|
The S-Curve of Technology Application A recent article by Jerry L. Jordan (2001), "Riding the S-Curve: Thriving in a Technological Revolution," Economic Commentary, Federal Reserve Bank of Cleveland, January 1 (Available on Web at www.clev.frb.org/research), presents some very familiar discussion on the process of creative destruction. Jordan is the President and CEO of the Federal Reserve Bank of Cleveland and a frequent writer on the topic of economic growth. Jordan describes some previous technological revolutions, including the spread of electricity that we discuss in the Case Study 6-1 in Chapter 6. With regard to the new communications technologies, Jordan positions us at the point where we are about to begin enjoying huge productivity gains. Writes Jordan: ...if history repeats itself, we now stand on the cusp of enjoying the transformation's most substantial gains. Consider that it has been about 25 years since the development of the microprocessor, halfway through the process as indicated by historical experience, and only recently have we seen an explosive expansion of the U.S. capital stock. As in past episodes, this new capital did not initially yield the productivity growth it seemed to promise. Between 1975 and 1994, annual U.S. productivity growth fell from about 3 percent to 1.5 percent. But it is increasingly clear that productivity growth is now rapidly gaining momentum again. In the past two and a half years, nonfarm business productivity has grown at a pace of approximately 3.5 percent annually–its best showing in about 30 years. And if past technological revolutions are a good indication, this may be only the beginning. If Jordan is correct, the information technology revolution has yet to provide us with its full benefits in the U.S. Few other countries have moved as far up the S-Curve as the U.S. has. |
Most developing economies have yet to begin the ascent of the S-Curve. The Economist Technology Quarterly The newsweekly The Economist has launched an interesting new supplement that will appear on a quarterly basis, called The Economist Technology Quarterly. The introduction to the first supplement, which was part of the December 9, 2000 issue of The Economist, was appropriately entitled "In Praise of Disruption." "The Economist Technology Quarterly will offer readers a foretaste of what new developments are threatening–no, guaranteeing–to disrupt the way business is done in the years ahead," the introduction promises. In recognition of the Schumpeterian process of innovation, the introduction continues, "...disruption is as much about opportunity as it is about destruction. Joseph Schumpeter, an Austrian-born economist best remembered for his notions of ‘creative destruction,' would be applauding the way that even the biggest computer manufacturers, chip makers, drug firms and telephone companies are feeling the heat of feisty upstarts and are having to reinvent themselves continuously to stay alive." The first issue covered, among other things, internet privacy, soft radios, extreme programming, teraflops from cyberspace, Bluetooth and 3G, wireless communications technologies, digital ink and interactive t-shirts, micromachines, the human genome, and a petaflop computer. If these things fascinate you, look up the The Economist Technology Quarterly, which is available at www.economist.com free of charge (Click on "Science and Technology on the left side of the home page and then scroll through each article). |
|
The Black Box
Recall our discussion at the beginning of Chapter 6 about the "black box." We encouraged you to
look at technology as a black box into which we put all of the economy's productive resources and
out of which emerge all of the welfare-enhancing goods and services that determine our standard of
living. We borrowed the description of technology as a "black box" from the title of Nathan
Rosenberg's (1994) book, Exploring the Black Box, Technology, Economics, and History,
Cambridge University Press, Cambridge. Because the black box represents the economy's production
function, it becomes clear that "technological progress" refers to anything that shifts the
production function upward. The idea that the level of technology in the economy depends on everything that influences how the economy is able to transformthe quantity of machines, labor, and resources at its disposal into welfare-enhancing output is still often not appreciated even by economists. For example, in a recent article Huagang Li (2001), "The Relative Importance of Effort, Organization, and Technological Change in Chinese Factories," Contemporary Economic Policy, Vol. 19(1), pp. 99-108, concludes: Data collected from Chinese state factories reveals very rapid value added growth despite little investment and a shrinking labor force. A standard computation points to rapid TFP growth rates. However, a more careful analysis suggests that the technological level of the state factories, at least the portion not collinear with effort and organization, did not improve significantly during the 1980-1991 period. Competitive pressure of a thriving market economy pushed state-sector workers to work much harder; the sudden labor mobility due to reforms also yielded significant efficiency gain. Together, these two effects almost accounted for all the TFP growth observed. Considering that nearly all Chinese reform policies emphasized the incentive for the workers and very little was done to boost the industry technologically, this result is not unexpected. (P. 107) Huagang Li's study is actually very well done and quite relevant to the discussion on technology in Chapter 6. But note how Li carelessly suggests |
