Technology and Global Change 37 surfing requires more hardware (a modem), software, a

telephone line, a local telephone network, and the internet itself. To ecologists the notion that
“everything depends on everything else” might be familiar. However, to students of technology
and policymakers, interdependence and interrelatedness create formidable challenges. It is
impossible to manage change through attention to just a few “key” technologies. In fact,
because of technological interrelatedness, it may even be easier to manage change where few
technologies and related infrastructures exist, such as in many developing countries. Consider,
for instance, the example of cellular or satellite telephones that can be put in place everywhere,
compared to a conventional telephone network system. This is the essence of the argument
that latecomers to development may have genuine advantages too in terms that they can “leap-
frog” (Goldemberg, 1991) older technology systems altogether. Conversely, countries “locked-
in” to large existing technology systems face difficulties to move rapidly to newer systems. A
historical example (England) and model for such entrenchment in old technology systems was
first given by the economist Marvin Frankel in 1955 (Frankel, 1955). As a contemporary example,
consider the introduction of “zeroemission” vehicles, already mandated in California. They are
not a technological novelty. Applicable inventions and innovations have existed since the turn of
the century. Thus the difficulty lies not in producing electric cars, but in solving the chronic
problem of power supply and storage. Without significant progress in batteries, for instance, the
speed and range of electric cars is severely limited and costs are high. And a new infrastructure
is also required for charging or exchanging discharged batteries. Technologies depend
increasingly on infrastructures of transport, energy, and communication. The service these
provide is much larger than the usually modest costs charged to users. We notice them most,
however, when we miss them most – when they fail. Thus infrastructures and related
technologies are important examples of what economists call “network externalities”. Consider
your telephone: even with all costs paid, it would be useless if only you owned a phone. Rather,
the utility of your phone increases with the number of participants in the telephone network and
the more people and services you can access, e.g., to enquire about a flight departure, to order
a pizza, or to chat with family and friends. Because costs are shared among all participants of
the network, but each participant has the full benefits (utility) of being able to communicate
throughout the network, the real value of the service remains “exogenous” to the price paid by
an individual. This presents serious issues when new infrastructure networks need to be put in
place. The high initial costs are incurred when benefits are still comparatively low; 38 Arnulf
Gr¨ubler if no one is prepared to incur the initial set-up costs, future benefits cannot arise.
Distributive issues are also raised because those who incur the initial high costs are not the
same people who reap the ultimate full benefits. Thus like the air we breathe, for which we pay
nothing, but without which we could not exist, infrastructures create important “externalities”.
These can be ignored in the microeconomic calculus, but they cannot be ignored by those
studying or aspiring to direct technological change. With the terminology and these four central
tendencies of technological change in place, we can now turn to the most exciting feature of
technology: technological dynamics or the mechanisms and patterns of technological change
over time. 2.2. Technological Change Some 10,000 years ago humans survived as nomadic
hunters and gatherers. This required considerable sophisticated (technical) knowledge. (If you
doubt this, try making a living today by hunting and gathering.) However, the first revolution in
technology – the development of agriculture – changed the nomadic lifestyle dramatically. The
development of markets and of money (institutional and organizational innovations or
“technologies” in a larger sense) set people free from the need to be self-sufficient, enabling
them to benefit from division of labor and specialization. Markets and agriculture (more
precisely agricultural surplus production) were fundamental drivers for the emergence of cities.
