ABSTRACT
Contemporary industrial capitalist economies have come to expect consistent significant increases in labor productivity, and a roughly proportional sharing of productivity gains between labor and capital incomes. This expectation is the ground on which much of the political economic drama in these societies plays out. These presumptions are so strong that even relatively small variations in the rate of growth of labor productivity or the wage share in national income are the object of intense scrutiny and debate. These features of contemporary capitalism would have surprised Thomas
Malthus and David Ricardo. They believed strongly in the pervasive importance of diminishing returns to labor and capital with capital accumulation in the face of fixed land resources, and thus would have expected advanced industrial capitalist societies to confront a chronically diminishing, rather than a consistently rising, average productivity of labor. They also saw no sustained basis onwhich themajority workers could bargain for higher (real) wages in the face of pervasive competitive pressures from overpopulation and immigration. Karl Marx, on the other hand, foresaw the tendency for capitalism to
be a technologically progressive mode of production, with pervasive pressures toward labor-saving innovation in production (Marx, 1976, ch. 12; 1981, Part Three). Ricardo had shown (in his chapter on Machinery) that competitive pressure on individual capitalists to lower costs could lead to a general increase in labor productivity through the substitution of machinery for labor, but seems to have regarded this process as offering only a temporary relief from the pressure of diminishing returns to both capital and labor due to limited land resources. Marx developed this mechanism of induced
innovation into a general theory of the tendency of capitalist production to increase labor productivity through the replacement of “living” by “dead” labor, that is the displacement of workers by machines. Through this process Marx saw capitalism as fulfilling its historicmission of developing the forces of production to the point where age-old problems of material scarcity could be eliminated. In his earlywritingsMarx sharedMalthus’ andRicardo’s pessimism about
the ability of workers to achieve substantial increases in wages within the framework of capitalist social relations, though he emphasized different mechanisms as being responsible. If labor productivity had risen continuously without a proportional rise in wages, capitalist society would soon have faced the revolutionary crisis Marx pinned his hopes on. A constantly growing disparity between the productive power of labor and the standard of living of workers would provide a powerful impetus to the project of socializing the surplus product. Bourgeois society could absorb the rapidly growing social surplus only through a constant increase in conspicuous capitalist consumption, which would reinforce the social resentments latent in the class divisions of the society, and make the stabilization of democratic political institutions problematic. A socialist regime coming to power under these circumstances would find it relatively easy to arrange for a steady rise in workers’ standard of living from a very low base, on the basis of a high productivity of labor. Both the motivation for and the feasibility of socialist revolution would have grown had the pattern of rising labor productivity and stagnant wages Marx foresaw in his early writings come to pass. Even as Marx was preparing his critique of political economy for publication as Capital in the 1860s, however, there were clear signs of substantial increases in wages in leading capitalist economies, particularly Britain. Marx saw that this historic development could undermine his political project, and struggled for the rest of his life to come to terms with it both theoretically and politically. Curiously, it is Adam Smith, the earliest of the great classical economists,
whose lifetime offered the smallest experience of full-blown industrial capitalism, who would have been least surprised at the emergence of continuous rises in labor productivity and parallel proportional increases inwages. Smith centered his analysis of productivity on the widening division of labor (see Smith, 1937, ch. I and VIII). Although he characteristically evades a sharp confrontation between the increasing returns to the accumulation of labor
and capital inherent in the division of labor and the diminishing returns due to limited land resources, Smith seems to have believed that the division of labor could predominate for a long time. Smith’s theory of wages is equally delphic, but he does conjure up the comforting vision of rising wages in a progressive capitalist society in which accumulation is steadily increasing population and output. Modern political economy provides at least one powerful explanation of
why industrial capitalist economies achieve systematic increases in labor productivity and stabilize wages as a proportion of national income. If the rates of increase of labor and capital productivity depend on the relative shares of labor and capital in the costs of production, there is a powerful feedback mechanism linking distribution to productivity increases. If the wage share, for example, rises, the rate of increase of labor productivity would also rise, tending to reduce the demand for labor and putting downward pressure on the level of wages. This feedback would act as a kind of social thermostat to stabilize the wage share at the level at which the induced increase in capital productivity is close to zero, which is a necessary condition for a stable relationship between accumulation and growth of the labor force. This theory has remarkable implications. It suggests that the wage share in the long run is completely independent of the forces of capitalist thrift leading to accumulation and of labor force growth, being determined entirely by factors relating to the bias of technical change. In this sense it provides a sharp alternative to the neoclassical vision of distribution reflecting the social scarcities of inputs to production. While this theory has great explanatory power, it also raises important
unresolved questions about the foundations of the theory of induced technical change. The classical distinction between labor, capital, and land inputs to production rests on their quite different conditions of reproduction, that is, in modern terms, on their different mechanisms of supply. The theory of induced technical change, on the other hand, rests on the pressures on individual capitalists to reduce costs in general. The individual capitalist has no reason to distinguish labor, capital, and land inputs to production, since they all appear simply as elements of cost. Why, then, should the rates of increase of productivity of labor and capital respond particularly to their shares in costs? The answer may lie in the generalizability of techniques that increase the productivity of any particular form of labor to other forms of labor, since all labor is the productive effort of conscious human beings.
