynamics which mathematical analysis has only just become
capable of handling.
I will approach the positive arguments sequentially (and a bit
of negativity is necessary early on, in terms of pointing out
accepted tenets of marxism that I abandon).
A: Theory of value
Since the basis of the theory of value I put forward is
dialectics, there is a greater respect for dialectical
philosophy in general in my analysis than was the norm after
Marx (the key aspects I want to focus on here are limited to
economic applications). I will draw a contrast in what follows
with conventional, "neo-classical" economics--which I'll assume
(probably in error) that most of you are not familiar with.
Neoclassical economics argues that the subjective utility
enjoyed by the consumer of a commodity plays a role in
determining the price that consumer is willing to pay for it
(Chris Sciabarra would probably want to add that this is also
the perspective of Austrian economics). The underlying idea
is that the consumer maximises the gap between the satisfaction
received from consuming a commodity and the cost of purchasing
that commodity, and that the maximisation of utility is the
object of economic society.
Their focus is exchange, and what they teach is that in any
exchange, both sides balance the ratio of the marginal (this
reason for marginal, and what it means to an economist,
is explained in the next paragraph) utility they receive
from what is exchanged with the marginal cost of producing it.
They also eliminate money from their analysis, arguing that
what is really going on is a barter of real goods. In the
case of the labor market, they are arguing that the laborer
is balancing the marginal "disutility" of work against the
marginal cost of producing him/herself, and the capitalist is
balancing his/her foregone marginal utility from the commodities
with which he buys the worker's time against the marginal cost of
producing those commodities.
[The reason for the marginal bit is that, if you draw two
curves, the biggest gap between them occurs when the tangents
to the two curves are parallel. Thus if you draw a curve to
represent the total cost of manufacturing a commodity, and
you draw a curve which represents the _quantified subjective
utility_ (see next paragraph) a commodity gives, the gap
between them is biggest where their tangents are parallel.
Since that slope corresponds to the increment to both cost
and utility of the last commodity made/bought, they call
these tangent the "marginal cost", "marginal utility"]
[This idea of quantified subjective utility is something I
intend ridiculing below; I expect that one of the many reasons
my analysis of Marx generates such hostility amongst Marxists
is that when I talk of use-value being quantitative under some
circumstances, it sounds similar to the above analysis. In fact,
it is "at right angles" to the above. To continue:]
Extended to all commodities, they teach that a free market
eventually reaches the outcome that the (marginal) cost
of producing each commodity is equated to the (marginal)
utility it generates:
--- = ---
and this means that (social) welfare is maximised at minimum
cost under capitalism.
That analysis is quite seductive to young minds when first
exposed to economics--something I know from my own initial
reaction to it as a 16 year old at school, and from having
spent many years teaching it (and trying to undermine it).
Only a small minority reject it and search out alternatives, of
which Marxism is one.
Using the exchange-value/use-value dialectic, it is possible to
put forward an (I hope) equally seductive initial perspective,
which also holds the neoclassical perspective up to easy
Capitalism is not about the maximisation of utility: capitalism
is about the accumulation of capital. Producers in capitalism
couldn't care less about the utility of the commodities they
produce: in the case of the worker, the "utility" of leisure
isn't available to him/her--since without his/her own capital,
not working means starving. In the case of the capitalist,
the utility of the commodities his/her factory produces aren't
of any interest: what is of interest is maximising the gap
between the exchange-value of the inputs, and the exchange-
value of the outputs.
How to explain that a gap occurs? This is where the exchange-
value/use-value dialectic comes in, preceded by an analysis which
argues that trade under capitalism is quite different to trade
under previous societies (whereas neoclassical economics argues
that its insights are "universal").
Under pre-capitalist society, when trade initially occurred, it
is quite feasible that the "utility" of what was exchanged affected
the price. Two cultures meeting "at the border" exchange superfluous
items which are peculiar to each culture, and the eventual price set
may well reflect subjective valuations of the goods.
However, as trade becomes routine, commodities are produced not
for the satisfaction of the individual producer, but for exchange
and profit. The cost of production becomes the dominant factor
in determining price, and the use-value of the commodity is
irrelevant. This is step one in ridiculing the neoclassical
perspective: it is really an analysis of pre-capitalist exchange.
