Full Terms & Conditions of access and use can be found at
https://www.tandfonline.com/action/journalInformation?journalCode=rcso20
Critique
Journal of Socialist Theory
ISSN: (Print) (Online) Journal homepage: https://www.tandfonline.com/loi/rcso20
The Transformation Problem: Labor Theory or
Command Theory of Value?
Barry Finger
To cite this article: Barry Finger (2021) The Transformation Problem: Labor Theory or Command
Theory of Value?, Critique, 49:3-4, 233-255, DOI: 10.1080/03017605.2021.2000755
To link to this article: https://doi.org/10.1080/03017605.2021.2000755
Published online: 14 Feb 2022.
Submit your article to this journal
View related articles
View Crossmark data

The Transformation Problem: Labor
Theory or Command Theory of Value?
Barry Finger
Values and prices are two dimensionally different measurements of the weights of
commodities in exchange: one expressed in units of time, the other in money.
Between the two there is a factor of proportionality such that the annual sum of
activated productive labor time is re-expressed as a denite quantity of monetary
output. This is maintained both at the level of individual commodity units in their
exchange with money and for the totality of commodity units turned over in a given
period of time. This factor of proportionality obviously uctuates over time. The
exchange of commodities at prices proportional to values does not also imply that
the sum of values should ever be equal to the sum of prices, nor the sum of surplus
values be equal to the sum of prots. Yet Marx, in Capital III, nevertheless
elaborates a specic analytical break between the conditions of exchange that
dominate simple commodity production and capitalist commodity production, but
claims to have reconciled the two by invoking these criteria. He does so by
constructing an alternative system to account for the average rate of prot in which
the value of capitalist commodities are represented, not by their embodied labor
content, but by the labor content of the money commodity against which they are
exchanged prior to becoming future components in the circuit of productive capital.
This paper offers an alternative solution to the transformation problem and
incorporates the average rate of prot within the pricing framework of Capital I. It
seeks to show why other attempts, both hostile and friendly, to resolve the issue
along the proposed lines of Capital III necessarily replace the labor theory of value
with a command theory of value. In this, Marx failed to transcend the classical
dilemma and became, instead, a prisoner to it.
Keywords: Value; price; prot; surplus value; rate of prot; transformation problem
Why revisit the relationship of values to prices in Marxs system? Hasnt this been
turned over endlessly in the past 150 years? What more can be said? Havent Marxists
and those interested in Marxian political economy, as allies and/or critics, chosen
Critique, 2021
Vol. 49, Nos. 34, 233255, https://doi.org/10.1080/03017605.2021.2000755
© 2022 Critique

sides? Marx is both wrong and his framework self-contradictory, if not useless as an
explanation of prices, regardless of the other valuable insights he might otherwise
offer; or Marx is correct but misunderstood and the record needs to be set straight.
What is proposed here is a third alternative: that Marxs approach to the pricing
question is wrong for having abandoned a consistent approach to the labor theory
of value and that it can be corrected and the labor theory of value salvaged only by
jettisoning the errant remnants of classical economic theory that distorted Marxs
analysis.
I. There is no transformation problem: the source of Marxs analytical error
The crux of the transformation problem is this. In volume 1 of Capital Marx
addresses the problem of capitalist production as a whole; in the 3rd volume, he
approaches the surface phenomena of market relations between the various units
of capital, whose interactions reproduce the conclusions previously derived.
Capital as a whole deals with aggregatestotal capital divided functionally into con-
stant capital and variable capital and total surplus value. Marx is not interested there
in individual prices or in the secondary deductions from surplus value in the form of
merchant capital, interest and rent. The rate of prot is measured against the aggre-
gate investment, constant capital (assuming there is no stock of variable capital), the
rate of exploitation is measured against the aggregate wage bill of productive workers
(variable capital). Since the productive economy as a whole is the subject of consider-
ation at this point, there is only one aggregate rate of prot and one aggregate rate of
exploitation.
In the third volume, the many rates of prot of the individual capitals in compe-
tition tends towards convergencetowards a uniform, average rate; as Marx, in
effect, posits when treating capital as a whole in volume I.
The question then arises. If the ratio of capital invested per productive worker, the
organic composition of capital, varies from sector to sector, how does a uniform rate
of prot emerge? If the competitive process yields a uniform rate of exploitation,
there cannot be convergence in prot rates. If there is a uniform rate of prots,
there cannot be convergence of rates of exploitation. Conversely, these two prop-
ositionsuniform rates of prots and of exploitationboth of which Marx insisted
on juggling simultaneously can only be valid if there are no deviations in sectoral
organic compositions of capital, which is a special circumstance and not the
general case.
1
1
We can represent the general rate of prot(p)asΣs/ΣC, where s is surplus value and C is the stock of
advanced capital (for simplicity it is assumed that there is no stock of variable capital). The organic composition
of capital (Q) is C/ (v + s), where v and s are ows of variable capital and surplus value during one average turn-
over. The annual rate of exploitation (s) is ns/v, here n is the average number of yearly turnovers and s/v is the
average rate of exploitation. From this p = ns/Qv (1+ s/n) = s/Q (1 +s/n). The rate of prot is, therefore, a
function of the organic composition of capital and the average annual rate of surplus value. From this, it can
be seen that all the sectoral rates of prot are equal if and only if all the organic compositions are uniform
234 B. Finger

Faced with a logical dilemma of his own making, Marx created (actually inherited
from classical political economy) a new category, a price of production, which he
treated as a modied value. It is modied, according to Marxs awed presentation,
because part of the value generated in one branch of production is realized in
another branch of production in accordance with the need to level the differing
rates of prots. But this seeming abandonment of the labor theory of value, in
which prices are no longer proportional to values (labor times), is said to be never-
theless reconciled at the level of the totality because the rate of prot and the rate of
exploitation in monetary terms remain equal to their corresponding rates expressed
in values and because the totality of prices is equal to the totality of values and the
sum of prots still equals the sum of surplus values.
Unfortunately, these assertions cannot withstand scrutiny. Price and value are
measured in different dimensions: money versus socially equalized expenditures of
time. So, there is no a priori reason why any expenditure of x units of time should
be equal to an expenditure of x units of money. Therefore, no reason why the sum
of prices should equal the sum of values, nor the sum of prots equal to the sum
of surplus values.
More important, when the prices of commodities are no longer proportional to the
values of commodities, the labor time expended (worked) to create these goods no
longer correspond to the labor-time needed to purchase them. In other words,
when prices are no longer proportional to values, the ratios that initially derive
from the assumption of proportionality (the rate of prot, the rate of exploitation
and the organic composition of capital) can no longer generally adhere in both mon-
etary and value forms.
The problem that binds these deciencies in logic resides in Marxs inability to
convincingly isolate and specify what market mechanisms deliver value from
spheres of production with low compositions of capital to those with high compo-
sitions. He simply attributes this to the movement of capital in search of prot advan-
tages. He assumes what he needs to prove.
It is undeniable that capital searches for differential prot opportunities and that
that may indeed lead to secondary changes in the social allocation of labor. But by
linking those secondary changes to the transference of value, Marx left himself
open to the absurd suggestion that accumulation is biased towards capital with
lower compositions of capital and, given the assumption of equal rates of exploita-
tion, higher rates of prot and toward innovations that are ever more labor intensive
than existing techniques. This, needless to say, contradicts the entire sweep of Marxs
analysis.
And it does so at a fundamental level. Changes in values (and their proportional
changes in prices) are rooted in changes in production techniques. If the expansion
and all the annual rates of surplus value are equal; or, if, given different sectoral organic compositions, there are
offsetting annual rates of surplus value that differ in each sector forcing individual sectoral rates of protability
to converge.
Critique 235

