Subject: [OPE-L:1866] Re: Re: Re: A Review of Lapides' Marx's Wage theory
From: Patrick L. Mason (pmason@garnet.acns.fsu.edu)
Date: Wed Dec 08 1999 - 13:40:12 EST
Ajit has done a great job of asking two important questions regarding
Marx's (and Marxian) wage theory.
1. "How come a rising rate of unemployment be accompanied with a rise in
real wages within a Marxist
framework?"
2. "In addition to Marx's famous statement in his 1865 lecture (VPP) where
he says '... the general tendency
of capitalist production is not to raise, but to sink the average
standard of wages, or to push the value
of labour more or less to its minimum limit' (p.61); in *Capital 1* the
whole of section 5 of chapter 25
(68 pages in total) is devoted to documenting a declining tendency of
real wages in England (for the period
1846-66) and Ireland (for the period 1860-65). Most interestingly, Marx
puts a lot of stress on the
deteriorating condition of housing for all strata of workers. Since
housing constitutes a fair share of the
real wage basket, the case for a declining tendency of the real wage in
this period is very strong. Meek (1967)
also agrees with our position in general, though he does not explicitly
take into account the evidence I have
alluded to above." (Sinha, Ajit 1998, *History of Economics Review* no.
28, summer, f.n. 1, p. 110)
First, a minor correction. Marx wasn't simply concerned with unemployment,
but with joblessness - that is, both underemployment and unemployment and
nonparticipation. Nevertheless, to re-ask Ajit's question with
"joblessness" substituted for "unemployment" makes it even more unlikely
that real wages will rise.
Second, within the spirit of Ajit's question, I will assume that all
workers are equally skilled or equally unskilled. Without this assumption
one could very easily move from the theoretical perspective that Ajit is
concerned with to an empirical answer. For example, if allow differences in
skill levels then it is entirely plausible that the employment and wage
levels of (say) the least skilled are declined will the wages and
employment levels of the more skilled are increasing. If the position wage
and composition effects among the more the skilled out-weigh the negative
wage and composition effects of the least skilled then the average
unemployment (or joblessness) rate will rise even as the average (mean)
wage rate is also increasing.
Given these two caveats, here are my responses to Ajit's question.
Question 1.
At a given point in time, say during a given business cycle, Ajit is
correct, that is, both real wages and joblessness will not increase. It is
likely that real wages will fall as joblessness increases.
Back to my simple model of Marx's wage theory
1. Average real wage = f(ability to pay, ability to make pay).
With rising unemployment, the ability to make employers pay (bargaining
power or the ability of workers to wage class struggle within the labor
market) will decline or at least be non-increasing. Further, in a deep
recession the ability to pay by employers will also decline.
However, if we examine a series of periods, say 2 or 3 business cycles of 8
years each, it is not necessary for the extent of joblessness to move one
way or the other. It depends on whether the economy is experiencing a
secular upswing, downswing, or secular stagnation. It also depends on
what's happenning with the organization strength of workers, or more
broadly, the social and historical conditions of workers.
Question 2.
Ajit's is suggesting that Marx's argument is
2. Average real wage = f(value of labor power).
But, this isn't correct. Marx's argument is
3. Average real wage = f(value of labor power, social and historical
conditions of workers).
My simple model of Marx (equation 1) is a popularization of equation 3.
So, Ajit is correct that technological change pushes down the value of
labor power. But, technological changes also introduces new goods into
workers' consumption bundle and it changes the condition of labor supply
and labor demand.
For example, technological change (and the accompanying accumulation and
centralization of capital) forced the urbanization of America and other
capitalist states. This process of urbanization and the expansion of the
welfare state and the existence of union induced private pension funds lead
to dramatic changes in family size. Notably, families have been decreasing
in size for at least one hundred years. Urban children cost more than rural
children because they produce less and they consume more. With the advent
of pensions large numbers of children are no longer necessary for old age
income. Children are less necessary for housework when both parents work
because wash and wear clothing (no ironing) and the enormous infusion of
technology into every facet of housework. With few children to care for and
technological improvements in home work instruments (dryers instead of
clothes lines), married women are now more likely to work than remain home.
Hence, although technological change decreases the value of labor power the
real wage real will not fall, unless we further assume that there are no
changes in the social and historical conditions of workers (along with no
changes in the intensity of labor and a number of other factors).
This is my short response. A longer response is attached below.
