Title: Knowledge and Competitive Advantage: The Coevolution of Firms, Technology and National Institutions
Author: Johann Peter Murmann
Scope: 3 stars
Readability: 4 stars
My personal rating: 5 stars
See more on my book rating system.

Topic of Book

Murmann overviews one of the most important early industries in 19^th Century industrialization: the synthetic dye industry in Britain, Germany, and the United States.

If you would like to learn more about the interaction between technology, skills and corporations, read my book From Poverty to Progress: How Humans Invented Progress, and How We Can Keep It Going.

My Comments

Murmann creates a theory for how industrial firms use technology to gain a competitive advantage and uses one industry to illustrate its usefulness.

Key Take-aways

• The synthetic dye industry in the late 19th Century was important for Germany to industrialize and catch up to the UK.
• Germany was able to gain early advantages over foreign competitors and protect those advantages to develop a new industry.
• Germany invested heavily in university-level education in chemistry, organic chemistry and chemical engineering. This quickly made Germany world leaders in the skills needed for the emergent industry.
• The German synthetic dye firms in the 1880s established the first corporate R&D labs. University professors and their students played an important role in staffing these institutions. These employees stayed in close contact with universities
• The few German companies that had the strongest ties to universities were more competitive than those with weaker ties. There was strong cross-fertilization of people and ideas between top professors, their students and the most competitive companies.
• Because British and American companies were located far from the centers of chemical learning, they had trouble attracting top candidates.
• German patent law enabled corporate labs to protect their research and give them a monopoly in the German market.
• Germany also established protective tariffs to enable the German synthetic dye industry to evolve into being competitive on the world market.
• Innovation came from two distinct levels of selection. The first winnowing process occurred inside firms, with companies testing dyes for perceived suitability. The second level was the marketplace which reduced the set to an even smaller number of dyes that could make a profit.

Important Quotes from Book

Frequently, enterprise, technological, and economic developments are examined in separate studies. In this book I bring together these different literatures and investigate an important chapter in business, economic, and technological history. I argue that a complex evolutionary process led to very different degrees of industrial success by dye firms in Britain, Germany, and the United States. The study identifies differences in educational institutions and patent laws as the key reason for the long-lasting German leadership position in this industry. When the German synthetic dye industry had pulled ahead of its foreign competitors, its superior performance allowed it to lobby government agencies to enhance educational institutions and patent regulations, creating a cumulative spiral of competitive advantage… At the level of the individual firm, a key finding is what the winners in all three countries shared in common: In contrast to the losers, they had strong ties to the centers of knowledge about organic chemistry.

At the heart of the theory lies the concept of coevolution…  I believe a coevolutionary theory that models firms as interacting with their social environment takes a significant step toward explaining how industrial leadership is gained and lost and how small initial differences in performance can translate into large differences over time.

One of the key propositions of this book is that the creation of German dominance in the synthetic dye industry before World War I cannot be understood without coming to terms with successful and unsuccessful patent law, science funding, and tariff lobbying efforts in the three countries.

After working in a professor’s university laboratory, chemists often moved from academia to industry, from one firm to the next, and sometimes back to a university position. This created an informal network of ties that connected players in industry and academia. Mapping the network on a worldwide scale for the period before World War I reveals not only that this informal network was overwhelmingly constituted of Germans but also that the central positions were occupied by players from Germany. Explaining the shift in industrial leadership in the synthetic dye industry is intimately bound up with being able to account for the strong and weak ties in what I will call the academic–industrial knowledge network.

Expanding on Campbell’s variation, selection, and retention model of evolutionary change, Durham (1991, p. 22) identifies five system requirements for an evolutionary theory of change:

R1. Units of transmission

R2. Sources of variation

R3. Mechanisms of transmission

R4. Processes of transformation

R5. Sources of isolation

An evolutionary explanation needs to identify clearly a unit of transmission – for example, genes, ideas, values, words, or even entire languages (R1). It has to specify how these units are transmitted through time and space – for example, sexual intercourse in biology or social intercourse in culture (R3). It needs to say where variations come from – for example, gene mutation in biology or invention in culture (R2). And it has to articulate clearly the process that transforms the system through selection – for example, changes in the frequency of a trait in a population based on natural selection through differential birth or death of variant organisms in biology or differential adoption of variant ideas in culture (R4). To account for stable differences in systems of the same kind (for example, how it is that two societies speak different languages), an evolutionary explanation needs to provide for sources of isolation (R5).

