Institutions and technological progress
Beyond the interaction of different kinds of knowledge was the further level of interaction and feedback between human knowledge and the institutional environment in which it operates.
Before 1750, economic progress of any kind had tended to run into what could best be called negative institutional feedback. One of the few reliable regularities of the pre-modern world was that whenever a society managed, through thrift, enterprise, or ingenuity to raise its standard of living, a variety of opportunistic parasites and predators were always ready to use power, influence, and violence to appropriate this wealth. Such rent-seekers, who redistributed wealth rather than created it, came either from within the economy in the form of tax-collectors, exclusive coalitions, and thugs, or from outside as alien pillagers, mercenaries, and plunderers. Before 1815 the most obvious and costly form of negative institutional feedback was, of course, war. Rentseeking and war often went in hand-in-hand. Britain, France, the United Provinces, and most other Continental powers fought one another constantly in hugely costly attempts to redistribute taxable real estate, citizens, and activities from one to the other - a typical “mercantilist” kind of policy.[107] Economic growth indirectly helped instigate these conflicts. Wealth accumulation, precisely because it was mostly the result of “Smithian Growth”, was usually confined to a region or city and thus created an incentive to greedy and well-armed neighbors to engage in armed rent-seeking. It was surely no accident that the only areas that had been able to thwart off such marauders with some success were those with natural defenses such as Britain and the Netherlands. Yet the Dutch United Provinces were weakened by the relentless aggressive mercantilist policies of powerful neighbors.[108] The riches of the Southern Netherlands - unfortunately easier to invade - were repeatedly laid to waste by invading mercenary soldiers after 1570. More subtle forms of rent-seeking came from local monopolists (whose claims to a right to exclude others were often purchased from strongmen), guilds with exclusionary rights, or nobles with traditional rights such as banalites. A particularly harmful form of rent-seeking took the form of price controls on grain that redistributed resources from the countryside to the city by keeping grain prices at below equilibrium levels [Root (1994)].Had institutional feedback remained negative, as it had been before 1750, the economic benefits of technological progress would have remained limited. Mercantilism, as Ekelund and Tollison (1981, 1997) have emphasized, was largely a system of rentseeking, in which powerful political institutions redistributed wealth from foreigners to themselves as well as between different groups and individuals within the society. The political economy associated with the Enlightenment increasingly viewed the old rent-seeking traditions of exclusionary privileges as both unfair and inefficient. Mercantilism had been a game of international competition between rival political entities. To defeat an opponent, a nation had to outcompete it, which it often did by subsidizing exports and raw materials imports, and imposing a tariff on finished goods. As it dawned upon people that higher productivity could equally outcompete other producers, they switched to a different policy regime, one that economists would certainly recognize as more enlightened.[109] In the decades around 1750, mercantilism had begun to decline in certain key regions in Western Europe, above all in Britain, where many redistributive arrangements such as guilds, monopolies, and grain price regulations were gradually weakening, though their formal disappearance was still largely in the future. The Age of Enlightenment led to a few pre-1789 reforms on the Continent thanks to the enlightened despots, but it was the French Revolution and the ensuing political turmoil that did more than anything else to transform Enlightenment ideas into genuine institutional changes that paved the road for economic growth [Mokyr (2005a)].
The Enlightenment also advocated more harmonious and cosmopolitan attitudes in international relations, which may have contributed to the relative calm that settled upon Europe after the Congress of Vienna. Political reforms that weakened privileges and permitted the emergence of freer and more competitive markets had an important effect on economic performance. The institutional changes in the years between 1770 and 1815 saw to it that the Industrial Revolution was not followed by a surge in rent-seeking and violence that could eventually have reversed the process [Mokyr (2005a)].The positive Bsedbackbetweentechnological and institutional change is central to the process of historical change. The co-evolution of technological knowledge and institutions during the second Industrial Revolution has been noticed before.[110] Above all, three kinds of institutions were important in facilitating the sustained technological progress central to economic growth: (1) those that provided for connections between the people concerned mostly with propositional knowledge and those on the production side; (2) those that set the agenda of research to generate new propositional knowledge that could be mapped into new techniques; and (3) those institutions that created and safeguarded incentives for innovative people to actually spend efforts and resources in order to map this knowledge into techniques and weakened the effective social and political resistance against new techniques. As noted above, even some of the formal endogenous growth models require a growing proportion of labor in the “invention sector”, a condition that clearly demands that their profits not be expropriated altogether.