The value of labor power [VLP]= "the collection of use values [di, i =
1,2,...,n] consumed by workers and the unit values [li]of these use values
[Slidi]."
VLP = Slidi
Each of these is socially determined and the average wage paid in each
industry is derived from this socially determined norm. For example, both
the nature and the extent of workers' consumption bundle are clearly which
related to the bargaining power of labor, which is related to nonmarket
employment opportunities.
It is obvious that [di], the consumption bundle or standard of living, is
socially determined. The average level of productivity combined with the
class struggle between labor and capital as a whole determine [di].
"However, it is not primarily the social nature of the standard of living
of the working class that makes the value of labor power a socialized
variable." (page 71). Given the standard of living, it is socially
necessary abstract labor time of society as whole which determines [li],
the unit values of the use values consumed by workers. These use values
then depend upon the overall technological development of society via the
social productivity of labor in each industry which produces workers'
consumption goods and the means of production used to produce those goods.
"The rate of surplus value exists first for capital as a whole, since both
the working day and necessary labor time are determined at this level of
analysis." (page 72).
There are also two additional factors which enter into the determination
of the value of labor value of labor power: (1) the cost of training the
laborer; and, (2) the labor market participation of labor of women and
children, which "makes a great difference in the cost of maintaining the
family of the laborer, and in the value of the labor-power of the adult
male." (Capital, volume I, page 569).
VLP = Slidi + ljtj + g, where tj represents the commodities and living
labor used to train workers and g represents the extent of labor market
participation of women and children.
Given the value of workers' consumption bundle, the cost of training, and
the extent of labor market participation of women and children, the
relative magnitudes
of surplus value and the price of labor power (the wage rate) are
determined by (1) the productiveness of labor; (2) the intensity of labor;
and, (3) the length of the working day.
WAGEk <== f(VLP, labor intensity, productivity, length of the workday)
(wage rate for firm k)
A workday of a given length and average intensity always creates the same
amount value, regardless of the productivity of labor. Hence, a ceteris
paribus increase in productivity of labor will lower the value of labor
power (VLP) (if the productivity increase is for a wage good) and raise the
mass of surplus value (S). It is precisely the reduction in the value of
labor power which raises the mass of surplus value. (The amount of the
reduction in V is exactly equal to the increase in S, but DV/V ¹ DS/S).
However, the decrease in the value of labor power consequent upon an
increase in productivity does not imply that the price of labor power
(wage) will decline by the same amount of even decline at all. The lower
limit of the potential decline in the wage is determined by the decline in
the value of labor power; however, the actual extent of the decline is
determined by the bargaining power of workers.
Initial Situation New Situation
8 hour workday 8 hour workday
(4 hours necessary labor, (4 hours necessary labor,
4 hours surplus labor) 4 hours surplus labor)
Output = 80 units Output = 160 units
l = 8hr/80q = 0.10 hr/q l = 8hr/160q = 0.05 hr/q
lg = 2 hr/oz lg = 2 hr/oz
p = l/lg p = l/lg
= 0.10(hr/q)/2(oz/hr) = 0.05(hr/q)/2(oz/hr)
= .05 oz/q = $5.00/q = .025 oz/q = $2.50/q
(1 oz of gold = (1 oz of gold =
2 hr of labor = $100) 2 hr of labor = $100)
VLP = l40q = 0.10*40 = 4hr VLP = l40q = 0.05*40 = 2hr
V = 4hrs = 40q = $200 V = 2hrs = 40q = $100
S = 4hrs = 40q = $200 S = 6hrs = 120q = $300
total value = 8hr = $400 total value = 8hr = $400
With sufficient bargaining power, workers may be able to keep their wages
from falling to $100. The rate of exploitation (S/V) in the old and new
situations is 1 and 3, respectively. If workers are able to keep the rate
of exploitation at 1, then their wages in the new situation will remain at
$200. Clearly, in this instance, wages have exceeded the value of labor
power. Note however even if the price of labor power were to fall
completely to its new lower limit, i.e., fall to $100, the amount of output
that laborers would be able to purchase is unchanged. In both cases, the
extent of the means of subsistence is unchanged. The fall in the value of
labor power is accompanied by a decrease in the price of output; hence, the
total mass of output workers are able to purchase in unchanged. Hence, an
increase in productivity can provide the space for both an increase in what
is commonly called the real wage (the money wage divided by the price level
= units of output workers are able to purchase) and an increase in the rate
of surplus value, even though the value of labor power declines with an
increase in productivity (with constant labor intensity and constant length
of the workday); whether real wage increases or remains constant however
depends on the relative strength of labor and capital.