Evolutionary theories try to overcome unrealistic assumptions about human beings in rational choice theories. Instead of viewing individual human beings as perfect calculating machines that can consider all possible alternatives for a particular decision, know the future perfectly (at least in probabilistic terms), and then calculate the optimal course of action, evolutionary theories see human beings as fundamentally limited in their ability to consider alternatives, foresee the future, and make error-free calculations.

For the most part, organizations act on a simple principle: If a given routine works, let’s do more of it; if it does not work, let’s do less.

The GM example highlights that change within an organization often comes about by creating new, relatively autonomous, divisions, by removing resources from existing divisions, by selling off divisions, or by closing them down altogether.

One key implication is that significant change at the industry (or population) level is likely to come from the birth of new organizations with different organizational routines and the exit of established organizations. Efforts to encourage existing organizations to transform themselves radically, as in the case of converting defense contractors to makers of consumer products, may be bound to fail.

I use the prefix “co-” in coevolution not in the restricted sense that two things are evolving together but in the broader sense that multiple things are jointly evolving.

Two evolving populations coevolve if and only if they both have a significant causal impact on each other’s ability to persist.

Because laboratory skills were involved in making dyes, most important dye innovations came from university-based researchers and were then commercialized by industrial firms.

After these research chemists turned out economically successful dyes, firms hired more and more chemists and pioneered an entirely new corporate function, formally organized research. The birth of corporate research and development (R&D), which today is a standard activity in high-tech industries (for overviews see Nelson, 1962; Freeman, 1982; Rosenberg, 1982), can be traced to the German synthetic dye firms in the 1880s. By the 1890s the vast majority of dyes were being discovered in the R&D laboratories of Bayer, Hoechst, and BASF.

Whereas in the early days of the industry a firm could exist by copying dyes invented somewhere else, patent laws made the systematic application of science within the boundaries of a firm a critical dimension of remaining a leader in the industry.

The organization of innovation evolved from hiring one chemist to employing a cadre of chemists who would systematically search for new dyes to complement the existing product portfolio of the firm or would find a novel synthesis that could circumvent the patent protection a competitor had on a specific dye. From the mid- 1880s onward, firms needed to master the forefront of synthetic organic chemistry to remain significant players.

I highlight the fact that the German educational and training system gave German firms a large advantage, particularly after science became a more precise tool in developing dyes.

I dissect in some detail the academic–industrial knowledge network that developed alongside the industry. The analysis shows that the central players in this network were professors such as Hofmann and Baeyer. Because the international network had its centers in Germany, German firms were, on average, closer in social (and geographic) distance to the sources of new knowledge and inventions than were British and U.S. firms. Being located more on the periphery of the knowledge network, Britain and the United States possessed an inherent competitive disadvantage in recruiting talent and discovering new technological threats and opportunities.

Inventing new products, however, is not sufficient for commercial success. Firms also have to manufacture the product efficiently. German firms apparently were more frequently able to make the transition from having a foreman control the shop floor to letting chemical scientists and engineers make the calls about how to organize and manage production. Bringing scientific methods (i.e., rational and systematic analyses) to the shop floor allowed German firms to push the forefront of production efficiency.

Because Britain offered patent protection on dyes in 1857, firms with patents on particular dyes, such as Perkin & Sons, were shielded from fierce competition and realized large profits based on their monopoly. By contrast, the absence of effective patent protection in Germany until 1877 led to a very different selection regime. Firms could enter freely, and the forces of competition would eliminate firms that could not keep up with the efficiency gains of the best producers.

The most important institution in the early success of the German dye industry was the university system, but patent laws were a second key factor that allowed the German firms to capture a dominant position.

The winners in all three countries shared one thing in common: In contrast to the losers, they had strong ties to the centers of organic chemistry knowledge.

Country-Level Performance

The dye industry thus is not only an example of a few firms achieving dominance but also a striking case of firms from a particular nation achieving dominance.

In fact, in many industries, successful firms are clustered in very few nations, often in a single one. Take as other examples movies, fast food restaurants, or credit cards. These industries are all dominated by American firms. The popular music, oil, and cigarette industries are dominated by British and U.S. firms. French firms dominate the high fashion and luxury goods industries. Japanese firms dominate consumer electronics. Furthermore, it is quite often the case, not only in the dye industry, that particular firms dominate their industry for long periods of time. In the nineteenth century the Singer company dominated the international sewing machine industry and International Harvester dominated the farm equipment industry. General Motors and Ford dominated the international auto industry for many decades in the twentieth century; only in the last two decades, with the rise of Japanese producers, have they lost their unique position. IBM dominated the mainframe computer business for decades after World War II in the same way that Microsoft today dominates the operating system business for personal computers.