The formal institutions that created the bridges between prescriptive and propositional knowledge in late eighteenth and nineteenth century Europe are well understood: scientific societies, universities, polytechnic schools, publicly funded research institutes, museums, agricultural research stations, research departments in large financial institutions.
Improved access to useful knowledge took many forms. Cheap and widely diffused publications disseminated it. All over the Western world, textbooks of applied science (or “experimental philosophy” in the odd terminology of the time), professional journals, technical encyclopedias, and engineering manuals appeared in every field and made it easier to “look things up”. Technical subjects penetrated school curricula in every country in the West (although Britain, the leader in the first Industrial Revolution, lost its momentum in the Victorian era). The professionalization of expertise meant that anyone who needed some piece of useful knowledge could find with increasing ease someone who knew, or who knew someone who knew. Learned technical journals first appeared in the 1660s and by the late eighteenth century had become one of the main vehicles by which prescriptive knowledge was diffused. In the eighteenth century, most scientific journals were in fact deliberately written in an accessible style, because they more often than not catered to a lay audience and were thus media of education and dissemination rather than repositories of original contributions [Kronick (1962, p. 104)]. Review articles and book reviews that summarized and abstracted books and learned papers (especially those published overseas and were less accessible), another obvious example of an access-cost reduction, were popular.86 In the nineteenth century, specialized scientific journals became increasingly common and further reduced access costs,technology of mass production and continuous flow. In their pathbreaking book, Fox and Guagnini (1999) point to the growth of practically-minded research laboratories in academic communities, which increasingly cooperated and interacted successfully with industrial establishments to create an ever-growing stream of technological adaptations and microinventions. Many other examples can be cited, such as the miraculous expansion of the British capital market which emerged jointly with the capital-hungry early railroads and the changes in municipal management resulting from the growing realization of the impact of sanitation on public health [Cain and Rotella (2001)].
86 This aspect of the Industrial Enlightenment was personified by the Scottish writer and mathematician John Playfair (1748-1819) whose textbooks and review essays in the Edinburgh Review made a special effort to incorporate the work of Continental mathematicians, as witnessed by his 1807 essays on the work ofMechain and Delambre onthe Earth’s meridian, and his 1808 review ofLaplace’s Traite deMecanique Celeste [Chitnis (1976, pp. 176-177, 222)]. at the cost of requiring more and more the intermediation of experts who could decode the jargon.
To be sure, co-evolution did not always produce the desired results quickly. The British engineering profession found it difficult to train engineers using best-practice knowledge, and the connections between science and engineering remained looser and weaker than elsewhere. In 1870 a panel appointed by the Institute of Civil Engineers concluded that “the education of an Engineer [in Britain] is effected by... a simple course of apprenticeship to a practicing engineer... it is not the custom in England to consider theoretical knowledge as absolutely essential” [cited by Buchanan (1985, p. 225)]. Afew individuals, above all WilliamRankine at Glasgow, argued forcefully for more bridges between theory and practice, but significantly he dropped his membership in the Institute of Civil Engineers. Only in the late nineteenth century did engineering become a respected discipline in British universities.
Elsewhere in Europe, the emergence of universities and technical colleges that combined research and teaching in the nineteenth century simultaneously expanded propositional knowledge and reducing access costs. An especially good and persuasive example is provided by Murmann (2003), who describes the co-evolution of technology and institutions in the chemical industry in imperial Germany, where the new technology of dyes, explosives, and fertilizers emerged in constant interaction with the growth of research and development facilities, institutes of higher education, and large industrial corporations with a knack for industrial research.[111] Institutions remained a major determinant of access costs.