of production in one sector leads to a fall in value, this is attributable to some com-
bination of a greater efciency in the employment of capital or the introduction of
new techniques that allow more output to be produced with less expenditure of
capital per unit of output.
If, on the other hand, the price of output falls without any change in value (pro-
duction time), this suggests that the entire expenditure of labor cannot be deemed
socially necessary. But these market uctuations, changes in supply and demand con-
ditions, are not what Marx is getting at. Marx is clear that supply and demand have to
be abstracted from before value creation can be properly investigated. In other words,
it must rst be assumed that the entire expenditure of labor is socially necessary.
2
When it is not, part of the labor expended and the capital employed for its expendi-
ture is not transferred, it is wasted and lost.
The more i nteresting case is when rents are earned. Here temporary restric-
tionsofsupplypermitpricestoariseabove v alue (to deviate from labor-time
proportionality) and this suggests that at least s ome other capitals must concomi-
tantly realize less value as prot. Again, this has to do w ith transitory market
uctuations attributable to the workings of supply and demand (and not, in
thecontextunderexamination,totheexistenceofdurablemonopoliesornon-
competitivemarkets).Butanevenmoreimportant objection is this. Unless it
can be demonstratedand it cannot that the sectors that earn rents are also cor-
related with higher than average compositions of capital and the sectors that sell
at prices below value are conversely associated with lower compositions, this line
of approach cannot be said to support Marxs explanation of prices of production
as modied values.
Compounding these deciencies, rent taking is an add-on to the average prots
embedded in competitive prices. But Marx considers prices of production to be
just such prices. So again, Marxs approach is self-contradictory. Rents have no sig-
nicance without the assumption of pre-existing prices of productionto which they
are added, and prices of production, if we are to follow this line of thinking, cannot be
explained without recourse to rents, the very things that are eliminated through
market competition.
2
It might, again, be tempting to argue that short-run shifts in supply schedules could cause market-prices to
diverge disproportionally from values, for social labor to be squandered and for rents to be accrued. But, Marx is
clear; this would be an erroneous manner of approaching the subject. Why? To be able to study phenomena in
their fundamental relations, in the form corresponding to their conception, that is, to study them independent
of the appearances caused by the movement of supply and demand. And he continues, For a commodity to be
sold at its market-value, i.e., proportionally to the necessary labour contained in it, the total quantity of social
labor used in producing the total mass of this commodity must correspond to the quantity of the social want for
it, i.e., the effective social want. Competition, the uctuations of market-prices which correspond to the uctu-
ations of demand and supply, tend continually to reduce to this scale the total quantity of labour devoted to each
kind of commodity (Capital, III, New York: International Publishers, 1968, pp. 190 and 192). Equilibrium
prices (prices of production), in Marxs writings, are not simply a product of supply and demand as in main-
stream presentations, but rather of the alignment of production techniques with the demands of capital accumu-
lation for prot maximization over time.
236 B. Finger

These are the roots and effects of the inconsistencies in Marxs presentation, or at
least Marxs presentation as assembled by Engels.
3
But they are not insurmountable
contradictions with regard to the labor theory of value. Competitive prices can still be
salvaged by an approach consistent with Marxs theory. Part of his analysis must be
modied. Prot (abstracting from secondary deductionsinterest, rent, retail prot)
cannot be treated as a distributional relationship among capitalists in the spheres of
production, a form of capitalist-communism as Marx suggests. In other words,
surplus value cannot be said to be shared in proportion to the capital invested,
because Marx did not isolate any social (market) process that supports this
conceit. The prot-rate leveling process, therefore, does not involve re-division
through the expedient of repricing.
Lets take a step back. In and of itself, a price of production is nothing other than
the monetary expression of a value with special internal features. It is characterized
by an equal sectoral appropriation of prots in proportion to advanced capital. This
result bears no a priori logical necessity for values to be transferred among capitalists.
What can be modied is not the gross value that a commodity commands in
exchange, but the class distribution of its net value within each sector in accordance
with the demand for prot rate convergence. This means, rst of all, that prices of
production are still monetary values proportional to labor times embodied in com-
modities. But the distribution of value between workers and capitalists within the
sectors where values originate must be modied to accommodate prot rate
convergence.
Capitalist prices embodying a uniform rate of prot, in other words, can be per-
fectly consistent with value proportionality, once, contra Marx, the presumption of
a uniform rate of sectoral exploitation is rejected. Marx correctly pointed out that
the problem of prot rate leveling could not be resolved in isolation. If prot per
unit of capital invested is equal, sectors with higher organic compositions of
capital are compelled to modify sectoral working hours, rationalize the production
process and increase the intensity of work to compensate for the relatively dimin-
ished base of exploitation. This is also mitigated, to a certain degree, by the employ-
ment of skilled workers, who can even at lower rates of exploitationgenerate more
value and therefore more surplus value in a given unit of time, and which thus, also,
has the knock-on effect of lowering the organic composition of capital. Finally, com-
petition pulls capital out of those applications with low rates of exploitation and into
those where the opposite is the case, that is out from where the rates of exploitation
are too low to appropriate the average rate of prot and into those where the rates of
exploitation are high enough to support above-average prot rates. The expansion of
output, at one pole, causes the turnover times to increase and therefore the annual
rate of surplus value to fall and, with it, the rate of prot; the shrinkage of supply,
at the other pole, causes turnover times to decrease with a concomitant increase in
3
See Geert Reuten, The Productive Powers of Labour and the Redundant Transformation to Prices of Pro-
duction: A Marx-Immanent Critique and Reconstruction, Historical Materialism, 25:3 (2017), pp. 335.
Critique 237