Now we turn to looking at exchange under capitalism to explain profit.
In general--in all exchanges of non-unique objects--the exchange-
value of the object determines its price. Use-value in this analysis
is an objective thing, not a subjective valuation--the use-value of
a chair is the fact that you can sit in it, not the comfort it
provides the sitter. That means that in general, exchange-value and
use-value are incommensurable under capitalism; whereas the
neoclassical analysis makes them commensurable through the device of
The next problem is to explain how things can be bought and sold at
their exchange-values, and yet result in a profit for their capitalist
This is where conventional marxism brings in the labor exchange, and
the distinction between abstract and concrete labor. It is where I
bring in the ev/uv dialectic. Those of you who reject my analysis are
going to find the next bit the most objectionable part of my argument.
The capitalist buys inputs to production at their exchange-value; their
objective use-value is irrelevant to the price he pays for them. In
production, he exploits a commodity's objective use-value--which is
its ability to produce new commodities. Whereas in normal consumption,
use-value is qualitative, in the case of productive consumption,
use-value is quantitative. Since the exchange-value of a commodity
and its use-value are unrelated under capitalism, there will be a
difference between the two, and this difference is the source of
The easiest example to give is the labor-power exchange, where the
gap between the hours it takes to produce the subsistence commodities
needed to sustain a worker can be compared to the hours that a worker
will be required to work once he/she has sold labor-power. But in
general, the distinction can be drawn between the productive
capacity of an input to production, and its depreciation.
The final stage of pedagogy is to point out that this explains how
a capitalist can turn an initial sum of money into a larger sum of
money: it explains the circuit M-C-M+, by filling in the essential
stage of production and generation of a surplus, and it justifies
the characterisation of capitalism as a system where the accumulation
of capital is the driving force.
It also allows the final stage of ridiculing the neoclassical
perspective. The neoclassical idea--that capitalism is about the
maximisation of utility--can now be shown to concern simply the
circuit C-M-C. In fact, neoclassical analysis analyses a capitalism
without capitalists, since no-one in their model wishes to
accumulate money (capital). Nor is it possible to do so--all that
can happen is that the mass of commodities can expand, if an
individual producer decides to produce more of a given commodity
and bring it to market, where it will automatically find a buyer--
In contrast, Marx's analysis easily points out the fallacy in
Say's Law--the market consists of two classes of actors, those
in the C-M-C circuit (workers) and those in the M-C-M+ circuit.
The former must come to the market, since they only seek the
means of subsistence; the latter will only come to market if they
expect a profit. Thus capitalists are the ones who are likely to
"pull out" if they don't perceive the possibility of a profit,
and their behaviour explains the cyclical and crisis-ridden
nature of capitalism.
The theory of value I put forward reaches the conclusion
that all inputs to production can be sources of surplus value
to the capitalist. The transformation problem, and the
tendency for the rate of profit to fall, both depend on labor-
power being the only source of surplus value.
With respect to the former, since all inputs are potential
sources of surplus value, an higher organic composition
of capital carries with it no implication of a lower rate
of surplus, or of profit. There is, however, a different kind
of transformation problem, of which more later.
With respect to the TRPF, again since all inputs can be a source
of surplus value, whether there is or is not a secular trend
to increasing ratios of non-labor to labor inputs (yep, I'm
being colloquial here) carries no implication for the overall
societal rate of surplus. Questions of achieved surplus have
more to do with cyclical issues than secular ones, and with
the realisation of surplus, rather than the generation of it
in the first instance.
C.1 The realisation problem--a second dialectic between
exchange-value and use-value
The ev/uv dialectic as I've outlined it in the value debate
gives an explanation for the origins of surplus value. But this
is only the first aspect of the dialectic between exchange-
value and use-value. It is possible to find extensive references
by Marx to a second aspect: that once value and surplus value have
been generated in the factory process, the status of exchange-
value and use-value become in a sense reversed. To actually realise
the exchange-value contained in the commodities which have been
produced, those commodities must be use-values in both the
individual sense--the commodities must fulfil some need--and
collective sense--the quantity of use-values produced must
not be too great relative to aggregate demand for them. If either
of these conditions is breached, then the capitalist will fail to
realise the surplus, and a loss will be made.