To understand the evolution of knowledge, we need to ask who talked to whom and who read what. Yet the German example illustrates that progress in this area was halting and complex; it needs to be treated with caution as a causal factor in explaining systematic differences between nations. The famed technische Hochschulen, in some ways the German equivalent of the French polytechniques, had lower social prestige than the universities and were not allowed to award engineering diplomas and doctorates till 1899. The same is true for the practical, technically oriented Re- alschulen, which had lower standing than the more classically inclined Gymnasien. Universities conducted a great deal of research, but it goes too far to state that what they did was a deliberate application of science to business problems.[112] Universities and businesses co-evolved, collaborating through personal communications, overlapping personnel, and revolving doors. The second Industrial Revolution rested as much on industry-based science as on the more common concept of science-based industry [Konig (1996)].Designing institutions that create the correct ex ante motivations to encourage invention is not an easy task. Economists believe that agents respond to economic incentives. A system of relatively secure property rights, such as emerged in Britain in the seventeenth century, is widely regarded as a prerequisite. Without it, even if useful knowledge would expand, the investment and entrepreneurship required for a large scale implementation of the new knowledge would not have been forthcoming. On a more specific level, the question of the role of intellectual property rights and rewards for those who add to the stock of useful knowledge in generating economic growth is paramount. Some of the best recent work in the economic history of technological change focuses on the working of the patent system as a way of preserving property rights for inventors. In a series of ingenious papers, Kenneth Sokoloff and Zorina Khan have shown how the American patent system exhibited many of the characteristics of a market system: inventors responded to demand conditions, did all they could to secure the gains from their invention and bought and sold licenses in what appears to be a rational fashion. It was far more accessible, more open, and cheaper to use than the British system, and attracted ordinary artisans and farmer as much as professional inventors and eccentrics [Khan and Sokoloff (1993, 1998, 2001) and Khan (2005)].
Whether this difference demonstrates that a well-functioning system of intellectual property rights was essential to the growth of useful knowledge remains an open question. For one thing, the American patent system was far more user-friendly than the British system prior to its reform in 1852. Yet despite the obvious superiority of the U.S. system and the consequent higher propensity of Americans to patent, there can be little doubt that the period between 1791 and 1850 coincides roughly with the apex of British superiority in invention. The period of growing American technological leadership, after 1900, witnessed a stagnation and then a decline in the American per capita patenting rate. Other means of appropriating the returns on R&D became relatively more attractive. In Britain, MacLeod (1988) has shown that the patent system during the Industrial Revolution provided only weak and erratic protection to inventors and that large areas of innovation were not patentable. Patenting was associated with commercialization and the rise of a profit-oriented spirit, but its exact relation to technological progress is still obscure.[113]
What is sometimes overlooked is that patents placed technical information in the public realm and thus reduced access costs. Inventors, by observing what had been done, saw what was possible and were inspired to apply the knowledge thus acquired to other areas not covered by the patent. In the United States, Scientific American published lists of new patents starting in 1845, and these lists were widely consulted. Despite the limitations that patents imposed on applications, these lists reduced access costs to the knowledge embodied in them. The full specification of patents was meant to inform the public. In Britain this was laid out in a decision by chief justice Lord Mansfield, who decreed in 1778 that the specifications should be sufficiently precise and detailed so as to fully explain it to a technically educated person. In the Netherlands, where patenting had existed from the 1580s, the practice of specification was abandoned in the mid-1630s but revived in the 1770s [Davids (2000, p. 267)].
In at least two countries, the Netherlands and Switzerland, the complete absence of a patent system in the second half of the nineteenth century does not seem to have affected the rate of technological advance [Schiff (1971)]. Of course, being small, such countries could and did free-ride on technological advances made elsewhere, and it would be a fallacy to infer from the Dutch and Swiss experience that patents did not matter. It also seems plausible that reverse causation explains part of what association there was between the propensity to patent and the generation of new techniques: countries in which there were strong and accessible bridges between the savants and the fabricants would feel relatively more need to protect the offspring of these contacts. Lerner (2000) has shown that rich and democratic economies, on the whole, provided more extensive patent protection. The causal chain could thus run from technological success to income and from there to institutional change rather than from the institutions to technological success, as Khan and Sokoloff believe. It may well be true, as Abraham Lincoln said, that what the patent system did was “to add the fuel of interest to the fire of genius” [cited by Khan and Sokoloff (2001, p. 12)], but that reinforces the idea that we need to be able to say something about how the fire got started in the first place.