that the economic reforms that
increased effort and efficiency are somehow different from technological progress! Improved
economic incentives are just as much technology as a more efficient machine or an improved
product design. Institutional reform is technological progress too. Learning by Doing: New Research The Case Study 6-4 on the Liberty Ship Program presents the classic example of learning-by-doing. It was also studied at length because it resembled a controlled experiment: A number of shipyards built an identical product for a single customer, the U.S. government, who kept detailed records on production in order to calculate the cost-plus prices that it paid for the ships. Of course, there were many similar wartime industries, and the obvious question is whether those industries also exhibited similar gains from learning. In fact, learning also took place in wartime plants producing aircraft, submarines, jeeps, rifles, ammunition, and the many other armaments whose production was greatly expanded over a short period of time. The study by Kazuhiro Mishina (1999), "Learning by New Experiences: Revisiting the Flying Fortress Learning Curve," in Naomi R. Lamoreaux, Daniel M. G. Raff, and Peter Temin, (eds.), Learning by Doing in Markets, Firms, and Countries, Chicago: University of Chicago Press, pp. 145-184, examines the gains in productivity over the production run of 6,847 B-17 "Flying Fortress" bombers by the Boeing Company in its Seattle, Washington plant from 1942 to 1945. Mishina interprets the data on productivity costs as being related to the expansion of the scale of production; when production slowed down toward the end of the war, learning stopped. The increase in labor productivity was not the result of increasing returns to scale, because productivity did not decline when production was produced, it merely stopped growing. Mishina also concludes that it was not the workers that learned because the workers |
|
changed frequently over the 4-year production run of the B-17. In fact, on average
workers
became less experienced rather than more experienced toward the later phases of the production
run as experienced workers were drafted into the armed forces and replaced by new workers, most
of whom had never worked in factories of any type before. Nor did improved equipment have much
to do with the improved productivity, as may have been the case of the Liberty Ship program;
there was little additional equipment installed after production began in 1942. According to
Mishina: The agent of learning is the core managers of control functions in the plant, that is, those who coordinatevarious aspects of the plant operations to ensure that work in progress flows smoothly without interrupting events so the shop manager can concentrate on his or her job of meeting production schedules while supervising the direct workers. (P. 175) Interestingly, Mishina also compares the production of B-17s by Boeing to the production of the B-24 by Ford Motor Company: Ford introduced its patented assembly line methods to aircraft manufacturing, and failed to show anywhere near as great a gain in productivity. Hence Mishina's interpretation that the productivity gains came from improved organization of production, which Ford was not able to carry out because it fit production into its established ways of doing things. In his "Comment" on Mishina's article, Russ Thomson points out that "Productivity growth is especially rapid when the new commodity comes to be produced in a radically different kind of production process." (P. 183). Ford had introduced such a radically different production process for automobiles earlier in the century, but it failed to do so for aircraft. Boeing did introduce a new flexible production method that it later used to become the industry leader in commercial aircraft. Note: Mishina's article is just one of many in a book that should interest anyone seeking to delve further into learning by doing at the firm, industry, or economy-wide levels, Naomi R. Lamoreaux, Daniel M. G. Raff, and Peter Temin, (eds.), Learning by Doing in Markets, Firms, and Countries, Chicago: University of Chicago Press. |
These studies make it clear that doing is not automatically accompanied by learning. Also,
the learning comes in many forms. The constant learning parameter that we assumed in the
equations 6-12 through 6-15 are meant to help you understand the basic mechanism through which
learning-by-doing can generate economic growth, the learning process is not a constant in the
real world. Also, don't forget the interesting website on the Liberty Ship program maintained by Peter Thompson at the University of Houston: www.uh.edu/~pthompso/liberty/liberty1.html |