Since that time, many further technological revolutions in fields such as materials,
construction, navigation, and military technology have dramatically influenced the course of
history. The past 300 years – the “age of technology” – have witnessed more momentous
technological changes than any previous period in human history. Anthropologists, historians,
and philosophers were quick to take an interest in technology and its role in shaping societies
and cultures. Surprisingly, economists only came later to the study of technological change
(Rosegger, 1996). Observing the Industrial Revolution from its midst, classical writers in
economics from Adam Smith to Karl Marx could hardly fail to see the importance to economic
growth of technological change, of new products and new production processes. But
technological change – the “industrial arts” – was not seen as an integral element of the
economic process. Even Karl Marx, who argued that transformations in the material structure of
production determined changes in social relations, and who wrote extensively on technology,
said relatively little about the sources of such changes (Rosegger, 1996). Technology and Global
Change 39 Two economists deserve special credit for pioneering our thinking on technology:
Thorstein Veblen and Joseph A. Schumpeter. Veblen (1904, 1921, 1953), perhaps best known for
his Theory of the Leisure Class (first published in 1899), was the first to focus on the interactions
between humans and their artifacts in an institutional context. He considered technology not as
an exogenous force on entrepreneurs, engineers, or workers, but rather part of material and
social relationships. Technology was developed and shaped by social actors, while at the same
time shaping social values and behavior. Such a “circular” model of interactions was
revolutionary at a time when technology was viewed as the exclusive domain of inventors,
engineers, and “heroic” entrepreneurs (a kind of naive, romantic fascination adhered to even by
the early Schumpeter). Such a unified view of technology contains a revolutionary message
today, when many social scientists are trapped in a futile polarization between extreme
positions of technology shaping society, or in turn society shaping technology.9 More widely
acknowledged are the contributions of the Austrian economist Joseph A. Schumpeter (1883–
1950),10 who started his successful scientific career in Austria, passed through failed stages as
an entrepreneur, served a short, unsuccessful interlude as Austrian finance minister, and
completed his career at Harvard University. Schumpeter’s Theory of Economic Development,
published in 1911 and translated into English in 1934, is a landmark in considering the sources
of technological change as endogenous to the economy. His later publications, in particular the
monumental Business Cycles (1939) and the still eminently readable Capitalism, Socialism and
Democracy (1942), deepened and extended the treatment of technology in his earlier work. For
Schumpeter the essence of technological change is “new combinations”, particularly those that
represent a discontinuity, i.e., new combinations that cannot be achieved by gradual
modifications of existing artifacts, practices, and techniques. This Schumpeterian notion of
technical change is referred to as “radical” technical (as opposed to incremental) change
below. ... to produce other things or the same things by a different method, means to combine
these materials and forces differently. In so far as the “new combination” may in time grow out
of the old by continuous adjustment in small steps, there is certainly change, possibly growth,
but neither a phenomenon nor development in our sense. In so far as this is not the case, and
new combinations appear discontinuously, then the phenomenon characterizing development
emerges. ... [the latter] ... is that kind of 9These extreme positions are referred to as
“technological determinism” (e.g., Gille, 1978) versus the “social construction” of technology
(e.g., Smith and Marx, 1994). 10For an excellent biography on the life and work of Schumpeter,
see Swedberg (1991). 40 Arnulf Gr¨ubler change arising from within the system which so
displaces its equilibrium point that the new one cannot be reached from the old one by
infinitesimal steps. Add successively as many mail coaches as you please, you will never get a
railroad thereby. [Joseph A. Schumpeter, Theory of Economic Development, 1934:64–66] For
Schumpeter the essence of technological change is “changes in techniques and productive
organization”, i.e., changes in technological hardware and software. As the above quote
emphasizes, such changes are inherently “nonlinear”. They entail both quantitative and
qualitative characteristics that cannot be produced by simply adding linearly “more of the
same” to existing technologies and practices. Schumpeter also draws an important distinction
between changes that emerge from an accumulation of small gradual changes (referred to as
incremental improvements in the next section) and those that represent radical “new
combinations”. He gives five examples (1934:66), listed as follows: 1. The introduction of a new
good or product, or of a new quality of a good or product. 2. The introduction of new methods of
production, not tested yet by experience in the relevant branch of manufacturing. New
production methods may be based on a new scientific discovery, or on a new way of handling a
commodity commercially. 3. The opening of a new market, either one that did not exist before or
one that has previously not been entered. 4. Obtaining (Schumpeter uses the rather
inappropriate term “conquest of markets”) new sources of raw materials or semimanufactured
goods. The new source may already exist, or it may have been newly created. 5. New forms of
organization, e.g., the establishment or the break-up of a monopoly. It cannot be stressed
enough that any technological change, whether incremental or radical, arises from within the
economic system as a result of newly perceived opportunities, incentives, deliberate research
and development efforts, experimentation, marketing efforts, and entrepreneurship.