the annual rate of surplus value and rate of prot. These synchronous movements
interact to bring prot rates into convergence. None of these movements are incom-
patible with value stability in pricing.
The operations of the theory of value, as propounded by Marx in the rst volume
of Capital, remain, in this way, fully operable and logically consistent as the purview
is expanded, as it is in Marx, from capital in general to the many capitals in compe-
tition in the third volume.
II. Marxs critics
IIA. Bortkiewicz solution and its variants
According to Marx, price is value (labor-time) in the form of money. So the trans-
formation of values into prices of production rests on the understanding of some
functional relationship between the expenditure of social labor time and the unit
of account (the pricing unit). If prices are proportional to values, the labor time
newly worked up (V + S) corresponds to its monetary equivalent $(V + S) as an
accounting identity. From this identity, the labor time content of the monetary
unit (or conversely the monetary equivalent of a unit of expended labor time) is
established in accord with the proportionality of labor-time in exchange.
Marx breaks the proportionality basis of this accounting identity in volume 3,
leading many critics to question the consistency of his analysis. Others pounced
on this apparent contradiction to offer alternative interpretations. Under the most
famous re-interpretation, the Bortkiewicz algorithm
4
and subsequent variants of
that algorithm, the recalculation of input values are simultaneously transformed
with output values into prices of production. For Marxs critics, this alleged
deciency was the most glaring. But if input values have to be re-priced, it is only
because Marxs erroneous contention is blindly accepted that surplus values are
and have been simultaneously (or previously) redistributed through market forces.
This redistribution must therefore be consistently carried through, at some point,
both on the input side as well as on the output side. Bortkiewicz and his followers
argued that this redistribution should be simultaneous; Marxs defenders, that this
should be sequential and temporal.
The proposed correction by Bortkiewicz establishes a series of exchange ratios that
can only be normalized as distinct prices by the selection of an appropriate numer-
aire. This numeraire fullls the function of money, insofar as it is a unit of account
and a medium of exchange.
The choice of this numeraire is also utterly arbitrary, since it has no other signi-
cance other than to transform ratios into absolute numbers. So, for example, if there
4
Ladislaus von Bortkiewicz, On the Correction of Marxs Fundamental Construction in the Third Volume
of Capital in Paul Sweezy (ed) Karl Marx and the Close of His System and Böhm-Bawerks Criticism of Marx
(New York: Augustus M Kelley, 1949) pp. 199221.
238 B. Finger

are three commodities, there are two independent exchange ratios: say, A/B and B/C
from which A/C can be inferred; and therefore the logical space to insert a pricing
unit. The numeraire (money unit) allows A, B and C to be expressed in absolute
prices while preserving these ratios, rather than as a list of discreet exchange ratios
relative to other commodities.
Bortkiewicz posited a monetary unit, in which the labor time incorporated in the
surplus product (Σs) is equated with its monetary equivalent $(Σs). So, in his example,
the 200 hours of labor time that are needed to generate the aggregate surplus product
(equal, also, under conditions of simple reproduction, to the total labor time
expended in the gold sector), before the transformation of values into prices, are arbi-
trarily represented by $200 worth of monetary units after the transformation. Com-
modity prices, value in the form of money, are now represented by the labor-time
embodied in the gold for which they command in exchange and not by the labor-
time directly expended on their own production.
According to the other variants that equate total value with total price, the labor
hours expended on the gross product Σ(c + v + s) is made the equivalent of an
equal sum of currency units $Σ(c + v+ s), so that x hours, living and dead, expended
on gross output are equated with x dollars.
The Bortkiewicz correction is mathematically coherent and economically mean-
ingless. Either of the invariance principals appealed to (total values in labor-time
equal to total monetary prices, or total surplus values in labor- time equaling total
monetary prots) is chosen simply to appease the thirst for orthodoxy. The deeper
signicance in the Bortkiewicz example itself is that the prices of all commodities are
normalized in terms of the abstract quantity of labor embodied in the gold for which
they exchange. That Bortkiewicz adds the utterly nonsensical insistence of making
1 hour expended on the surplus product represent $1 of prot, however, has been
pored over and debated as if something of overwhelming marxistical signicance
were at stake.
The total price to total value equivalency, on the other hand, signies that the price
of individual commodities are instead normalized in terms of the fractional part of the
gross labor time expended on the corresponding aggregate product against which these
individual commodities are exchanged. Or to state that proposition in different terms,
$1 is equated to the gross value added (overhead costs consumed and preserved plus
wages of productive workers plus prots) on average through the exertion of one
hour of productive labor time. Again, it is of no substantial signicance that the
sum of the fractional parts in dollars equals the sum of the fractional parts in
hours. Either way, individual prices simply represent the portion of total value com-
manded by commodities in the process of exchange.
A similar objection can be raised against the so-called new solution of Duncan
Foley.
5
According to Foley, although production prices deviate from corresponding
5
D. K Foley, The Value of Money: The Value of Labor Power and the Marxian Transformation Problem,
Review of Radical Political Economics, 14: 2 (1982), pp. 3747.
Critique 239

labor values, the ratio of the labor value of the net product (V + S) is equal to the pro-
duction price of the total net product $(V + S). The production prices of each element
of the wage bill and the surplus product are not equal to or proportional to the value
of the individual commodities, but this deciency is said to be reconciled when taken
as a whole.
This just raises the same problem in a slightly modied form. For again, there is no
meaningful economic signicance to the proposition that 1 hour of productive labor
time must be set equal to $1 of net product.
It is Marx who rst insisted on an invariance between total prices and total values
and total prots and total surplus values, despite the logical deciency behind that
claim: by virtue of the fact that values and prices are measured in two different
dimensionstime and money. There is no reason for these to directly equate. For
the labor theory of value to be coherent, they need to be proportional, not identical.
That is why these pricing units serve no explanatory purpose and clarify nothing
about the distribution of social labor time, not only among the many sectors of the
economy, nor of the distributions of dead to living labor, and paid to unpaid
labor. These are the central functional distinctions that the law of value is said to
reveal. Once the proportionality principal in exchange is dropped, the assumption
of an averageor in Marxs case, uniformrate of exploitation and rate of prot
must be modied. There is no longer any way of imputing a meaningful labor-
time content arising from production to any of the elements of c, v and s. And
why would anyone attempt to, since they have no operational, no lawful, signicance?
For in fact, all these distinctions are effectively effaced by these corrections.
These algorithms did not redeem the labor theory of value; they laid the foun-
dation for its later abandonment as a needless redundancy under the impact of the
neo-Ricardian school, or dismissed altogether as an erase and replace procedure
on the part of mainstream economists.
IIB. The neo-Ricardian critique
One of the Srafan schools principal aim, and the one whose implications concern us
here, is to demonstrate the inconsistency of any theory of prot based on the concept
of a marginal productivity of capital. And by capital they mean the physical means of
production. It is rst of all a critique of neo-classical economics. As in neo-classical
price theory, there is no capital accumulation and no technological change. Unlike
neo-classical theory, the given array of means of production and workers are not
assigned alternative production purposes in response to changes in consumer prefer-
ences. These would otherwise change the composition of output by expanding the
production of some commodities while reducing that of others. By positing a static
output structure as it interacts with changes in class distribution the conditions
that give rise to the context of marginal productivity are eliminated.
This has nothing to do with the labor theory of value, quite simply because the
relationship between time and moneyvalue and pricepivotal to Marxs analysis,
240 B. Finger