The puzzle of distribution Wage levels are the outcomes of implicit or explicit bargaining between individual workers and employers. Looking at this process from the worker’s subjective point of view it is not hard to understand why the classical economists thought wages would tend to be pushed down to some minimum level. Workers in capitalist societies typically feel insecure, feel that they face vigorous competition from a large number of other equally-or better-qualified workers, and as a result tend often to accept whatever wage is on offer. The wage bargain, as Marx is at pains to point out in his analysis of laborpower as a commodity (Marx, 1976, ch. 6), is not an agreement between the capitalist and worker to share the value added by the worker’s labor. The worker surrenders control over her or his capacity to produce for a set period of time in exchange for the wage, regardless of how well the capitalist actually succeeds in turning the capacity to produce into a product and sales revenue. Thus, there is no reason to think that capitalist employers will automatically reward workers for higher productivity with higher wages. Furthermore, workers form a large and dispersed group that faces major
difficulties in controlling the boundary conditions of the labor market. A rising wage easily attracts potential workers from other activities, such as subsistence farming, and from other regions and countries into the labormarket, as Marx emphasized in his discussion of reserve armies of labor. While the Malthusian mechanisms of fertility and mortality play a much smaller role in regulating the supply of labor-power to an industrial capitalist society than in sustaining a demographic equilibrium in pre-capitalist societies, the world population explosion that followed the rise of industrial capitalism reminds us of their latent force. The Classical political economists were hard-headed enough to reject the
idea that social solidarity, morality, or good-will would do much to ensure that workers as a class shared in general rises of labor productivity. Even if some individual capitalists were to make a practice of sharing productivity gains with their workers, they would quickly find themselves at a disadvantage with respect to even marginally less generous or responsible competitors. Thus, Malthus, Ricardo, and Marx looked for some ultimate floor to
the wage, with the expectation that wages would normally be forced down to this floor. Malthus wanted to locate this floor in biological terms, in
the minimal standard of living required for the successful reproduction of workers as a class. Marx, following Smith, puts more emphasis on the “social and historical” factors mediating the material requirements for the reproduction of workers, without explaining very clearly how these social and historical factors actually work. Smith acknowledges the notion of a minimum subsistence standard of living as a wage regulator, but also seems to believe that wages might be pulled up well above this level indefinitely in an expanding and prosperous capitalist economy. To what mechanisms, then, can we attribute the actual historical expe-
rience in industrial capitalist societies of wages rising roughly at the same rate as labor productivity, so that wage shares have remained roughly constant over long periods of time? Three fundamental mechanisms suggest themselves. First, there might be a systematic tendency for the subsistence minimum
standard of living of workers to rise along with labor productivity, due to rising requirements of training, health, and social skills at higher productivity levels. It is hard to imagine, for example, ragged and intermittently starving workers running sophisticated modern technology. But the experience of “newly industrializing countries” in the last half of the twentieth century shows that the range of worker standards of living compatible with advanced technology is very broad. There is also the question of the degree to which technology has been shaped to workers’ styles and conditions of life rather than the reverse. Second, workers’ efforts to organize themselves politically and economi-