Given the dramatic growth in market size up to 1914, there was much room for new firms to come into the industry. Apparently, long-term success for national dye industries was directly proportional to the number of corporate failures a country could sustain without jeopardizing the strength of the industry as a whole. More failures and more successes seem to have been parallel processes. Experiencing a larger number of start-ups, the German dye industry had more room to experiment with different firm strategies and structures. Even though most of these experiments turned out failures, the successful ones evidently had found the right recipe for capturing a large portion of the world market.

Period 1, 1857–1865: Early synthetic dyes

Period 2, 1866–1885: The rise of scientific theory in dye innovation

Period 3, 1886–1914: The age of corporate R&D laboratories

Taking advantage of the second industrial revolution required large investments in educational facilities and the creation of large managerial hierarchies. But Britain’s governmental bureaucracies had no particular competencies in organizing universal education or implementing proactive industrial policies.

Yet borrowing technology often does not require the same social organization as creating the technology in the first place.

Germany and the United States were largely unencumbered by an existing model of industrialization and thus could go down a path that differed from that of Great Britain. In both countries, the large managerial enterprise rather than a large number of small, personally controlled firms came to dominate complex and capital-intensive industry.

Before the birth of the dye industry, Germany was the home of a small, by later standards, but nonetheless very good research and teaching system in organic chemistry.

The system became better over time and provided German firms with highly qualified scientists not available to the same degree in Britain and the United States before 1914.

Given that knowledge in synthetic organic chemistry was a critical, hard-to-imitate resource, German firms enjoyed a substantial advantage that they were able to translate into a domination of world dye markets… But until World War I, American universities offered almost no advanced training in synthetic organic chemistry, which formed the key knowledge base for the dye industry.

In the first period the advantage of residing in an institutional environment that provided more organic chemists was related to a larger number of start-ups that could experiment with the right formula of success for the new industry; in the second and third periods, however, the chief advantage had to do with the ability of dye firms to hire academically trained chemists who could apply scientific theory for the development of new dyestuffs.

The innovative capabilities of the German dye industry rested to a large extent on the human capital that was embodied in the highly skilled organic chemists and engineers employed in corporate R&D laboratories to create new products and processes.

Neither the German nor the British dye industry would train their scientists and engineers fully in-house. They would rely rather on universities, polytechnical institutes, and trade schools to provide their scientific and technical personnel with fundamental education. Firms then hired these well-educated individuals and taught them the skills peculiar to the specific tasks required in the industrial context of the synthetic dye industry. The key point here is that dye firms in both countries relied (as they still do today) on their social environment for the training of their scientific and technical human resources.

The German system from the 1830s onward became the center of the most advanced system of higher education in the world for the next 100 years.

Britain excelled in producing a large number of world-class “private” scientists, who were often associated with learned societies and academies, but did not develop the institutional framework for producing large numbers of qualified students who could be employed by industry. German universities, in contrast, were often educating more chemists than society needed.

For business enterprises, the establishment of the polytechnic schools that sprang up throughout the German states in the 1830s was at least as important as the universities in creating the technical talent that could propel the German dye industry into a leadership position. Whereas laboratories like that of Liebig channeled a significant number of students into industrial firms (17 percent of his students in Giessen went into industrial manufacture, 30 percent went into pharmacy, and 14 percent took academic jobs in chemistry; Fruton, 1988, p. 17), polytechnic schools formed the backbone of the institutional structure for the training of engineers and skilled craftsmen.

A very large number of contacts between academia and industry were initiated and cemented through teacher–student relationships.

One important reason why the German system of higher technical and scientific education was superior to that of the British lies in the higher level of financial support by German governments. At the end of the nineteenth century, the British government paid £26,000 to universities for all purposes. Prussia alone, albeit the largest German state, supported her universities with £476,000. The respective figures for the academic year 1911–1912 amount to £123,000 and £700,000.

Utilitarian goals (making the German states stronger) and national crisis (being run over by Napoleon) are primarily responsible for societies’ support of higher education.

Starting with the Land Grant Act in 1862, the United States began to build a very large university system, creating many new campuses and upgrading existing ones. The Land Grant colleges in the individual states were typically set up with the explicit purpose of training students to serve the practical needs of agriculture and industry within the state. Reflecting the pragmatic attitude of the larger culture, engineering dominated the new universities.

By 1900, the face of American higher education had been changed dramatically. Research in science had become institutionalized among the leading universities. But I emphasize that this research was driven by practical concerns. The U.S. university system did not develop strength in pure science that was in any way comparable to what was achieved in Germany before the First World War.