plays no role in neo-Ricardianism. Their models are built on the simultaneous deter-
mination of rates of prot and relative prices, once the technical conditions and phys-
ical wage structure are specied (or, alternatively, with a posited rate of prot,
physical wages and relative prices can be derived consistent with that rate of
prot). The rate of prot, wage goods, means of production and the technical struc-
ture of production are all in physical units. Relative prices (exchange ratios) are then
converted into absolute prices, by the selection of a suitable numeraire condition or
simply by arbitrarily setting the price of some physical commodity equal to 1.
It is an exercise in mathematical logic: to examine how the patterns of prices
behave in a given production-output structure, when either the (material) wage or
prot share is run from zero to maximum. Prices are then shown to be dependent
solely on the distribution of net output between workers and capitalists. The
height of the rate of prot is purely a function of the degree to which real wages
can be squeezed. Again, this does not purport to be a correction of Marx, as
much as a dismissal of Marxs approach and its replacement with an alternative fra-
mework in the spirit, perhapsor at best, of Marxism.
Nevertheless, it is the very dismissal that speaks to the argument that the labor
theory of value is a needless detour, because the labor theory of value, it is held,
cannot support an adequate theory of competitive prices. Prices are only proportional
to values under the heuristic, but hypothetical and unrealistic case that prot rates (or
wages) are zero. Otherwise, the neo-Ricardian presentation of prices consists, again,
as a series of exchange ratios normalized by a numeraire that that is a stand-in for
money.
In their dismissal of the labor theory of value, the neo-Ricardians embed certain
assumptions that make a consistent presentation of prices in the framework of the
labor theory of value unattainable. They make no distinction between skilled and
unskilled labor hours, differences in labor intensity and differences in rates of exploi-
tation. That is, they, like Marx, operate under the assumption that all hours worked
are hours expended under conditions of average skill and intensity. This volume I
expedient, where capital as a whole is the subject, is then carried through to
volume 3, where the many capitals in competition are the subject. But by carrying
this assumption over, Marx and those who either criticize or defend Marx, box them-
selves in. If only labor of average skill is employed and under equal levels of intensity,
a uniform rate of exploitation is the unavoidable corollary.
Consider, for purposes of illustration, an example posed by Ian Steedman in Marx
After Sraffa.
There are three spheres of production: iron (means of production), gold (luxury
goods, but more pertinently, money
6
) and corn (consumption goods).
6
Steedman adds, Since one unit of gold is produced by one unit of labor (), it follows that the price of
commodity is numerically equal to the labour embodied in the gold with which that commodity exchanges.
Again, Steedman like Bortkiewicz and as we will see, like Marxs defenders, resorts to a command theory of
Critique 241

Iron
Labor Iron Gold Corn
Iron 28 56 56
Gold 16 16 48
Corn 12 8 8
The value of iron therefore is equal to 56 hours of labor + 28 units of iron = 56 units
of iron. Or 2 hours of labor is equal to 1 unit of iron. From this, we can state that 1
unit of gold equals 1 hour of labor, and 1 unit of corn equals 4 hours of labor. Now we
can re-present this in terms of labor-times.
Iron:C + (V + S) = 56 + 56 = 112
Gold:C + (V + S) = 32 + 16 = 48
Corn:C + (V + S) = 24 + 8 = 32
Steedman then adds the assumption that wages consist of 5 units of corn. Because a
unit of corn is the equivalent of 4 hours, the total wage bill is 20 labor hours. And,
since total labor expended is 80 hours, total surplus value consists of 60 hours.
The rate of exploitation is uniformly given as 60/20 or 300 percent. The aggregate
rate of prot is 45 5/11%
So in complete form, c + v+ s:
Iron: 56 + 14 + 42 (individual rate of profit, 60%)
Gold:32 + 4 + 12 (individual rate of profit, 33%)
Corn:24 + 2 + 6 (individual rate of profit, 23%)
It is intuitively obvious that this scheme, with equal rates of exploitation, cannot yield
equal rates of prots, because the ratio of labor to capital (the organic composition of
capital) differs among the 3 branches.
But appearances conceal unacknowledged assumptions.
For we have no reason to assume that cross-sectoral production structures have a
uniform rate of exploitation. This cannot be the general case. Nor can we simply sti-
pulate a real wage. The rate of exploitation and the real wage are outcomes not pre-
conditions of both market activity and sectoral interclass power relations. Each
branch of capital consists of a different mixture of skills and labor intensity. Steed-
mans schema only approaches minimal sense if we think of each sectoral v + s as
socially equalized labor. Otherwise, nominally equal expenditures of labor time by
differing skill-mixes exerted with varying intensity are erroneously treated as
having an equal ability to expand value in a given period of time. By what reason,
in other words, could we justify the assumption that an hour of labor collectively
expended by some combination of ironworkers and machinists, gold miners and
value. For the labor embodied in the gold for which a commodity exchange represents the value of the commod-
ity only if exchange takes place according to value equivalents.
242 B. Finger

goldsmiths, farm hands and tractor mechanics yields the same value? No one vaguely
familiar with the labor theory of value would, for instance, argue that a gem cutter
generates the same value in an hour as a ditch digger.
Once this is conceded there is also no basis for the additional assumption of a
uniform rate of exploitation.
What market forces compel the employers of ironworkers and machinists, gold
miners and gold smiths, farm hands and tractor mechanics to divide the net
income uniformly? The rate of prot is the key variable. And the only control that
an individual capitalist unit has to effect change in it, short of technological inno-
vation, is through altering the work process, by raising productivity, eliminating inef-
ciencies, hiring skilled workers, or conversely, by deskilling and raising the intensity
of work and by lengthening or shortening the amount or duration of shifts in the
workday.
What they cannot do is poach surplus value from sectors that are more protable
than theirs. Yet Marx, his critics and defenders, operate under the tacit assumption
that there is some automatic reallocation mechanism embedded in the logic of
market optimization.
If we, in contrast, assumearbitrarily for instance that each branch earns a 25%
rate of prot,
7
then unmodied values that have the characteristic of prices of pro-
duction result:
Iron:56c
i
+ 33.6v
i
+ 22.4s
i
= 112
i
, (rate of exploitation 66%)
Gold:32c
g
+ 6.4v
g
+ 9.6s
g
= 48
g
(rate of exploitation 150%)
Corn:24c
c
+ 1.6v
c
+ 6.4s
c
= 32
c
(rate of exploitation 400%)
Here we see that prices of production can be proportional values, without the
assumption of a uniform organic composition of capital. Prices of production are
a special form of value, but not a departure from value. Real wages emerge from
the competitive struggle. They are not predetermined. Should 5 units of corn be con-
sumed uniformly as wages, they are therefore distributed in proportions of 4.04 to the
iron sector: .77 to the gold sector and .19 to the gold sector. Since wages exceed that
needed to buy 5 units of corn, the remainder is devoted to luxury goods (represented
in this scheme by gold).
There are also some peculiarities worth noting. The maximum rate of prot (again,
dened for purposes of this illustration as the ratio of prots to prime costs) for the
iron industry is 100%; for the gold industry 50% and for the corn sector 33%. There-
fore, if there is to be a uniform rate of prot, given the assumptions of a unitary rate of
turnover in all sectors and no stocks of capital, that rate must be between 0 and 33%.
For it is not possible for the corn sector to raise its rate of prot beyond the point
where it pays workers the equivalent of air.
7
Following Steedman, we represent the rate of prot here in ow terms as s/(c + v). It is assumed that there
is no capital stock.
Critique 243