cally through political parties and unions to control the boundary conditions of the labor market might give them enough leverage over wages to secure a claim on a proportion of productivity gains. The immense political struggles over the right to unionize and its limits in advanced capitalist economies suggest that both capitalists and workers perceive this as a critical social dynamic. But the degree and influence of unionization varies immensely even among the advanced industrial capitalist societies, much more than the wage share or the elasticity of the wage with respect to labor productivity. A sharp increase in the wage share in the course of rapid capital accumulation has been a common experience in many newly industrializing countries with weak or repressed labor rights, such as South Korea. Third, the forces of capital accumulation might be so strong as to tend
constantly to outrun effective supplies of labor-power, thus forcing wages to rise to ration excess demand for labor. This Smithian mechanism is the
simplest andmost plausible of the explanations, but carries with it subtle and difficult questions of interpretation and ideology. Even advanced industrial capitalist economies tend to operate with a significantmargin of unemployed labor, which can balloon up for long periods to large absolute numbers of unemployed. This observation tends to argue against the assumption that full employment or labor scarcity plays the dominant role in mediating wage increases. This explanation tends all too easily to slide into an apology for capitalist social relations in the form of “trickle-down” economics, the argument that strengthening profitability and capital accumulation are the best way to advance the interests of workers as a class. The three mechanisms are far from mutually exclusive, and in fact tend
strongly to reinforce and interact with each other. A tight labor market provides a favorable ground for union formation and bargaining. Unionization tends to professionalize the workforce and create pressures to build higher standards of living into workers’ expectations and self-image. On the other hand, a sharp rise in the wage share in income reduces the ability of capitalists to accumulate and sets in motion a process of weakening labor demand and increasing unemployment. In periods of rapid accumulation the third mechanism, labormarket scarcity, may play the key role in advancingwages, which get built into historical and social expectations of workers’ standards of living. In periods of slumping accumulation, the difficulty capitalists face in renegotiating wage levels with their ongoing employees and the resistance of unions may significantly slow the fall in wages. Marx, without denying the importance of “over-accumulation” of capital
and a consequent rise in wages as one aspect of the capitalist business cycle, argued that over the long period deeper forces regulated the size of the reserve army of labor and the rate of exploitation (the ratio of the profit to the wage share in income) (Marx, 1981, ch. 13). A rise in the wage share, in Marx’s analysis, tends to be self-limiting because by reducing profitability it reduces the rate of capital accumulation and hence the growth in the demand for labor. In addition, a rise in the wage share tends to hasten the growth of labor productivity and thus create more technological unemployment, renewing the reserve army of labor. (We will return to this theme below.) The inverse forces tend to correct a fall in the wage share as well over the long period, though Marx tended to put less emphasis on this implication of his analysis for obvious political reasons. The systematic explanation of wage movements both over the long period
and over the business cycle is a key problem for the modern development of
Classical and Marxian political economy. Despite the existence of a lot of research examining this problem in specific historical situations, the political and policy effectiveness of these ideas has been hobbled by the reluctance of the Classical school as a whole to acknowledge the pervasive positive effect of the accumulation of capital and periodic scarcity of labor-power onwages. Putting this mechanism in its appropriate place in a synthetic understanding of wage dynamics is critical to forwarding Classical and Marxian positions.
The Goodwin model In analytical terms the most elegant expression of these dynamics in the Classical/Marxian literature is Richard Goodwin’s model of the capitalist labormarket (Goodwin, 1967). Denoting employment byN and the potential labor force by L, Goodwin assumes that the rate of change in the wage per worker, w, will be proportional to the excess of the employment rate e = N/L over an institutionally given level e0, at which the real wage would not change. Writing wˆ = w˙/w = (1/w)(dw/dt) for the growth rate of the wage, and δ for the factor of proportionality:
