2nd Draft March 2005
The ideas of economists and political philosophers, both when they are right and when they are wrong, are more powerful than is commonly understood. Indeed the world is ruled by little else. Practical men, who believe themselves to be quite exempt from any intellectual influence, are usually the slaves of some defunct economist. Madmen in authority, who hear voices in the air, are distilling their frenzy from some academic scribbler of a few years back.
John Maynard Keynes (1883–1946),
The General Theory of Employment, Interest and Money, 1936, 383
1. In the Standard Model of economics, technological change refers to the effect of new knowledge on the production function of a firm or nation. The nature and source of the knowledge is not a theoretical concern; only its effects on the production function. However, new knowledge has many sources and varying effects. It may be productive increasing output on the shop floor; it may be managerial reducing costs or increasing sales; or, it may be entrepreneurial realizing a vision of future markets, products and/or other opportunities. It may flow from the natural and engineering sciences (physical technology), the humanities and social sciences (organizational technology) or the Arts (design technology). In economic theory, however, it does not matter what form new knowledge takes; it does not matter from whence it comes; the only thing that matters, in terms of calculatory rationalism, is its impact on the production function.
2. In response to technological change,
the production function for existing output may shift upwards or downwards, i.e., technology can be lost as happened
with the fall of
3. In the 20th century, technological change became recognized as the most important source of economic growth, i.e., increase in output – absolutely, or, per capita. Our understanding of such change, however, remains limited. We do not fully understand why some things are invented and others are not; why some are successfully innovated and brought to market, and others are not. The contribution of technological change has, in theory, traditionally been treated as a ‘residual’, i.e., after measuring total growth of output, the contribution of an increased quantity and quality of capital, labour and natural resources are factored out and the residual is called technological change. Technological change, in this sense, is a residual amounting to an error term, or, a measure of our economic ignorance. [AA]
4. Furthermore, in the Standard Model technical knowledge is treated as a public rather than a private good. Thus when new knowledge is published or otherwise made known, others cannot be easily excluded from acquiring it, i.e., it is non-excludable. Furthermore, when shared, the quantity of knowledge is not reduced, i.e., it is non-rivalrous. In fact, the more knowledge is shared, the more knowledge is created. In this sense, knowledge exhibits increasing returns to scale.
5. The effects of technological change in the Standard Model on the production function are conventionally broken out into two dichotomous but complimentary categories: disembodied & embodied; and, endogenous & exogenous technological change.
1. Disembodied technological change has implicitly dominated economic thought since the beginning of the discipline. It refers to generalized improvements in methods and processes as well as enhancement of systemic or facilitating factors such as communications, energy, information and transportation networks. Such change is disembodied in that it is assumed to spread itself out evenly across all existing plant and equipment in all industries and all sectors of the economy.
2. Conceptually embodied technological change traces back to Adam Smith’s treatment of invention as the result of the division of labour (1776). It refers to new knowledge as a primary ingredient in new or improved capital goods. The concept was refined and extended by Marx and Engels (1848) in the 19th and in the 20th century by Joseph Schumpeter and his concept of ‘creative destruction’ (1942). No attempt was made, however, to measure it until the 1950s (Kaldor 1957; Johansen 1959). And it was not until 1962 that the term ‘embodied technological change’ was introduced into the economic lexicon, and by contrast, disembodied change formally recognized (Solow 1962). [BB]
3. The concept of embodied technological change emerged out of the ‘scientific’ research and development (R&D) efforts of the Second World War followed by post-war organized industrial R&D programs. This experience demonstrated that scientific knowledge could be embodied in specific products and processes, e.g., the transistor in the transistor radio. The conceptual development of embodied technological change has, however, “lost its momentum” (Romer 1996, 204). Many theorists, according to Romer, have returned to disembodied technological change as a force locomotif meaning: “Technological change causes economic growth” (Romer 1996, 204). [CC]
1. While embodied/disembodied refers to the form, endogenous and exogenous refers to the source of technological change. Such a distinction is current as the ‘internalist/externalist’ debate in the history, philosophy and sociology of science (Fuller 1992). The source of exogenous technological change is outside the economic process. New knowledge emerges in response to factors such as the curiosity of inventors and pursuit of ‘knowledge-for-knowledge-sake’. In effect, exogenous technological change [DD], from the point of view of the firm or nation, falls from heaven like manna (Scherer 1971, 347).
2. By contrast, endogenous technological change emerges from within the economic process itself - in response to profit and loss. For Marx and Engel, all technological change, including that emanating from the natural sciences, is endogenous. Purity of purpose such as ‘knowledge-for-knowledge-sake’, like religion, was so much opium for the masses cloaking the inexorable teleological forces of capitalist economic development. The term itself, however, was not introduced until 1966 (Lucas 1966) as was the related term ‘endogenous technical change’ (Shell 1966). [EE]
3. Endogenous change is evidenced by formal industrial research and development or R&D programs. It therefore includes what are usually minor modification and improvement of existing and new capital plant and equipment called ‘development’ (Rosenberg & Steinmueller 1988, 230). [FF] In this way industry continues the late medieval craft tradition of experimentation. R&D varies significantly between firms and industries. At one extreme, a change may be significant for an individual firm but trivial to the economy as a whole. On the other hand, ‘enabling technologies’ such as computers or biotechnology may radically transform both the growth path and the potential of an entire economy. How to sum up the impact on the economy of the endogenous activities of individual firms remains, however, problematic.
4. With respect to the nation-state, endogenous and exogenous technological change has a different meaning. They refer to whether the source is internal, i.e., produced by domestic private or public enterprise, or external to the nation, i.e., originating with foreign sources.
1. Out of the decades’ long debate over embodied vs. disembodied and endogenous vs. exogenous technological change, a new theory emerged in the 1980s called the New Growth Theory. Initiated by Paul Romer (1986), it is explicitly endogenous and implicitly embodied.
2. Like other ‘new’ forms of economics such as the New Institutionalism (Coase 1992), New Economic History (North & Thomas 1970), New Economic Geography (Krugman 1983; Martin & Sunley 1996) and the New Economics of Science (Dasgupta & David 1994), New Growth Theory appears, at least to this observer, as an exercise in re-calibrating the Standard Model to include descriptive, empirical, institutional and historical evidence previously excluded because of its qualitative rather than quantitative nature.
3. While welcomed, the professional urge remains to fabricate such new evidence into quantitative proxy indicators to be plugged into mathematical models. Romer thus calls for more sophisticated mathematical modeling without expectation of testing because “these kinds of facts tend to be neglected in discussions that focus too narrowly on testing and rejecting models” (Romer 1994, 19-20). So much for Positivism in econometrics!
4. Beyond admitting additional sources of evidence, new growth theory introduces the concept that technological change involves non-rival ‘ideas’ that can “be stored in a bit string” (Romer 1996, 204), implicitly referring to computer programs, a form of soft-tooled knowledge. [GG] His concept, however, presents, to my mind, a confusion between information (measurable) and knowledge (immeasurable) and a failure to acknowledge the distinction between the short-run and long-run with respect to intellectual property, i.e., between knowledge residing in the private domain in the short-run but in the long-run entering the public domain.
5. With respect to information and knowledge, the ‘bit’ abstracts from content and fails, as has been demonstrated, to provide a homogenous unit measure of knowledge, or what Kenneth Boulding called ‘the wit’ (Boulding 1966, 2). With respect to intellectual property, in the short-run technical knowledge is rivalrous to the degree that copyrights, patents and other state-sponsored intellectual property rights provide protection. In the long-run, however, all intellectual property rights expire and knowledge enters the public domain where non-rivalrousness exists. Given new technical knowledge is continually being copyrighted and patented, one faces an ever moving horizon between rivalrousness and non-rivalrousness, a horizon that can never be reached. Or, put in terms of Lord Keynes’ famous aphorism: “In the long run we are all dead” (Keynes 1923).
1. As with most theoretical debates that concerning technological change in economics involves antagonists favouring mutually exclusive positions. To this observer, it appears everybody is right – to one degree or another, at one point in time or another place in space. Technological change can be disembodied in that the production function can be materially affected by systemic changes in facilitating factors such as communications, energy, information processing and transportation. The impact of the ‘B2B’ or business-to-business internet fits this category. In David’s terms, such networks constitute techno-economic regimes predominantly external to the individual firm and/or country (David 1990). Similarly, technological change can be embodied, e.g., the transistor in the transistor radio. It can also be endogenous resulting from the internal R&D programs of firms or exogenous resulting from activities elsewhere in society, e.g., in the universities or elsewhere in the world. In this sense, the production function of a knowledge-based economy rests on the composite effect of disembodied, embodied, endogenous and exogenous technological change.
2. This composite effect can be expressed as an economic Quibit of disembodied, embodied, endogenous, exogenous technological change called a FLX (pronounced ‘flex’). I coined the term in recovering a word from Newton’s original but now obsolete fluxion meaning “the rate or proportion at which a flowing or varying quantity increases its magnitude” (OED fluxion, 5). The modern term is ‘differential’. I also adopt and adapt the term corresponding fluxion or “rates at which two interdependent quantities may change simultaneously” to the rates of four quantities, i.e., a Qubit of technological change and its continuous effects on the production function of a firm or nation-state. At any point in time all four forms of technological change are entangled and at work affecting the production function. In this sense, unlike the Standard Model, there is no short-term in a knowledge-based economy during which factors of production are fixed. [HH]
3. However, the FLX can be lumpy and uneven. It can also be negative in that ‘de-industrialization’ can occur whereby knowledge moves ‘off-shore’ and is lost to a nation or firm or through ‘de-skilling’ whereby traditional praxis is embodied in a new instrument and similarly lost but this time to a machine. Such a loss of knowledge is somewhat analogous to the ‘Kuhnian loss’ in scientific revolutions shifting from one paradigm to another (Fuller 2000).
4. On the one hand, output may be increased or cost reduced; on the other hand, however, there is a loss of knowledge as domestic production is replaced by foreign or advanced machine production. This is one reason for not using the conventional term ‘flow’ which conveys a sense of constancy. Exogenous changes like innovation of a new general purpose tool may transform the entire economy (David 1990, 335). On the other hand, endogenous tinkering on the shop floor, the ‘D’ or development in ‘R&D’, may only contribute to a specific firm. A FLX is a measure of all four types of technological change or ‘new knowledge’ as it affects the production function of a nation or a firm. Such new knowledge may emerge from the NES as physical technological change, from the HSS as organizational change, from the Arts as design change or some combination thereof. It may also enter the production function in the form of any or all of personal & tacit labour, codified & tooled capital and/or newly recognized toolable natural resources.