Other institutions have been widely recognized as aiding in the generation of new techniques. Among those are relatively easy entry and exit from industries, the availability of venture capital in some form, the reduction of uncertainty by a large source of assured demand for a new product or technique (such as military procurement or captive colonial markets), the existence of agencies that coordinated and standardized the networked components of new techniques, and revolving doors between industry and organizations that specialize in the generation of propositional knowledge such as universities and research institutes.
There is a fundamental complementarity between knowledge growth and institutional change in the economic growth of the West. Augmenting and diffusing knowledge produced the seeds that germinated in the fertile soils that economic incentives and functional markets created. Without these seeds, improved incentives for innovation would have been useless. Commercial, entrepreneurial, and even sophisticated capitalist societies have existed that made few important technical advances, simply because the techniques they employed rested on narrow epistemic bases and the propositional knowledge from which these bases were drawn was not expanding. The reasons for this could be many: the agendas of intellectual activity may not have placed a high priority on useful knowledge, or a dominant conservative religious philosophy might have stifled a critical attitude toward existing propositional knowledge. Above all, there has to be a belief that such knowledge may eventually be socially useful even if the gains are likely to be reaped mostly by persons others than those generating the novel propositional knowledge. Given that increasing this knowledge was costly and often regarded as socially disruptive, the political will by agents who controlled resources to support this endeavor, whether they were rich aristocratic patrons or middle-class taxpayers, was not invariably there. The amounts of resources expended on R&D, however, are not the only variable that matters. Equally important is how they were spent, on what, and what kind of access potential users had to this knowledge.
One specific example of an area in which technological innovation and institutional change interacted in this fashion was in the resistance of vested interests to new technology [Mokyr (1994, 2002)]. Here institutions are particularly important, because by definition such resistance has to operate outside the market mechanism. If left to markets to decide, it seems likely that superior techniques and products will inexorably drive out existing ones. For the technological status quo to fight back against innovation thus meant to use non-market mechanisms. These could be legal, through the manipulation of the existing power structure, or extralegal, through machine-breaking, riots, and the use of personal violence against inventors and the entrepreneurs who tried to adopt their inventions.
At one level, eighteenth-century Enlightenment thinking viewed technological change as “progress” and implicitly felt that social resistance to it was socially undesirable. Yet there was a contrary strand of thought, associated with Rousseau and with later elements of romanticism such as Cobbett and Carlyle continuing with the Frankfurt school in the twentieth century, that sincerely viewed industrialization and modern technology and the Enlightenment that spawned them as evil and destructive. Such ideological qualms often found themselves allied with those whose human and physical capital was jeopardized by new techniques. Mercantilist thought, with its underlying assumptions of a zero-sum society, was hugely concerned with the employment-reducing effects of technological progress. The ensuing conflict came to a crashing crescendo during the Industrial Revolution. The Luddite rebellion - a complex set of events that involved a variety of grievances, not all of which were related to rent-seeking - was mercilessly suppressed. It would be a stretch to associate the harsh actions of the British army in the midlands in 1812 with anything like the Enlightenment. All the same, it appears that rent-seeking inspired resistance against new technology had been driven into a corner by that time by people who believed that “freedom” included the freedom to innovate and that higher labor productivity did not necessarily entail unemployment.