Technological change does not fall like “manna from heaven”. Schumpeter also emphasizes the
nonequilibrium nature of new combinations. Technological change is not simply “more of the
same”; it radically changes the relations between economic inputs and outputs, and it changes
the constraints under which these can evolve. As we will see in the next section most
macroeconomic models still largely ignore these two fundamental features of technological
change, that is: (i) evolution from within (i.e., technological change should not be exogenous
Technology and Global Change 41 to the model); and (ii) the inherently dynamic and
nonequilibrium nature of technological change, which static equilibrium models fail to capture.
With this up-front pessimism about the treatment of technological change in much of economic
modeling, let us return to Schumpeter’s own words: ... Capitalism, is by nature a form or
method of economic change and not only never is but never can be stationary. And this
evolutionary character of the capitalistic process is not merely due to the fact that economic life
goes on in a social and natural environment which changes and by its changes alters the data of
economic action; this fact is important and these changes (wars, revolutions and so on) often
condition industrial change, but they are never its prime movers. Nor is its evolutionary
character due to a quasi automatic increase in population and capital or the vagaries of
monetary systems of which exactly the same thing holds true. The fundamental impulse that
acts and keeps the capitalistic engine in motion comes from the new consumers’ goods, the
new methods of production or transportation, the new markets, the new forms of industrial
organization that capitalist enterprise creates [italics added]. ... The history of the productive
apparatus of a typical farm, from the beginnings of the rationalization of crop rotation, plowing
and fattening to the mechanized thing of today – linking up with elevators and railroads – is a
history of revolutions. So is the history of the productive apparatus of the iron and steel industry
from the charcoal furnace to our own type of furnace, or the history of the apparatus of power
production from the overshot water wheel to the modern power plant, or the history of
transportation from the mail coach to the airplane. The opening of new markets, foreign or
domestic, and the organizational development from the craft shop and factory to such concerns
as US Steel illustrate the same process of industrial mutation – if I may use this biological term –
that incessantly revolutionizes the economic structure from within, incessantly destroying the
old one, incessantly creating a new one. This process of Creative Destruction is the essential
fact about capitalism. [Joseph A. Schumpeter, Capitalism, Socialism and Democracy, 1942:82–
83] After setting the scene about the importance and essence of technological change, we can
now introduce the finer conceptual and terminological detail in the following section, which
presents a taxonomy of technological change. 2.2.1. A taxonomy of technological change11
Incremental Improvements Occurring more or less continuously across all industry or service
activities, incremental improvements resulting from scientific research and development,
engineering, and learning effects improve the efficiency of all factors 11This section is based on
Freeman and Perez (1988) and Freeman (1989). 42 Arnulf Gr¨ubler of production. Although the
combined effect of incremental improvements is extremely important, no single improvement
by itself will have a dramatic effect. The accumulation of small incremental innovations in long-
term overall productivity growth is extremely important, but the steps of individual
improvements are difficult to document in detail. As a rule they can be documented through
resulting aggregate productivity increases. Typical examples include reduced labor, materials,
or energy requirements. The associated model is the “learning” or “experience” curve – with
accumulated experience, humans learn to make things better, faster, and with fewer defects
(see Section 2.3). Economists call this “learning by doing” (Arrow, 1962) and “learning by using”.
The extent and rate of such learning effects vary according to the kind of learning involved. Most
importantly they are not “autonomous”. They should not be represented as an exogenous time-
trend function, as is frequently the case in models trying to capture technological change.