Organizations that were important to dye industry:

• Professional Organizations
• Trade Organizations
• Academic and Trade Journals
• Corporate Research Laboratories
• State Policy
  • Technical Higher Education
  • Taxes
  • Tariffs

The center of the academic–industrial network became located in Germany because the laboratories of Hofmann (Berlin), Baeyer (Berlin, Strasbourg, Munich), and later Emil Fischer (Strasbourg, Munich, Berlin) became leading centers of organic chemistry where new knowledge was created at a faster rate than in British laboratories.

Students from the United States and Britain would very frequently come to Germany to receive advanced training in organic chemistry. But a reverse flow did not take place once Hofmann had left London.

The student–teacher relationship was the most important mechanism that created the strong links in the network. Spending long hours together in the laboratory, making joint scientific discoveries, and publishing papers under joint authorship not only transferred the knowledge (often tacit) of how to do organic chemistry but also created strong emotional ties between teachers and students. It was not uncommon that a student would marry into the teacher’s family, and sometimes a teacher would marry into the family of the student.

[Research labs] meant that German firms were more protected against innovations made abroad because they would be quickly informed about them and possessed the capabilities necessary to copy or build upon the innovation.

Many German firms appear to have possessed an important advantage over their British competitors because they academically trained chemists in charge of the organization and management of the shop floor, whereas in Britain the traditional foremen continued to rule over the production process and the associated labor practices.

Because the argument I develop is complex, I want to summarize it at the outset.

The German dye industry benefited from not being able to obtain patents in Germany before 1877 because the fierce competition in Germany forced German firms to build superior capabilities in the manufacturing and marketing of dyes. Once Germany had captured a dominant position in the global synthetic dye industry because its leading firms had developed the best organizational capabilities, the German dye industry benefited from the arrival of a German patent law in 1877. The leading firms could then cement their dominant position by developing systematic corporate R&D laboratories that kept them the most innovative firms in the world.

Hindsight makes it clear that entering the synthetic dye industry early was to be of great competitive advantage, in terms of both gaining a strong position with dyers and printers and building up effective skills in the manufacture of dyes. No firm founded after 1870 became a world leader.

Individual Firms

Production equipment had to be devised from scratch. Specialized chemical apparatuses simply did not exist at the beginning of the dye industry. The earliest dye plants tried to scale up laboratory equipment by using vessels and ovens that were developed for other industrial purposes. Those firms that experimented a lot with different production arrangements, different equipment, and alternative chemicals had a very good chance of uncovering more efficient methods.

No firm had a master plan of how it was going to operate in the new synthetic dye business. Rather, firms would be guided by the most pressing problems and opportunities – or what Cyert and March have called problemistic search – proceeding largely by trial and error.

Second, firms did not possess a systematic R&D strategy, in that they lacked formalized procedures for developing dyes on a routine basis. Once a new dye was created somewhere in the larger community of colorists, dyers, printers, or entrepreneurial and academic chemists, firms would scramble to get access to the new product by acquiring the patent (Britain) or by buying the information on how the dye was prepared… The key factors to success were process experimentation on the shop floor – bringing production costs down – and establishing a customer base that would provide revenue and profits that could finance investments in organizational capabilities. Both innovators (Perkin & Sons and SM&N) and imitators (Bayer and J¨ager) did well as long as they found customers who would place a steady flow of orders. It was the firms without a solid customer base that were simply pushed out of the market, accounting for many early failures in the industry.

Because over time theory in organic chemistry became better equipped to understand the chemical structure underlying dyes (one example is Kekulé’s famous benzene ring model, widely used by 1880), the search for new colors during this second period could be guided to some extent by scientific theory… Although process innovations in the manufacture of existing products still took place predominantly in industry, the vast majority of radical product innovations such as synthetic alizarin originated in university laboratories during this period.

The door was open for firms to make scientific and technical talent the chief weapon for competing in the marketplace. Given this shift of innovative activity, the key strategic requirement for firms was to get close to the generators of new knowledge in universities.

In the 1870s and early 1880s, the most important dye innovations came out of university laboratories.

The Age of Bayer (1886–1914)

The period from the mid-1880s until the beginning of World War I can be called “the Age of Bayer”; during this time, the firm caught up with its larger rivals BASF and Hoechst in Germany and left Levinstein, BS&S, and Jager far behind. Building on the firm’s existing position and the science capabilities of the German social environment, the managers of Bayer pursued the strategy of aggressively investing in all functions of the corporation, although particular investments continued to be made on an ad hoc basis. Especially by investing heavily in R&D, the firm pushed dye technology to unprecedented levels of sophistication and rendered many existing dyes obsolete.