The British example is quite telling.[114] In the textile industries, by far the most resistance occurred in the woolen industries. Cotton was a relatively small industry on the eve of the Industrial Revolution and had only weakly entrenched power groups. There were riots in Lancashire in 1779 and 1792, and a Manchester firm that pioneered a powerloom was burnt down. Yet cotton was unstoppable and must have seemed that way to contemporaries. Wool, however, was initially far larger and had an ancient tradition of professional organization and regulation. Laborers in the wool trades tried to use the political establishment for the purposes of stopping the new machines. In 1776 workers petitioned the House of Commons to suppress the jennies that threatened the livelihood of the industrious poor, as they put it. After 1789, Parliament passed sets of repressive laws (most famously the Combination Act of 1799), which in Horn’s (2003) view were intended not only to save the regime from French-inspired revolutionary turmoil, but also to protect the Industrial Revolution from resistance “from below”. Time and again, groups and lobbies turned to Parliament requesting the enforcement of old regulations or the introduction of new legislation that would hinder the machinery. Parliament refused. The old laws regulating the employment practices in the woolen industry were repealed in 1809, and the 250 year old Statute of Artificers was repealed in 1814. Lacking political support in London, the woolworkers tried extralegal means. As Randall has shown, in the West of England the new machines were met in most places by violent crowds, protesting against jennies, flying shuttles, gig mills, and scribbling machines [Randall (1986, 1989)]. Moreover, in these areas magistrates were persuaded by fear or propaganda that the machine breakers were in the right. The tradition of violence in the West of England, writes Randall, deterred all but the most determined innovators. Worker resistance was responsible for the slow growth and depression of the industry rather than the reverse [Randall (1989)]. The West of England, as a result, lost its supremacy to Yorkshire. Resistance in Yorkshire was not negligible either, but there it was unable to stop mechanization. Violent protests, such as the Luddite riots, were forcefully suppressed by soldiers. As Paul Mantoux put it well many years ago, “Whether [the] resistance was instinctive or considered, peaceful or violent, it obviously had no chance of success” [Mantoux (1961 [1928], p. 408)]. Had that not been the case, sustained progress in Britain would have been severely hampered and possibly brought to an end.
90
In other industries as well resistance appeared, sometimes from unexpected corners. When Samuel Clegg and Frederick Windsor proposed a central gas distribution plan for London, they were attacked by a coalition that included the eminent scientist Humphry Davy, the novelist Walter Scott, the cartoonist George Cruickshank, insurance companies, and the aging James Watt [Stern (1937)]. The steam engine was resisted in urban areas by fear of “smoky nuisances”, and resistance to railroads was rampant in the first years of their incipience. Mechanical sawmills, widely used on the Continent, were virtually absent from Britain until the nineteenth century.[115] Even in medical technology, where the social benefits were most widely diffused, the status quo tried to resist. When Edward Jenner applied to the Royal Society to present his findings, he was told “not to risk his reputation by presenting to this learned body anything which appeared so much at variance with established knowledge and withal so incredible” [Keele (1961, p. 94)].[116] In medical technology, in general, resistance tended to be particularly fierce because many of the breakthroughs after 1750 were inconsistent with accepted doctrine, and rendered everything that medical professionals had laboriously learned null and void. It also tended, more than most other techniques, to incur the wrath of ethical purists who felt that some techniques in some way contradicted religious principles, not unlike the resistance to cloning and stem-cell research in our own time. Even such a seemingly enormously beneficial and harmless invention as anesthesia was objected to on a host of philosophical grounds [Youngson (1979, pp. 95-105; 190-198)].
With the rise of the factory and the strengthening of the bargaining power of capitalists, authority and discipline might have reduced, the ability of labor to resist technological progress at least for a while. The factory, however, did not solve the problem of resistance altogether; unions eventually tried to undermine the ability of the capitalist to exploit the most advanced techniques. Collective action by workers imposed an effective limit on the “authority” exercised by capitalists. Workers’ associations tried to ban some new techniques altogether or tried to appropriate the entire productivity gains in terms of higher piece wages, thus weakening the incentive to innovate. On the other hand, laborers’ industrial actions often led to technological advances aimed specifically at crippling strikes [Bruland (1982) and Rosenberg (1976, pp. 118-119)].[117]
9.