Learning depends on the actual accumulation of experience. Without “doing” there is no
“learning”. Radical “New Combinations” Radical “new combinations” are discrete and
discontinuous events. In recent decades they have usually been the result of deliberate
research and development efforts in industry, government labs, or universities. They may make
quantum leaps in productivity possible and overcome resource limitations. Or they may enable
the development of entirely new materials and products. Although they depart radically from
existing engineering practice and technologies, they nevertheless often tie in with existing
industrial structures. They therefore require no radical changes in overall industrial organization,
although they do necessitate changes at the level of plants or even industrial sectors. The
introduction of the Bessemer process, offering the possibility of low-cost, mass production of
high-quality steel in the 19th century, the introduction of nylon, or the contraceptive pill both in
the 20th century, are illustrative examples. Despite their importance for individual industrial
sectors or submarkets, their aggregate economic impact remains comparatively small and
localized, unless a whole cluster of radical “new combinations” is linked together to give rise to
entirely new industries or services. Changes in Technology Systems Under this heading we refer
to far-reaching changes in technology, affecting several branches of industry or occurring
across several sectors of the Technology and Global Change 43 economy. Such changes
combine both radical and incremental innovations with organizational and managerial changes.
Technological change in one part of the economy triggers corresponding changes both
upstream and downstream in related branches. A good example is the introduction of industrial
electric motors (cf. Devine, 1982). Before their introduction, factories would have used a central
steam engine with power distribution via transmission belts. Electric motors provided a new
versatile decentralized source of motive power. They changed, first, the entire organization of
the shop floor. Second, they required changes upstream in the production and distribution of
electricity. Without such substantial changes in organization, both on the shop floor and in
upstream electricity supply, the electric motor’s impact on productivity would have remained
localized and limited. Devine (1982) estimates that the impact of the electric motor was
multiplied by a factor of three through such organizational changes. The overall energy
efficiency of a steam engine, coupled with mechanical power distribution, according to
Devine’s estimates is between 3% and 8%. If only the steam engine is replaced by self-
generated electricity, the overall energy efficiency remains at 3–6%. However, combining utility-
generated electricity and decentralized unit drives raises overall energy efficiency to 10–12%, or
by a factor of three at the lower end of the range. These estimates report 1920s efficiencies.
Current overall energy efficiencies for industrial drive systems are on the order of 25–28%
(Naki´cenovi´c et al., 1990), twice as large as 70 years ago. Clusters and Families Some
changes in technology systems are so far-reaching that they impact upon the entire economy
and nearly every aspect of daily life. Such changes involve whole clusters of radical and
incremental improvements and may incorporate several new technology systems. The
development of the automotive industry, for example, was contingent on developments in
materials (high quality steel sheets), in the chemical industry (oil refining), in production and
supply infrastructures (oil exploration, pipelines, and gasoline stations), in public
infrastructures (roads), and a host of other technological and organizational innovations. The
growth of the industry was based on a new way of organizing production, i.e., Fordist mass
production combined with Taylorist scientific management principles. These yielded significant
real-term cost reductions, making the car affordable to a wider social strata. This changed
settlement patterns, consumption habits of the population, 44 Arnulf Gr¨ubler leisure
activities, etc. And the automobile is just one among many consumer durables now considered
standard in industrialized countries. Clusters of interdependent radical innovations and
technology systems give rise to whole families of hardware and software innovations with
associated new institutional and organizational settings. Together they multiply the effects of
each other on the economy and society. Thus their collective effect is more than the sum of
their individual contributions. It would be impossible to calculate overall impacts even if
detailed data on individual components were to exist. Qualitative descriptions are more
appropriate. In the literature such clusters have been analyzed under the headings of “general
natural trajectories” (Nelson and Winter, 1977) and “technoeconomic paradigms” (Freeman and
Perez, 1988). Such clusters drive particular periods of economic growth, and will provide the
central organizing concept for this book’s analysis of technology and global change. A
Schumpeterian (1935, 1939) perspective on long-term economic growth and technological
change sees overall development coming in spurts, driven by the diffusion of clusters of
interrelated innovations and interlaced by periods of crisis and intensive structural change.12
The existence of a succession of a number of such clusters over time does not mean that there