Bayer’s R&D laboratories turned out new dyes at a rate that had never been seen before (Meyer-Thurow, 1982). To finance large R&D expenditures, the firm had to grow its sales volumes significantly, an approach that required the expansion of both plants and sales force. Bayer’s strategy was to become one of the few large players with powerful economies of scale and scope on their side and to knock out competitors.

Not making the same kind of investments, Jager was pushed into niche markets; BS&S went out of business altogether; and Levinstein and Schoellkopf could not match Bayer’s fast expansion.

One characteristic of the Age of Bayer was large-scale production that would lower the cost of making a particular dye. To go from manufacturing 100 dyes in 1878 to 1,800 dyes in 1913, Bayer had to expand dramatically the scale and scope of its production facilities. Since the very beginning of the synthetic dye industry, Bayer had faced a rapid decline in dye prices. The firm survived only by keeping up with the productivity levels of competitors. The key goal for Bayer was to improve productivity faster than prices were falling. As long as Bayer could decrease its cost more quickly than prices for dyes declined, it could maintain profit levels that would allow the firm to make larger investments in organizational capabilities than most rivals could.

Bayer outperformed all of the five other companies by developing a marketing organization that reached into every corner of the world. Along with Hoechst and BASF, it was engaged in a race to become the biggest dye firm in the world by systematically developing capabilities to serve very diverse customers. At the turn of the century, Bayer had sales subsidiaries in all the major textile centers of the world.

In a memorandum to all important German dye firms, Duisberg proposed a full merger of the industry into a single firm.

The experience of the synthetic dye industry shows that chance favors the prepared organization. By building large R&D laboratories, a firm such as Bayer was able to tame chance to a considerable extent. For those firms that could afford them, R&D laboratories were something of an insurance policy against the innovations that would almost invariably eventually destroy the value of existing dye products.

Those firms that had missed the boat of building research laboratories and in-house development departments when the first movers started to exploit this new organizational tool were simply left behind.

When Bayer became a leading firm in the dye industry, it did not need nearly as much luck as it did earlier to be successful. Rivals that had fallen behind (e.g., J¨ager) could not afford to make similar investments into organizational capabilities.

Success became self-reinforcing.

Success breeds more success, and failure breeds more failure.

Coevolution of National Industries and Institutions

This chapter develops empirical support for the argument that in the long run the firms in Germany were more successful than their British or American counterparts because they were more effective in collectively shaping their selection environment after the industry had taken off. Once German firms were ahead of their British and American rivals, superior economic performance and bigger size provided German firms with the resources and visibility to collectively influence their selection environment in a way that was not possible for the less successful British and American firms. A more favorable selection environment then set in motion powerful feedback processes that allowed German dye firms to extend their lead and eventually dominate the industry.

Between 1857 and 1914, scientific and technical training programs expanded significantly in Britain but not at the same rate as in Germany. A chief cause lay in the resistance of British firms to hiring many university graduates at the time that German firms were staffing large R&D laboratories with many chemists and hiring engineers to build efficient production processes. This underlines how coming relatively late into the era of industry was a blessing for Germany because firm managers were not encumbered by an outdated model that saw university graduates as an inappropriate tool for improving the efficiency of production.

This created a vicious cycle. When progressive firms (i.e., Levinstein) wanted to hire academic organic chemists and production engineers, they typically had to import German and Swiss talent before 1900 because British universities did not produce the same quality of graduates. Unless British firms paid these foreigners more than they would make at home, they would not be able to hire the best talent away from Germany and Switzerland. Either paying more or having lesser talent would put them at a competitive disadvantage with German firms. When British industry then lost more and more market share to the German firms, it did not have the resources to support universities and lobby to expand the system to the same degree as in Germany. Whereas in Germany success bred the resources for more success, in Britain the relative decline caused a lack of resources that led to further decline.

In Germany’s case the process of expanding universities was self-multiplying; in the British case it was self-limiting.

Fundamental scientific research as practiced at German universities and Technische Hochschulen was rare in the United States before World War I.

Two factors stand out. German dye firms and their industry association were more effective in organizing themselves and their lobbying efforts than was the British industry. Furthermore, just as in the case of the lobbying efforts in education, the bigger size and larger profits of the German industry gave it more resources and clout than its British rival industry had.