Yet, Marx, in Steedmans presentationand Steedmans interpretation is correct
sets the general rate of prot as:
60/(112 + 20)or at about 45 5/11%,
which is now not only the aggregate rate of prot, but also the individual sectoral rate
of prot.
Not to put too ne a point on it, neither Marx nor any of his defenders has ever
logically demonstrated a convincing process whereby surplus values are collectively
pooled and proportionally shared. Still, this assertion of capitalist communism is
behind the erroneous Marxian belief, mirrored inadvertently by Steedman, that the
totality of aggregate surplus value (or surplus product) is available for redistribution.
Its assertion is a mystication that remains unchallenged and unexamined. Surplus
value is a social relationshipcrystallized labor-timeembedded in specic material
or immaterial commodity form that only upon sale takes the form of abstract labor,
i.e. money. It is not a freestanding liquid that can freely ow as between communi-
cating vessels of uneven circumference in some logical black box after production
and before sale.
Redistribution, as such, is only possible through prior contractual agreement (such
as between retailers and producers, or lenders and borrowers, or landowners and
leasers) or through the assertion of monopolistic power. None of these forces are
at play here. And Marx was adamant that prices of production as a category had
to be explainable before these further concretizations could be considered.
But that is not Steedmans basic objection. In fact, the neo-Ricardian variant tacitly
builds this assumption into its own models. For Steedman, the logical aw in Marx is
rather the latters failure to simultaneously convert input values into prices consistent
with output prices. Nevertheless, the entire neo-Ricardian correction brings this
problem in again through the back door.
Steedman solves the pricing issue, based on a prior assignment of real wages, by
means of the following equations:
A.(1 + r)(28p
i
+ 56w) = 56p
i
,
where the price of iron and labor wages multiplied by one plus the rate of prot,
equals the price of iron industry output. Similarly, in the gold industry:
B.(1 + r)(16p
i
+ 16w) = 48,
where gold is the money commodity and therefore the pricing unit. This means that
the prices of all other commodities are measured by the amount of labor embodied in
the gold for which they (in this case iron and corn) exchange. (A unit of goldpre-
sumably some arbitrary weight, at some arbitrary level of purity’—is assigned a
244 B. Finger

monetary value of 1, say one dollar.)
C.(1 + r)(12p
i
+ 8w) = 8p
c
,
D. 80w = 5pc.
Thus the four equations determine the four variables r (rate of prot), w (monetary
wage), p
i
the price of iron and p
c.
the price of corn, such that
r = 52.08%, w = 0.2685, p
i
= 1.7052, p
c
= 4.2960
Now, plugging these numbers back into the physical input- output chart above:
Iron
Labor Iron Gold Corn
Iron 28 (1.7052) 56 (4.2960) 56 (1.7052)
Gold 16 (1.7052) 16 (4.290) 48
Corn 12 (1.7052) 8 (4.290) 8(4.2960
In the value scheme, capitalists in total obtain 48 units of gold plus 3 units of corn.
In the iron industry, the expenditure on wages of 14 hours purchases 14/4 units of
corn, or 3.5 units of corn; in the gold industry wages amounting to four hours pur-
chase 4/4 units of corn or 1 and in the corn sector 2 hours of paid time purchases .5
units of corn.
Capitalists in the iron industry can consume a maximum of 42 units of gold and no
corn or some combination of units of corn and gold equaling 42 hours of expended
surplus labor time.
Capitalists in the gold industry can consume a maximum of 12 units of gold or
some combination of gold and corn not exceeding 12 hours of surplus labor time.
Capitalists in the corn industry can consume a maximum of 6 units of gold or some
combination of gold and corn not exceeding 6 hours of surplus labor time.
But what happens when we look at the pricing scheme? Since r = 52.08%, capitalists
in the gold and corn sectors can now consume more value as prots and therefore as
material output, than their workers generate as surplus value and surplus product.
8
It is this that needs justication and, like the Marxists they criticize, no such jus-
tication is forthcoming, nor the need for such a justication even recognized. Neo-
Ricardianism is an example of where a seemingly impeccable exercise in mathemat-
ical logic breaks down for its failure to deliver a result consistent with the deciencies
in economic logic it presumes to correct.
But, as we will show below this is not the only place where logic breaks down.
8
The intuitive conclusion alluded to here is given formal substance by, among others, Anwar Shaikh,
Marxs Theory of Value and the Transformation Problem”’ in Jesse Schwartz (ed), The Subtle Anatomy of
Capitalism (Goodyear Publishing Company, 1977). Shaikh demonstrates that the same rate of prot can be cap-
tured through an iterative procedure of surplus-value redistribution over logical (as distinguished from histori-
cal) time as can be arrived at by algebraic shorthand.
Critique 245

IIC. Prices of production and non-reproductive commodities
One of the peculiarities of the neo-Ricardian approach is a specic corollary, that the
conditions of production of that subset of privately produced goods which enter
neither directly nor indirectly into the consumption basket of workers have no
effect on the average rate of prot. Generally, what are being referred to here are
waste goods, armaments, etc., and luxury goods. The theory of the Permanent
Arms Economy as a partial explanation, a source of stabilization, for the post war
boom was premised on this neo-Ricardian conclusion. PAE theorists argued that
increases in the organic composition of capital have no part in determining the
general rate of prot. So, as ever more surplus value is accumulated in waste pro-
duction, the system shields itself from the more fulsome effects of a falling rate of
prot that would otherwise accrue if investment were shifted to the basic (reproduc-
tive) sectors.
The goods in question, the means of warfare for instance, are narrowly non-repro-
ductive with respect to the reconstitution of labor power as a commodity, but in the
larger sense they are essential to the reproduction of the social and political nexus
the totality of class relationships that constitute class society in its entirety.
It is important to draw a distinction between these non-reproductive commodities
and nonproductive goods and services. Often these are wrongly treated as inter-
changeable. One category of nonproductive
9
use-values is goods/services that are
not produced by capitalists for sale on the market place. Examples of such are
public education, public transportation, public hospitals, infrastructure, basic
research and development, etc. These goods may and often are reproductive
insofar as they are necessary to maintain or enhance the value of labor power. But
the surplus labor time expended in their production is not appropriated by capitalists
and is not a source for additional private investment. Nevertheless, insofar as they are
provided to the private sector at tax rates considerably below what the market place
would otherwise charge for these services if privately offered, public goods can con-
stitute a partial subsidy to capital.
Of course, non-reproductive goodsif nationalizedwould remain non-repro-
ductive but they would now also be nonproductive of private surplus value. If such
nationalized goodssay, armamentswere sold or exported at prevailing levels of
prot above costs on the world marketthey would be revenue enhancing for the
state sector and could facilitate a reduction in private taxes. If they are sold to the
private domestic sector, on the other hand, they might, also, suppress the demand
for the aggregate output of (private) commodity production and thereby potentially
affect protability.
9
Another category involves the spheres of circulation, banking, accounting, bookkeeping and mercantile
capital. The capital (including wages) employed here are socially necessary overhead costs of capital. But it is
not primarily allocated to commodity productionmaterial or non-materialbut to the transfer of ownership
titles of preexisting commodities. It therefore produces no surplus value.
246 B. Finger