is a quasi-linear development path, e.g., from textiles to basic metal industries to mass-
produced consumer durables as alluded to in Rostow’s (1960) stage theory of economic
growth. Instead, such clusters are time-specific phenomena. The success of any one (in terms
of economic growth) and the drawbacks (in terms of environmental impacts) cannot be
repeated quasi-mechanistically at later periods in history or in different socioeconomic
settings. We adopt the concept of technology clusters and families to distinguish broadly
between various historical periods characterized by different driving forces and patterns of
technological change and their impacts. Our interest in global change issues together with
technological interrelatedness and interdependence explains why we have adopted a taxonomy
and perspective 12Such discontinuous paths of economic development have been
corroborated by empirical studies ever since the seminal contributions of Nikolai Kondratiev
(1926) and Joseph A. Schumpeter (1939). They received revived interest in the periods of
economic crisis in the 1970s and 1980s (see e.g., van Duijn, 1983; Freeman, 1983; and Vasko,
1987). Beyond the empirical corroboration of important historical discontinuities, however, the
interpretation and theoretical explanation of such long waves of economic and social
development remains fragmented and open to further research. In particular, debate continues,
first, on whether we are dealing with a recurring or cyclical phenomenon endogenous to the
economy, and, second, on what causes the long waves that have been identified. For an
excellent collection of classical, seminal papers of long wave theory including critical writings,
see Freeman (1996). Technology and Global Change 45 with a deliberately large boundary.
There are disadvantages to such an approach; we cannot dwell on the detail of individual
artifacts and techniques. Instead, we must analyze them as systems and address their
characteristics, and the scale and quality of their global change impacts, as a whole that is
more than just the sum of its parts. In Chapter 4 we present briefly empirical evidence on the
existence and timing of technology clusters, and identify appropriate indicator technologies
that can be used as pars pro toto for their respective technology clusters and families. We focus
on four major technology clusters since the beginning of the Industrial Revolution and identify a
possible fifth cluster that in the next millennium could transform our entire technological and
material base. 2.2.2. A taxonomy of global change: Impacts of technological change With
respect to (direct and indirect) global change impacts we group technological changes into four
categories: (i) those that augment resources; (ii) those that diversify products and production;
(iii) those that enlarge markets (output); and finally (iv) those that enhance productivity.
Technological Changes that Augment Resources The tremendous historical expansion of
industrial production has consumed enormous amount of natural resources in the form of raw
materials and fuels. Technological changes that augment the resource base have therefore
been essential. These include technologies that facilitate the discovery of new resource
deposits and that improve the accessibility and recoverability of existing resources;
technologies that represent new resource inputs altogether; and finally technologies that
substitute for existing material and fuel inputs. Technologies that increase efficiency (i.e.,
enable to produce more with less inputs) can also be considered to augment resources, but we
will discuss them separately under the general heading of productivity. The onset of
industrialization in 18th century England is usually associated with the emergence of coal as a
major new industrial fuel. Although coal had been used in the brewing industry and to evaporate
salt brines since the 13th century, its use remained limited because of restricted access to coal
resources and limited applications. Coal was basically used in the same way as the fuelwood it
was supposed to replace. Mining concentrated on comparatively shallow deposits, and coal
could only be transported from mines located near riverways and the seashore. Hence the use
of the term “sea coal” well into the 19th century. Two important technological innovations 46
Arnulf Gr¨ubler changed this situation. First was Abraham Darby’s discovery of the coking
process through which pig iron could be produced using coal instead of increasingly scarce and
expensive charcoal. Second, the invention of stationary steam engines (Newcomen-Savary)
allowed water to be pumped from greater depths than had been possible previously with
mechanical pumps driven by horses. This increased physical access to deeper coal resources.
These two technological innovations in turn paved the way for numerous subsequent
innovations. The coking process eventually gave rise to an entirely new coal-based chemical
industry that included city gas and synthetic versions of dyes like indigo. James Watt improved
the thermal efficiency of the Newcomen stationary steam engine. It subsequently was used in
mines not only for lifting water but also as a power source for mechanization, thus lowering
mining costs and improving the economic accessibility of coal resources. Most importantly it
became a mobile power source for railways. This further improved access to coal deposits and
drastically lowered transport costs. With ra