The first issue concerns the qualitative aspects of commodities
produced, advertising, etc. The second concerns economic cycles in
the context of a (normally) growing economy: tranquility is a word
which can almost never be used to characterise capitalism, so the
guarantee is there that both at the individual and aggregate level,
capitalists are never going to exactly fulfil their expectations.
C.2 The dialectic of labor-power
As I argued in the value debate, in my opinion Marx derives the
result that labor-power is _a_ source of surplus by considering
the aspects that labor-power shares in common with all commodities.
But at several levels, labor-power is both a commodity and a non-
While it has been conventional Marxist practice to use the non-
commodity aspects of labor-power to explain the origin of
surplus (the abstract/concrete labor case, the Bowles and Gintis
analysis circa 1980, Wolff's analysis, etc.), I use it to
argue that, in general, the payment to workers will _exceed_ the
value of the commodity they sell. In other words, the value of
labor-power (the subsistence bundle of commodities) determines
the _minimum_ wage. Again, I can provide numerous instance of
Marx characterising the value of labor-power as "the minimum
Since labor-power is both a commodity and a non-commodity, there
is a dialectical tension which manifests itself in the organisation
of the sellers of labor-power to fight for a payment above its
value, in individual and organised resistance over working conditions,
in politics, in the myriad manifestations of humanity amidst
Since my analysis argues that the system of production in general
generates a surplus, the class struggle is thus a struggle over the
apportionment of the surplus between the capitalist and working
class, and there is no upper limit--short of the entire surplus--
to the level that workers can claim.
However, the share of surplus that workers manage to achieve will
affect the actions of capitalists, both negatively--reducing their
profits and also their desires to invest--and positively--by
shaping the distribution of income and hence the pattern of
demand. This is another foundation for cyclical analysis.
Also, the fact that a major input into production will normally be
"priced" above its value means that there is a problem in
transforming values into prices. Unlike its "technocratic"
namesake of the last century, this is a real problem, which
will manifest itself in the rate of investment, the rate of
growth, and the rate of inflation.
C.3 The dialectic of money
Commodity money can be priced at its cost of production--an issue
which Marx spent quite some time discussing, in the context of
a gold standard of exchange. But paper and credit money is something
else again. The cost of production of fiat money is negligible: less
than 1% of its cost of production. Clearly, fiat money can't be
sold for its "cost of production" exchange-value: instead, money is
the one commodity whose use-value is its exchange-value: the cost of
a $100 note is $100.
By itself, this dialectic would be passe`. The point which Marx only
partly developed, however, is that the observations which apply to
money also apply to some degree to capital assets. To repeat the
quote I made recently in reply to a posting by Juan Inigo, this
insight exists in the following comment by Marx on Ricardo in TSV
>"The compensation _given_ for the mine or quarry, is
>paid for the _value_ of the coal or stone which can be removed
>from them, and has no connection with the _original_ and
>indestructible _powers_ of the land.
>No! But there is a very significant connection with the
>"_original_" and destructible _productions_ of the soil.
>The word "_value_" here is just as ugly as the phrase
>"_repaid_ himself with a profit" was above.
>Ricardo never uses the word _value_ for utility or usefulness
>or "value in use". Does he therefore mean to say that the
>"compensation" is paid to the owner of the quarries and coalmines
>for the "_value_" the coal and stone have before they are
>removed from the quarry and the mine--in their original state?
>Then he invalidates his entire doctrine of value. Or does
>_value_ mean here, as it must do, the _possible_ use-value and
>hence the _prospective exchange_-value of coal or stone? (p. 249.)"
Summarising the above, the concept with which Marx toys is that
the landowner is paid a price related to the "_possible_ use-value
and hence the _prospective exchange_-value" of the asset. Generalising,
since assets are bought by capitalists only on the expectation of making
a profit out of them, it follows that the price of such assets is set
by their use-value, and not by their exchange-value. Capital assets
(which par-excellence are entire businesses, but the case can also be
made to apply to individual items of machinery) have the dialectic that
their price is set by their (unknowable but estimated) quantitative
use-value--the cash flow that a capitalist anticipates from an asset
(its use-value) determines its exchange-value.