Conversely, nonproductive goods, whether reproductive or non-reproductiveif
privatized, would be a source of enhanced private accumulation. The singularity of
neo-Ricardian analysis, however, resides in the conclusion that the conditions of pro-
duction of privately manufactured non-basic goods do not play a part in the formation
of the uniform, general rate of prot. That subsidy to the private sector in the form of
reduced levels of taxation, raise after tax protability and enhanced taxes, say to sup-
press demand, may lower prot levels, and thereby protability is not in dispute. But
these differential impacts on protability, while often tied to the issue of nancing
non-basic goods, have nothing to do with production conditions.
More generally, non-basic goods (armaments, say) are privately produced for the
state. If the state guarantees a reasonable prot, then it would appear that the neo-
Ricardian corollary proves apt. But this is a political decision, not a conclusion dic-
tated by economic laws. The state agrees to purchase these commodities at a given
price. Subject to that agreement, private producers must adjust their rate of exploita-
tion to realize the average, or prevailing, rate of prot. That is, the nonreproductive
private sector attains its prot target in virtually the same manner as the reproductive
sectors.
Steedman argues to the contrary. In line with Bortkiewicz (and Sraffa), offers a
demonstration as to why the average rate of prot is independent of the production
conditions for non-reproductive or non-basic goods. To see this, he writes in foot-
note 12, p46,
ignore the second of our four equations (A-D in the previous section, BF) and
divide through each of the others by p
c
. There are then three equations and only
three unknowns, namely r and the ratios (w/p
c
) and (p
i
/p
c
). Thus r can be deter-
mined, as can (w/p
c
) and (p
i
/p
c
), quite independently of the production conditions
of the gold industry. The latter conditions merely determine, via the second
equation, the absolute levels of w, p
i
and p
c
.
The root problem with this conclusion is that the sweep of Steedmans generaliz-
ations is too broad. What his exercise shows is not that the average rate of prot can
be determined without reference to conditions in the gold (non-basic) sector. Rather,
it shows only that a general rate of prot in the basic sector can be derivedosten-
siblywithout reference to the gold sector. The rate of prot in the non-basic sector
is not generally determined or determinable with the data provided; neither is the
price of the non-reproductive goods. In this particular example, Steedman simply
assigned gold units a monetary price in advance. If, on the other hand, a non-mon-
etary non-basic good was added to the mix, there would be no pseudo-economic jus-
tication for the pre-assignment of any price.
Steedman, after assigning a pric e t o t he n on-reprodu ctive g ood, s imply assumes
a uniform general rate of prot and, based on that assumption, assigns the rate of
prot in the basic sec tor arbitra rily to the non -basic secto r. The rate of protinthe
gold sector, notwithstanding Steedman s conceit, can, in contrast, exceed, equal or
fall short of r. And because of this, r is not a necessary uniform rate of prot, and
neither therefore is r necessarily the average rate of prot, which is t he weight ed
Critique 247

average of the two sectors, the determined basic and indeterminate non-basic
sector.
And because the individual rates of prot in both sectors are products of the indi-
vidual rates of exploitation and organic compositions of capital, the system-wide
average rate of prot cannot, contra Steedman, et. al., be independent of the pro-
duction conditions in the non-basic sector.
Thus, far from offering a cohesive alternative explanation to that of Marx for estab-
lishing the uniform, average rate of prot and competitive prices, the neo-Ricardians
prove that such a determination cannot likely be established along the path they
chart. The non-basic system consists of just as many equations as there are non-
basic commodities. The actual unknowns are the prices of the non-reproductive
commodities and the general rate of prot of the non-reproductive subsystem.
All of which is to say that the general rate of prot of the non-basic sector and the
prices of non-reproductive commodities cannot be uniquely determined simul-
taneously with the prices of the reproductive commodity subsystem, without rst
assigning the non-reproductive commodities a price and a rate of prot, the rate
that prevails in the reproductive sector. This approach is, therefore, quite simply
question-begging.
Far from establishing the possibility of setting system-wide prices and prot rates
from the physical data, a more intensive investigation suggests that neo-Ricardianism
rests on a series of tautologies. It establishes neither an average nor a uniform rate of
prot nor a comprehensive set of prices of production in the absence of a hidden set
of unexamined and indefensible assumptions.
If neo-Ricardianism is a detour from the labor theory of value, it is a detour cul-
minating in a dead-end.
III. Marxs defenders
IIIA. Fred Moseley
The perspective defended here is that the transformation problem resides in the
misstep that Marx introduced in reconciling values and competitive prices, a
misstep that was rooted in the ambiguous legacy of classical political economy.
That legacy suggested to Marx an unresolved social tension between values and
the natural price of commodities. Ricardo and Smith often treated these as simply
overlapping. The thesis of this essay is that they are, indeed, overlapping, but there
is nothing straightforward or simple about this. Marx sought the resolution of that
tension in the market redistribution of surplus value, which rendered competitive
prices as modied values. Yet, as we have argued, there can be no such reconcilia-
tiongiven the specic criteria Marx raised as proof of that reconciliationwhen
competitive prices and values no longer align.
Marxs critics, in this sense alone, caught him dead to right, though his orthodox
followers have continued to wage a prolonged rearguard defense.
248 B. Finger

And few have been more heroic in that defense than Fred Moseley and Andrew
Kliman. Let us consider Moseley rst.
To illustrate Moseleys take on this, he compares simple value-prices with prices of
production that violate the rule of proportionality in exchange as does Marxs prices
of production. But there is a twist. Simple value-prices can be represented by M
c
+M
v
+M
s,
where Mc, + Mv are (monetary) cost-prices of the means of production and
wage goods. They simultaneously represent the hours need to produce the quantity
of gold used to purchase these commodities. These are distinguished from and c, v,
which represent the labor times need to produce these goods, such that Mc:Mv = c:v.
M
s
represents the monetary expression for the surplus value directly generated, such
that M
s
also signies the quantity of labor expended on gold production needed to
purchase the surplus product but not the labor time actually expended on its pro-
duction. It is this M
s
which is redistributed in accordance to the relative weight of
each capital in the aggregate that gives rise to prices of production, M
c
+M
v
+
rK/ΣK (where ΣK is the aggregate stock of invested capital and K is the sectoral
investment and r is the aggregate rate of prot(ΣM
s
/ΣK). Alternatively, rK/ΣK
can be represent by M
p
for each sector of production.
These monetized cost-prices, per Moseley, reect the incorporation of previously
redistributed surplus values, which is why they are no longer proportional to the
embodied values required for the production of these goods. Howas opposed to
wherethey are redistributed is not really addressed. But the twist is this. Moseley
introduces a dual accounting system, where the hours represented by prices of pro-
duction for the various classes of commodities in circulation are distinguished from
the hours represented by socially necessary production times.
So, how does this resolve the transformation problem? By reframing the issue.
Assume M
c
+M
v
(cost-prices) are given at the beginning of production, and that
the labor content of the monetary unit is determined not post factum by the relation-
ship between the price of the annual product $(V + S) that emerges and the pro-
ductive labor hours expended (V + S), as defended in this paper, but rather by the
amount of labor time represented by the sums of gold units invested as capital.
Again, keep in mind that the money unit (weights of gold, at a given level of
purity), for Moseley, acts as an exogenously existing ex ante constant that represents
a given quantity of abstract labor time.
On this basis we can begin to see how Moseley can claim that sum of simple value-
prices (Moseleys term) equals the sum of prices of production and the sum of surplus
values associated with simple value-prices equals the sum of prots corresponding to
prices of production. To his credit, Moseley in his interpretation of the transform-
ation problemunlike Bortkiewiczfully appreciates that values have to be
expressed in common price form, before being equated/compared with prices of
production.
They can be so associated because, per Moseley, any sum of M, regardless of the
actual labor-time incorporated in the commodities for which they exchange, rep-
resentby denition or presumptiona given quantity of abstract labor embodied
Critique 249