The tariff history of all three countries shows that the relative powers of supplier, producer, and user industries had important consequences for an industry’s ability to bring about favorable tariff regimes. The quality of an industry’s collective organization, not just its economic size, was a determinant of political influence. German firms were more successful in getting a tariff regime they desired not only because they were much bigger than their British rivals, but also because, through the network of contacts that linked firms and other centers of power, the German dye industry had the tools to engage in more successful collective action.

Toward an Institutional Theory of Competitive Advantage

I identify the exchange of personnel, the formation of commercial ties, and lobbying on behalf of the other social sphere as the more specific causal mechanisms that connected the evolution of national populations of firms with the evolution of national populations of universities.

Recall for this purpose that a sound evolutionary explanation occurs at the level of the population, not the individual, and has to meet the following requirements: First, to introduce novelty into the economic system, a mechanism must exist that creates variants of the existing structures. Without a constant source of novelty, a selection process cannot create new economic structures that may be better adapted to the economic requirements of society. Second, selection pressures need to be consistent, or more precisely, new variants must be created more frequently than new selection criteria because otherwise the evolutionary process cannot act as a “blindwatchmaker” that brings about through trial and error structures that are better adapted. The evolutionary system would degenerate into a random walk that on average could not be expected to lead to structures that are better adapted to their environment. Third, a retention mechanism must be present that transmits economic structures from the present into the future. Without such a retention mechanism, new developments could not build on previous adaptive achievements but would have to start from scratch; complex nonrandom structures would not be possible.

This articulation of evolutionary theory is much more general than the Darwinian formulation for biology. The biological case has the special features that variations are completely random and not guided by previous experience. In sociocultural evolution, by contrast, the process that creates variants is not entirely independent of the process that selects variants.

The first synthetic dye appeared in 1856 not because someone had a conscious plan to create such a product; rather, Perkin was searching for a synthetic route to a medicine and serendipitously discovered a purple substance. What made him different from other organic chemists who had seen colored materials in their test tubes was that he pondered its possible commercial value and set out to investigate whether the purple material would work as a dye for textiles. Once Perkin acquired some positive feedback from silk dyers, his family decided to invest funds into a production plant.

This description of how the dye population changed over time highlights the operation of at least two distinct levels of selection. The first winnowing process occurred inside firms, with companies testing dyes for perceived suitability. Of all the molecules a firm synthesized, only a few received the development and marketing funds necessary to introduce them into the market. The marketplace in turn selected from this reduced set an even smaller number of dyes. The dye that flourished in the market survived for some time until still better (i.e., higher quality or lower cost) synthetic dyes became available and replaced earlier ones. It is useful to conceptualize the relationship between these multiple selection processes as a nested hierarchy.

If, however, a firm used internal selection criteria that were inconsistent with the market selection criteria for too long, the firm would lose money and eventually go out of business… In contrast, those firms using internal selection criteria for dye molecules that were consistent with what the market wanted made profits that allowed them to increase their R&D budget and expand their production capacity. The retention mechanism in this process is easy to specify: Firms retain a successful dye in their product portfolios until the dye no longer finds enough customers.

Most frequently were selected out at the firm level and less frequently at the level of the marketplace. Lower-level selection criteria were eliminated by higher-level selection forces if the former yielded results that were inconsistent with the larger environment.

Starting in the mid-1870s, Kekulé’s benzene ring theory and later theoretical advances in organic chemistry such as Witt’s theory had a significant impact on the development of new dyes because the new structural understanding of molecules allowed chemists to reduce the search space for developing new dyes (Wilcox, 1966). Chemists still needed to synthesize a large number of molecules to find a useful dye, but they had a better grasp of what kind of molecule was more likely to work as a dye. This shortcut in developing new dyes – as we learn from Kekulé’s autobiographical sketches – emerged itself from a trial-and-error process.

Firms are selected for their levels of profitability…  Given that entrepreneurs differ in their psychological profiles and their experience bases, they tend to introduce variety into this population even when they consciously try to imitate the model of another firm… The larger the number of entrepreneurs that start new firms, the more varied the population is likely to be…

Business models of successful firms are preserved because they are written down in organizational rules and passed on as an organizational culture that tells members who they are in the context of the organization and in general terms what they are supposed to do.

Individuals store in their memory their respective roles in the larger organizational structure. When an individual leaves the organization, the new person assigned to the role will learn the job either through elaborate instructions or by learning from the know-how of other people who carry out similar tasks.