This has several repercussions. It tempers the basic case I made in
the value debate (and which that debate never gave me the chance
to qualify) that all inputs to production are sources of
surplus. That proposition remains true, but the price paid for the
machinery inputs to production are in some circumstances--during
booms--going to be priced well above their cost of production
(though this will be discounted by the prevailing cost of credit),
and in other circumstances--during slumps--they will be priced well
below their cost of production. Add to this the dialectic of labor-
power with its predictions for the payment to workers being above
value and also pro-cyclical, and the stage is set for some
extremely messy dynamics.
D Monetary Aspects
These "messy dynamics", I believe, provide a very good characterisation
of the cyclical crises of capitalism, and money and accumulation are
crucial in them. These aspects of capitalism have been the subject of
study by a number of post-Keynesian economists (Minsky, Moore, Wray),
as well as Cartelier. Capitalism is cyclical, these cycles are driven
by capitalist expectations of profit, and the banking sector provides
an additional crucial component of this, both in its expansion and
contraction of credit over the cycle, and the role the accumulation
of debt plays.
This is the "end bit" of the above argument on which I am currently
working. A first paper on this, entitled "Finance and Economic
Breakdown: Modelling Minsky's Financial Instability Hypothesis" is
due to appear in the _Journal of Post Keynesian Economics_, probably
in the Winter '95 issue. An electronic version of that paper (in
postscript format) is held on the csf.colorado.edu ftp server, in
the directory econ/authors/Keen.Steve
E Reform and Revolution
An argument which has been made by Minsky, and which has been
confirmed after a fashion in my modelling, is that "big government"
makes crashes of the Great Depression kind almost impossible. To
caricature the argument, the government's deficit spending during
a slump provides capitalists with a cash flow which enables them
to eventually repay the debts they accumulated during the
preceding boom. Without that deficit spending, the debts could
remain unrepayable, leading to a debt-deflationary spiral into
"Pure" capitalism, without this big government aspect, could quite
possibly plunge into a terminal crisis--whose root cause was
cyclical, not secular. Thus the "leftist overconfidence" in capitalism
that Louis mentioned in his post is truly misplaced, because in this
analysis, the reason capitalism survives is not because of its
"undiluted" aspects, but because of its contradictions: capitalism
needs the counterbalance of a state sector to survive its own
extremes. This is, if you like, the ultimate dialectic.
Some could construe this as an anti-reform argument--"let's
destroy the welfare state to hasten the collapse of capitalism".
I don't. While older than Chris Bailey thought, I'm not old
enough to have experienced the Great Depression--and it is an
experience that I don't want to see repeated. The hardship it
caused to ordinary working people was enormous, and as the
experience in Europe (and elsewhere!) showed, such a cataclysm
is as likely to lead to fascism as it is to socialism.
I instead construe this as a reason for radicals to be
involved in reform. This is intertwined with my final point--
where I can no longer argue that I'm working simply from Marx's
If the above dialectically-based analysis captures the reality of
capitalism reasonably well--which of course I think it does--then
the question is why? What is there in dialectics which makes it
an apposite model of capitalism?
I conjecture that what Hegel and Marx captured in dialectics has
only recently been "rediscovered" by modern mathematics and
science in the theory of chaos.
The basis of this analysis is that nonlinear relationships between
entities generates complex behaviour. One of the key consequences
of nonlinear relationships is that a tiny difference in initial
conditions makes an enormous difference to the final outcome;
another is that, for that reason, it's impossible to predict what
will eventuate after any sort of cataclysmic change in such a
system. Revolution is obviously a cataclysmic change in a social
system, and there is no guarantee--as history has shown us--that
what will evolve after such a change bears any semblance to the
reasons that change was undertaken. I thus see revolution as an
inherently dangerous route to social re-organisation.
That's about as brief a summary as I can make of the positive
side to a non-LTV marxism. There could be a lot more added to
this skeleton, some of which I'm undertaking at present, and
much of which has already been done by political economists
who don't call themselves Marxist (Minsky being a prime
example), but whose work I believe follows in Marx's spirit.
More information about the Marxism