An increase in labor intensity (above the norm) represents more abstract
labor per unit of the day. Say the intensity of labor doubles. In this
case, one workday represents 16 hours of abstract labor time rather than 8.
If output also doubles, the unit value of output remains the same. (Unlike
the case of an increase productivity where the unit value of output would
be reduced by half). Hence, individual prices do not fall and the total
money received from the sell of output will be doubled. Of course, with
twice as much revenue wages can also increase but whether or not they do
depends on the bargaining strength of labor. An increase in wages however
does not necessarily imply that the price of labor power is above the value
of labor power. "On the contrary, the rise in price may be accompanied by a
fall in value. This occurs whenever the rise in the price of labor power
does not compensate for its increased wear and tear." (page 575) If then we
take an above average intensity of labor as an unpleasant working
condition, Marxian economics does not suggest that workers will receive a
compensating wage differential -- unless they are able to extract this
differential because of greater bargaining power. Moreover, the maximum
potential wage consequent upon an increase in labor intensity may not be
sufficient to compensate for the increased wear and tear on workers.
Finally, one can see that all firms will have a constant incentive to
increase labor intensity -- especially those firms utilizing the most
backward means of production.
If the intensity of labor increases equally in all industries, this new
higher level of intensity would become the normal and cease to have an
impact on changes in the price of labor power and the amount of surplus
value. "But still, even then, the intensity of labor would be different in
different countries, and would modify the international application of the
law of value." (page 576).
Changes in the length of the workday also affect the extent of surplus
value and the price of labor power. Shortening the workday leaves the value
of labor power unchanged and therefore necessary labor time is unchanged,
but surplus labor time is decreased. Capital will be worse off unless they
are able to lower the price of labor power. An increase in the length of
the workday (if it isn't too great) will provide space for a rise in the
price of labor power along with magnitude of surplus value can increase.
However, an increase in the length of the workday may cause the value of
labor power to fall below the price of labor power, even if the price of
labor power has remained unchanged or risen slightly. The longer workday
implies an above average amount of wear and tear on the worker. This
greater wear and tear can be compensated for by higher wages, up to a
certain point. However, "beyond this point the wear and tear increases in
geometrical progression, and every condition suitable for the normal
reproduction and functioning of the labor power is suppressed. The price of
labor power and degree of its exploitation cease to be commensurable
quantities." (page 578)
Summarizing, Marx's argued that international differences in wage rates
(and hence also intra-national wage differentials) may occur because of
differences in :
1) the price and extent of the average worker's household consumption
bundle as historically and socially determined; 2) the cost of training the
laborer; 3) the part played by the labor of women and children; 4) the
productiveness of labor; 5) the intensity of labor and the length of the
workday; 6) nonmarket employment opportunities.
We have shown that wages, the price of labor power, in Marx' s model are
regulated by the value of labor power (Marx, 1906, 1977a, 1977b). In turn,
the value of labor power is determined by the value of the commodities that
are necessary to sustain the laborer and her family at the socially and
historically given level of subsistence.
Consistent with classical analysis, the wage rate does not reflect the
relative scarcity of labor in the economy; it represents the costs of
reproduction of labor power--which are inseparable from the costs of
reproducing the worker. Although the average wage rate cannot exceed the
average rate of productivity (for a sustained period of time), there is no
automatic relationship between the general wage rate and the average level
of productivity; an increase in productivity will secure a wage increase
only if workers are sufficiently organized to force capital to grant an
increase in compensation. Additionally, recent developments in economic
theory have also shown that there is no monotonic relationship between
individual wage rates and individual rates of productivity (Sraffa, 1960;
Eatwell, 1983; Nell, 1980).
However, the average level of productivity does provide an upper bound for
the general wage rate. One effect of technological change is to raise this
limit and thereby create additional space for the general wage rate to rise.
So, the value of labor power determines the average wage rate;
fluctuations in supply and demand cause the market price of labor to
oscillate around the long run supply price of labor power. Inter- and
intra-industry expansions or contractions in output are responsible for the
allocation of workers across different capitals in the economy.
This archive was generated by hypermail 2a24 : Sun Dec 12 1999 - 15:45:03 EST