Firms simply could not see very far into the future. They made investments based on short-term forecasts of how the market was likely to develop. But even forecasts for the next few years were often inaccurate because firms did not know how other firms would act. When new classes of dyes emerged, several firms correctly anticipated sizable demand for the product. But when many firms that were hoping for monopoly profits entered the same market, they found that they were all chasing the same customers.

To analyze the development of university systems – populations of universities at the level of individual nations – through an evolutionary lens is much more straightforward because universities, like firms, require material resources to maintain themselves.

It is quite appropriate to describe the organization of the first university as a marketplace, where disciplines that could attract students would flourish at the expense of other disciplines that lost out in the competition for students.

Over the last 1,000 years, various different funding and governance models for universities have been tried. But one can identify general processes that shaped the development of universities over the millennium. First, those universities that attracted financial support, first-rate scholars, and students would grow relative to other universities.

In terms of the birth of new universities, the second half the nineteenth century witnessed a significant number of new institutions in Britain, Germany, and the United States. One common reason for the formation of new universities in all three countries was that traditional universities were exhibiting substantial inertia in incorporating more science and engineering education into their curricula. In Britain no new university was formed between the University of Edinburgh in 1583 and the University of Durham in 1832, followed shortly afterward (1836) by the more significant University of London.

In the sixty-four years before World War I, however, six English universities (Manchester, Birmingham, Leeds, Liverpool, Sheffield, Bristol) and one Welsh (The University of Wales) were chartered. All new universities were much more focused on science and engineering and thereby increased the frequency of appointments in these areas. Similarly, the population of German universities was greatly changed through the creation of technical universities. Altogether, eleven technical universities were organized between 1868 and 1910 (Munich, Aachen, Darmstadt, Brunswick, Hanover, Berlin, Karlsruhe, Stuttgart, Dresden, Danzig, Breslau) that focused on teaching science and engineering subjects. Some of the German technical universities were created from scratch. Others were formed out of existing polytechnics and local colleges. After a long political struggle with traditional universities, the German technical universities were given the right to grant doctorate degrees in 1901 and thereby could compete even more successfully for students.

Beginning with the Morrill Act of 1862, through which the U.S. Congress appropriated to each state funds from the sale of federal land, the United States began to build a very large system of higher education, creating many new campuses and upgrading existing ones.

The more significant evolutionary transformation of national university systems took place, however, not through the birth and death of individual universities but rather through degree choices of students and faculty appointments of individual universities that tried to adapt to changing environmental conditions.

The most important causal flow linking industry and university was the exchange of personnel between these two social spheres. This exchange led to a flow of skills and knowledge beyond what was disseminated through the academic journals available across borders. Leading professors in the academic field trained students, many of whom later became leading chemists in industrial firms. Exchange also occurred in the opposite direction. Chemists working in firms were also recruited to join the faculty of universities. The flow of personnel between the two social spheres created a mutualistic relationship in which both sides benefited. The German dye industry overtook the British one in large measure because the much larger number of German organic chemists quickly led to a larger number of firm start-ups in Germany. The fierce competition that ensued among German firms forced them to acquire superior skills. Those German firms that survived this competition had acquired capabilities that allowed them, in turn, to beat their British rivals.

Industries compete with other industries for favorable regulation, tax treatment, and other forms of support from governments (Hirsch, 1975). Because science and education budgets are limited in every country, academic disciplines compete with other disciplines for resources. Given the possible joint benefits that can accrue to academic disciplines and their related industries, academics and industrialists have an incentive to lobby governments to increase support for their partner’s social sphere, or they can form coalitions on issues that concern both. Forming such a coalition creates a mutualistic relationship between academic and industrial partners, but at the same time it can create a competitive relationship between different academic disciplines (or between different industries) by pitting them against one another.

Coevolutionary theory is an advance over existing theories because it focuses expressly on how competitive leadership positions are gained and lost over time.

For a number of reasons coevolutionary processes cannot guarantee that the economy represents an optimal adaptation to needs, desires, and opportunities of society as a whole.

First, recall that evolutionary social processes cannot look far into the future.

Second, evolutionary processes create the future out of existing structures. As a consequence what institutions can be devised are significantly constrained by present arrangements.

Third, if domestic industries face international competition, the specific historical moment during which a new industry emerges may have significant long-term consequences that are difficult to reverse because the development locks in on a particular path.

In the synthetic dye industry, national university systems and patent laws were the key institutions that mattered for the relative success of national populations of firms. At other times or in other industries, tax laws, labor laws, employer–employee relations, financial institutions, or competition policies may play a decisive role in the long-term competitive success of firms. We have good reason to believe that coevolutionary processes also shape these latter institutional structures.

