The technology-productivity puzzle has generated a rancorous debate in economics. Many labor economists claim that computer-related investments have led to structural gains in productivity. Others are skeptical that such high-tech investments are behind recent surges in labor productivity. They argue that the relative increase in productivity growth rates since the early 1990s is largely explained by procyclical effects-an expansion of economic activity that leads to increasing returns and higher labor productivity growth rates. This paper examines the Veblenian explanation of how capitalists attempt to enhance their profitability by restraining production and how widespread accessibility to technology can frustrate "capitalistic sabotage," leading to economic expansion and higher labor productivity.
The analysis initiates a search for an alternative way of measuring the intensity or robustness of the economic system that is grounded in the labor market. The need for an alternative measurement stems from both a critique of the commonly used capacity utilization figures and the increasingly positive relation between work hours, economic output, and labor productivity. The interplay among technology, excess capacity, productivity, and work hours is admittedly convoluted, but Thorstein Veblen's analysis of reserve capacity lends illuminating insight into how technological advance may lead to cyclical productivity gains that rival, and possibly dwarf, those of a structural nature.
The investigation begins with a brief overview of the theoretical debate over the source of productivity improvements as well as the problematic nature of productivity measures. The following section explores Veblen's reserve capacity discussion as an explanation of why economic "slack" systemically exists within the productive capacity of market-based economies and suggests that labor productivity growth may be best accomplished through the expansion of economic output. The Veblenian perspective makes it possible to describe how technological advancement may reduce reserve capacity and lead to a tightening of the economic system that indirectly fosters labor productivity growth. The final section examines the usefulness of capacity utilization figures in measuring the "tightness" or "robustness" of the economy and suggests an alternative measure of economic intensity that places more emphasis on a variety of labor market measures such as work time.
Procyclical Productivity Growth
Growing controversy has surrounded the question of whether recent productivity gains are driven by computer-related investments or other unplanned influences such as greater capacity utilization. Kevin Stiroh contended that the productivity resurgence is widespread and claimed to observe a strong link between pre-1995 computer capital accumulation and post-1995 productivity growth across U.S. industries (2001). Karl Whelan likewise argued that when depreciation methods are adjusted to reflect the rapid value loss of computer equipment, the contribution of computer capital accumulation to economic growth is substantially larger than standard estimates (1999). The Bureau of Labor Statistics stated that "the recent growth in productivity appears to be more structural than cyclical ... new technology has engendered a pronounced rise in rates of return on high-tech investment, which has led to a stepped-up pace of capital spending and increased productivity growth" (2000).
By contrast, Robert Gordon contended that the computer revolution has not substantially influenced productivity outside of the manufacturing efficiency gains achieved within the computer industry itself (1999). In fact, labor productivity has not even returned to its most triumphant levels. Gordon found that "the recovery over the last three years (1996-1999) does not bring the economy back to the pre-1970 'golden age,' but appears to recover roughly two-thirds of the lost ground" (3). Gordon further contended that the productivity rebound "can be explained by three factors, (1) improved methods for measuring price deflators, (2) the normal procyclical response of productivity in periods like 1997-99 when output grew faster than trend, and (3) the explosion of output and productivity growth in durable goods, entirely due to the production of computers" (4). Gordon's second explanation, that of procyclical productivity, suggests that engineering a robust economy may be as fruitful in boosting labor productivity as expensive investments in technology or the various other forms of human capital that have infatuated labor market observers of late. When coupled with Veblen's contention that firms routinely possess reserve capacity, the role of robust economic activity takes on a new prominence in the labor productivity debate.
Although macroeconomic studies of aggregate output have recently ascribed greater importance to procyclical productivity, the notion has rarely spilled over into the direct analysis of total and labor productivity. In their model of economic growth, Federal Reserve economists John Fernald and Susanto Basu acknowledged that "productivity is procyclical ... that is, whether measured as labor productivity or total factor productivity, productivity rises in booms and falls in recessions" (1999, 5). Yet the authors contended that procyclical productivity has received little attention in the study of business cycles and that cyclical variations in utilization have not been viewed as central to the understanding of business cycles. With the exception of Gordon's analysis, formal studies of labor productivity have been equally remiss in emphasizing the cyclical phenomena. (1)
In neoclassical theory, stiff competition in product and factor markets forces firms to extract comparable levels of labor productivity from the work force. Through the discipline of the market, labor is reified into a homogenous commodity. Any hoarding of labor or industrial capacity is irrational in the neoclassical paradigm as it adversely affects firm profits. Consequently, productivity econometricians have almost axiomatically considered the effort of labor or variations in the workweek of labor and capital as either unobservable or inconsequential.
Social models of efficiency, however, have incorporated a procyclical approach to labor productivity that is not solely dependent on longer hours. These models emphasize that when utilization rates are below the level for which factories and job functions are designed, inputs cannot be employed effectively and low productivity will tend to be propagated through the economic system (see Bowles, Gordon, and Weisskopf 1983). As underutilized capacity tends to dampen new capital investment and technological innovation, the path-dependency effect of a slack economy will tend to slow down the rate of growth of productivity for years in perpetuity. The same hysteresis applies when the direction is reversed and utilization rates are above the trend. Social models of productivity thereby suggest a reverse causation from output to labor productivity--rather than the traditional belief that greater labor productivity is a principal source of more industrial throughput. At a minimum, feedback effects closely relate the scale and intensity of production to the rate of growth of labor productivity.
The management of production and the attendant impact on labor productivity is influenced by a myriad of factors. In the last two decades, labor productivity studies have primarily focused on technological issues and the innate productivity of individual workers. For instance, human capital theory is a familiar example of the efforts devoted to improving the inherent productivity of individual workers. Although issues of human capital and worker training are important, the attention given to such worker-specific issues has arguably led to a neglect of the employer-related causes of low productivity. The Veblenian notion of capitalistic sabotage provides a rationale for why firms may arrive at an output level that is optimal from the firm's pecuniary prospective but subaltern from a social or industrial viewpoint. Veblen's analysis underscores the importance of employer decisions concerning the scale, scope, and intensity of production and suggests that captains of industry may have greater culpability when it comes to keeping workers productive.
The Meaning and Measure of Productivity
Due to the controversy and confusion surrounding the topic, labor productivity has become a threadbare term in economics. Although many understand the importance of labor productivity growth in raising the material standard of living in a community, far fewer people are conversant in the estimation methodology and interpretation of labor productivity measures. Labor productivity is often used to describe physical-output-per-worker-hour. Yet, in our increasingly service-oriented, high-tech economy, the dollar value of physical-output-per-worker-hour seems to be a more useful measure of labor productivity. As such, any otherwise unspecified use of the term "labor productivity" in this discussion refers to the "dollar-value-of-physical-out-put-per-worker-hour."
More confusion is introduced by the circular fashion in which productivity is typically measured and discussed. The received wisdom in economics teaches that growing labor productivity is the only way to achieve real, long-lasting increases in output and living standards. Yet, if labor productivity is the dollar-value-of-physical-output-per-worker-hour, is labor productivity determining output, or vice versa? At the outset, the circular reasoning inherent in the very definition of the leading productivity measures suggests that robust economic activity is the best way to stimulate greater labor productivity and that workers are largely at the mercy of policy makers who either foster or arrest that activity.
Even attempts to discern...