in the monetary material (gold). Equal quantities of M regardless if expended on
commodities priced proportional to embodied labor-time or at prices that diverge
from embodied labor-time (such as his/Marxs prices of production) always represent
the same quantity of value (abstract labor time). Similarly, any given return on invest-
ment, ΔM, which corresponds to the given M invested at the beginning of the pro-
duction period, always represents, again by presumption, a given quantity of abstract
labor time in the form of money, regardless of the conditions under which the com-
modities composing the aggregate surplus product are produced.
Simple values-prices consist rst of all as the sum of monetized input values ΣMc
+ v (cost-prices) and this is the invariant structure that bridges the two subsystems
(simple value prices and prices of production). The only thing that changes is the dis-
tribution of monetized surplus value arising from the current period of production.
Previous redistributions are said to have been already accounted for in cost-prices. (I
have already criticized the question-begging notion behind this supposed redistribu-
tion and will not revisit it here.)
Stated differently, a given expenditure of M for productive inputsmeans of pro-
duction and wage goodsis an outlay of a quantity of the money material (gold) that
therefore implicitly corresponds to a given amount of abstract labor-time embodied
in the gold units for which they exchange. And this holds, regardless of the labor time
embodied in the commodities purchased by M. Similarly, ΔM represents a given
quantity of labor time bearing no necessary relationship to the socially necessary
labor time embodied in the commodity components of the surplus product. And it
too holds under any conceivable condition of exchange. As embodied labor-times,
c, v and s are effectively shelved. They are, perhaps, to follow Moseleys unstated
logic, a provisional approximation of reality that is abandoned as Marx incrementally
approached the surface conditions of market exchange. For what now matters above
all is the labor embodied in the monetary material.
There are now thus two completely different and incompatible expressions of the
labor-content of the monetary unit at play in Moseleys Marx: the exogenous one, a
value-xed pricing numeraire, employed by Moseley (and Bortkiewicz) and the
endogenous one immanent to the circulation of commodities as a collection of embo-
died values, which is employed by Marx in volume I. They imbue each commodity
with a different quantity of abstract labor time.
10
Moseleys measure goes beyond a mere restatement of Marxs contention that
prices of production are modied values. It suggests that a commodity value can
10
What Bortkiewicz and Moseley miss, in any event, is this. The value of gold is regulated by the labor-time
content of the money unit, which is determined by the relationship between the net sum of the social product ($
Σ (v + s)) and the expenditure of productive labor-time (Σ (v + s)). So, for example, if money is gold-backed and
1-ounce of gold is dened to equal, say, $35, we know nothing of the value of that 1-ounce. Moseley, on the
other hand, wrongly treats the labor-time content of gold as given, as if the dened exchange rate between a
xed weight of gold and the monetary unit also confers a xed value to that unit. But golds value varies as a
function of labor productivity, as do commodity values is in every other sector. Those mines that cannot
yield at least an average rate of prot will shut down; those that earn a higher rate can pay rents.
250 B. Finger

be modied by changes in its price, that is, by changes in the terms of exchange with
which a commodity is swapped out for gold.
If commodities exchange in accordance with the labor-time directly and indirectly
associated with the production of each commodity, money is simply the veilthe
universal equivalentby which the proportionality of labor-time in the marketplace
is socially constituted.
If, on the other hand, money is an exogenous standard of value determined by the
production conditions that prevail in the gold sector, prices are no longer the barom-
eter of labor time embodied in the production of commodities, but a measure of the
labor-time commanded by commodities in their exchange with gold. So, on the one
hand, commodities possess value by virtue of the fact that they are the products of
social labor. But the quantity of social labor employed in their production is no
longer the engine that regulates the amount of labor-time commodities can purchase
or command in the process of exchange.
This is a reversion to Adam Smith. As I. I. Rubin explained, Smiths analysis of
exchange value becomes bifurcated and ows along two methodologically different
channels: the one the discovery of what causes changes in value, the other the
search for an invariable measure of value. Each of these paths leads Smith to a par-
ticular conception of labor value or of labor as the basis of value. The rst leads him to
a concept of the quantity of labour expended on the production of a given product; the
second to a concept of the quantity of labour which a given commodity can acquire or
purchase through exchange (A History of Economic Thought (London: Ink Links,
1979), pp. 186187).
So, on the one hand, commodities possess value by virtue of the fact that they are
the products of social labor. But changes in the quantity of social labor employed in
their production are no longer the sole determinates that regulates the amount of
labor-time commodities can purchase or command in the process of exchange.
Moseleys money remains a unit of account, a medium of exchange and a store of
(gold) value, but does so at the cost of shedding its ability to act as a universal equiv-
alent. In effect, he rescues Marx by returning to Adam Smith and the classical
approach in search of an invariant measure of value. Labor values are replaced by
production costs, (M), and revenue streams in excess of costs, ΔM, because
Moseley, like Marx before him, and everyone who has trod Marxs path after him
and insist on reproducing Marxs logical deciencies, nd it impossible to make
the exchange of materialized labor equivalents otherwise accord with the capitalist
pricing system.
IIIB. Andrew Kliman
11
Kliman approaches the transformation problem from a somewhat different perspec-
tive than Moseley, though, I will argue, he reaches similar results. He considers the
transformation process to be dynamic, executed sequentially through the process
of production and the turnover of commodities from one period to the next. He
Critique 251

faults the followers of Bortkiewicz with simultaneously attempting to transform
inputs and outputs from values into prices, rather than treating the pricing of
inputs in any particular period as data from a previous period. That is, Kliman
and his school challenge Bortkiewicz and the neo-Srafans on the very grounds
where the supposed inconsistencies the latter attribute to Marx are said to be located.
There is an additional wrinkle that Kliman also reintroduces. He, perhaps alone
among modern defenders of Marxs approach, distinguishes the independent but
often wrongly conated criticism of Böhm-Bawerk, with that of Bortkiewicz. That
is, while the latter deals with the supposed deciencies in chapter 9 of volume III
of Capital (Formation of a General Rate of Prot), Böhm-Bawerk calls Marx to
task for the incompatibility in pricing formations between volumes III and I.
It should be clear from the preceding that I consider Böhm-Bawerk s criticism of
Marx to be both appropriate and pertinent, but not insurmountable. It is true that
Marx foreshadowed the transformation issue in volume I, suggesting that price/
value proportionality in exchange is a provisional assumption adopted for purposes
of exposition. But if individual prices cannot be held proportional to value, nor prots
proportional to surplus-value then there is no reason to assume that these dispropor-
tionalities can be meaningfully reconciled at an aggregate level without modifying
our understanding of how values are measured, a modication arising from
changes in the conditions which regulate the exchange of commodities.
Again, values are an expression of the expenditure of time; price a re-expression of
that expenditure in terms of money. Between the two we can deduce the labor-time
content of the monetary unit and its inverse the monetary equivalent of a unit of
expended labor-time. The better test of consistency, then, is if the ratios (organic
composition of capital, rate of exploitation and rate of prot) expressed in time
and in money are themselves proportional.
But is Klimans approach a proof, a demonstration of consistency, or, is it, as is
Moseleys, the wayward introduction of a parallel command system? Moseley, to
restate, takes the input values to have been previously converted into prices of pro-
duction with surplus values having been redistributed in previous circuits of pro-
duction and circulation. These transformed factors of production have a
monetary price and these prices represent given quantitities of gold and therefore
represent an equivalent quantity of labor-time crystallized in the monetary material
for which they were exchanged. What they do not represent is the labor times directly
expended in the production of these factors.
In what signicant way does Klimans approach diverge from Moseleys? Kliman,
like Moseley, holds that the input data of any successive period of production and
turnover are the prices of production that have been established in the previous
period. On this basis, he challenges any simultaneous procedure that further converts
input and output prices for the same period.
11
Andrew Kliman and Ted McGlone, The Transformation non-Problem and the Non-Transformation
Problem, Capital & Class, 12:2 (1988), pp. 5683.
252 B. Finger

Let us examine this one step further. Consider Klimans proposition under con-
ditions of simple reproduction, where total value (w
0
) is constant, such that w
0
=
c
0
+v
0
+s
0
in the initial period 0. In period 1, c
0
+v
0
are data and s
0
is again given
but redistributed to different sectors of production in accordance with the rule of
prot rate equalization before the next period of production commences. In period
2 the data is c
0
+ Δc
1
+v
0
+ Δv
1
, where the Δ may be positive or negative. Within
each discreet period, the sum of the values and prices of production are equal as
are the sum of surplus-values and prots. So, too are rates of prot, the organic com-
position of capital and the rate of surplus value identical in terms of both value and
price within each cycle. Eventually, through the process of iteration, an equilibrium
rate of prot emerges, whereby c
n-1
+ Δc
n
=c
n
and v
n-1
+ Δv=v
n-1
.
Here, too it would seem that, like Moseley, a consistent, alternative solution has
been found that illustrates Marxs understanding of the relationship of value and
price. Formally, this proceeds, unlike Moseley, along similar lines as other iterative
solutions. But these solutions have been wrongly interpreted, according to
Kliman, for comparing period 0 prices with period n prices. They differ because of
the incorporation of additional value into the social capital in the interim such
that w
n
>w
0
. But this interim is, in effect, a preliminary accumulation of capital
that establishes the precondition for simple (or expanded) reproduction to
advance along an equilibrium path in terms of values and prices.
But heres the rub. Kliman, like Moseley, wanders into the trap of a command
theory of labor value, which is inconsistent with the law of value. It cannot be
helped, which is why Böhm-Bawerk was onto something in his insistence that chan-
ging the conditions of exchange upends Marxs theory of value. Once commodities
no longer exchange according to the labor-time required for their production, the
prices that represent their values are determined instead by the value of the quantity
of the money material against which they exchange.
In volume I of Capital and elsewhere, the time worked to produce commodities
equals the time worked to purchase these commodities. Gold becomes the
measure of value, because all commodities measure their exchange values in gold,
in proportion as a certain quantity of gold and a certain quantity of the commodity
contain the same amount of labor-time; and it is only by virtue of this function of
being a measure of value, in which capacity its own value is measured directly in
the entire series of commodity equivalents, that gold becomes a universal equivalent
or money (Contribution to the Critique, p. 75). In a society characterized by a capi-
talist division of labor, the exchange of commodities is in essence the exchange of
equal quantities of social labor mediated by money. Or so Marx initially argues,
both in the aforementioned Critique and in Capital I.
But Marx violates this understanding by inadvertently accepting a dual ledger
system of values to account for a pricing concept that he inherited from classical pol-
itical economy: (labor) values as measured in production time, which are pro-
portional to monetary prices; but (command) values which discard labor values
(determined by production time) and substitute for them a measurement as
Critique 253

determined by the labor content of the money commodity against which individual
commodities exchange. The measure of value in the latter is no longer established by
the conditions under which the commodity is produced, but rather by the conditions
of its exchange, its power, in other words, to command other goods.
The apparent consistency in both Moseleys and Klimans arguments refers to the
consistency within the command theory of pricing. The sum of commanded mone-
tized values equals the sum of commanded prices; the sum of commanded surplus
values equals the sum of commanded prots; the command value rate of prot
equals the command price rate of prot. What they cannot do, any more that
could the neo-Bortkiewiczian, neo-Ricardian school, is to reconcile the two contra-
dictory theories of value.
IV. Conclusion
The transformation problem resides in Marxs presentation itself, where no systematic
attempt was made to establish the conditions under which an average rate of prot
could be resolved withi n the framework of his volume I analysis of prices. Instead he
tried to unpack classical political economy where he found it, by accepting their
conclusions, while reassessing their reasoning. He emphasized, the really difcult ques-
tion how is this equalization of prots into a general rate of protbroughtabout,
since it is obviously a result rather than a point of departure? (Capital III, p. 174).
Having posed that question, Marx was logically compelled, once he deferred to the
classical framework, to provide a convincing demonstration as to how surplus value
is redistributed from capitals with relatively low organic compositions of capital and
relatively high rates of prot to capitals with high organic compositions and low rates
of prot. To his credit, Marxlike Kliman sought resolution in a dynamic (tem-
poral) rather than in a static framework.
But rather than solving the problem, he introduced a new unforeseen wrinkle. If
the suggested market dynamics are pertinent to the transformation problem, it is
only by invoking short-term changes in supply and demand relationships, an expe-
dient that Marx, as we have shown, previously considered an inadmissible dodge.
And justiably so. For it is not through creating temporary gluts and scarcities
that prot rates are equalized under competitive conditions. The purpose of capital
movements, to the contrary, is to eliminate such conditions and reestablish competi-
tive prices.
All of which leads back to this. If competitive prices are proportional to socially
necessary labor times, and such prices also have the characteristic of affording the
units of production an average rate of prot, then the solution to the transformation
problem must reside in the internal reconciliation of these two propositions.
The specic market dynamic that Marx blinded himself to is rooted in the varia-
bility of sectoral rates of exploitation, changes that capitalists are compelled to
execute through various means of rationalization and innovation in response to
external protability challenges and restraints.
254 B. Finger

Unfortunately, Marx and his defenders sought to reconcile this by introducing a
deus ex machinamodied values, inadvertently introducing a new exchange prin-
cipal inconsistent with the labor theory of value.
Disclosure statement
No potential conict of interest was reported by the author(s).
Notes on contributor
Barry Finger has contributed to New Politics, Critique, New Left Project, The Bullet,
International Socialist Review and other socialist periodicals. He was a shop steward
and union activist with the Public Employees Federation.
Critique 255