The proximate cause for a firm’s competitive advantage lies in its ability to offer a product or service at lower cost or higher quality than its rivals. To achieve such a low-cost or high-quality position, a firm needs to acquire a variety of skills. Let us define firm capabilities as all those skills that allow a firm to deliver high-quality or low-cost products to customers.

In a rapidly changing environment, only firms that constantly reconfigure their activities to create innovative products, or firms that cut their production costs, will be able to sustain a competitive advantage. The central task of management is precisely to coordinate the diverse activities that occur within a firm and allocate its resources toward those activities that enjoy the greatest likelihood of success, given the capabilities the firm already possesses.

The central insight gained from the comparative analysis of the synthetic dye industry before World War I is that firms do not create their capabilities from scratch.

As long as nations differ systematically in their institutional structures, national institutions can serve as the ultimate source of competitive advantage because they provide firms with raw capabilities that are not available elsewhere at the same cost and quality.

Because the rules of the game are often not the same across countries, the same firm strategies can lead to very different outcomes in different national contexts. Two important implications follow for designing business strategies of individual firms. First, firms need to evaluate carefully in which national context to locate particular activities because differences in national contexts may ultimately decide which firms will win the industrial game.

As a general investigative rule, it is useful to keep the following in mind: no failures, no evolutionary adaptation through selection.

From an evolutionary perspective, creating an effective firm is equally a problem of acquiring the relevant skills and capabilities. Organizational change is a cumulative process. From this point of view, the future trajectory of a firm is not so much determined by external incentives as by internal capabilities. As long as a firm can obtain sufficient amounts of resources from the environment, it is much more likely to look for new opportunities in the vicinity of the existing activities than to enter markets that in theory offer the highest profit margins in the economy. When a firm makes negative profits for an extended period, it will not be able to obtain additional resources, and sooner or later it will die. Whereas in neoclassical theory, firms are typically viewed as homogeneous and constrained only by technological and market conditions, firms in evolutionary theory are seen as diverse and constrained by their own administrative heritage.

One of the key questions for the evolutionary perspective is, What kind of institutional structures are capable of generating a sufficient amount of novelty for progress in the economic system?

From an evolutionary perspective, societies are historically grown entities and exhibit a large amount of path dependence in their development. Where they can go tomorrow is significantly constrained by where they are today.

Another important question at the level of a country is how much reform and institutional innovation can be implemented without leading to a collapse or at least serious decline of the level of economic development already achieved.

By no means does the market mechanism always constitute a very stringent selection environment that will allow efficient firms to replace inefficient ones quickly.

One key contribution of this book has been to show that large firms are able to change selection criteria in their favor.

It is important to distinguish between regulation that aims to make markets work effectively and regulation that aims to undermine the working of market processes.

The former is desirable; the latter tends to reduce economic performance. Above all, policy makers need to remember that different industries require very different regulation. One size does not fit all economic activities that occur in the economy.

Early on, Hannan and Freeman (1977) saw individual organizations as fundamentally incapable of change and regarded industrial transformation as proceeding almost exclusively at the population (industry) level, through the birth and death of inert organizations. Managerial perspectives, in contrast, view organizations as quite readily changeable through managerial actions. Evolutionary theory in the tradition of Nelson andWinter (1982) takes an intermediate position.

The answer, in my view, is to see the managerial role as one of establishing internal selection criteria for the goals and actions of all the individual and collective agents that work within the organization.

Because large organizations are composed of many units – small groups, functional departments, production sites, divisions – another role of top managers is to make sure that the selection criteria used at these various levels of the organization are consistent with one another.

In this view, a CEO should see him or herself as the Chief Evolutionary Officer, responsible for ensuring that the firm engages in a sufficient amount of experimentation with products and organizational practices to increase the odds that it will be better aligned with the competitive environment. The key challenge from this perspective is to find the appropriate balance between experimentation and retaining existing practices. For the most part, managers seem to err on the side of not building enough experimentation into their firms to allow it to become an effective evolutionary system that is to a large extent self-organizing.

Recognizing, of course, that organizations, besides following rules, sometimes deliberate about new rules, Nelson and Winter (1982) introduced a hierarchy of decision rules (routines) in which higher-level, less frequently invoked rules control lower-level, more frequently invoked rules. Moving up the hierarchy, one finds that decision rules become less and less routine-like and deterministic of action.

If you would like to learn more about the interaction between technology, skills and corporations, read my book From Poverty to Progress: How Humans Invented Progress, and How We Can Keep It Going.