Enterprise & Society

Picking the title for this talk was the easiest part of preparing it. This phrase is one that I have used in teaching for many years, during what is essentially the climactic lecture of the core microeconomics course [End Page 710] which I have taught many, many times, and many of you have, if not taught, at least taken. The punch line of this course is that, in long-run competitive equilibrium, free markets produce an efficient outcome in which firms earn zero profits. The answer to the final exam question, “What will an efficient economy look like?” is answered by finding where profits are equal to zero. This is often puzzling to students, especially the MBA students I teach most of the time. At some point I realized that it was important to clarify to them that equilibrium is not a goal, but a result, and that their job as managers and leaders was to escape from equilibrium. Long-run equilibrium is not the goal but the result of competition. It is what competitive pressure pushes firms toward, but from the point of view of the firm, it is to be avoided. Competition is not a state; it is a force, and it is a force that, left unchecked, will leave you earning zero profits.

Firms attempt to stave off competition in many different ways. The response that is the focus of that core course in microeconomics is the decision to enter or exit a market or an industry. The standard story is very simple and not very interesting: enter if there are profits and leave if there are not. But the underlying message, that in order to understand entry and exit, one has to think about firms responding to and trying to escape from competition is critical. We have a small literature on exit in business history, but it mostly focuses on failure as something that is explained by the firm itself, what one might call an internalist explanation of exit.¹ But once one sees firms and potential firms as continuously under competitive pressure, then we see that often exit is not simply, or even primarily, about the individual firm, but rather the broader market context. It may be the result of declines in demand, either for a particular product or in the economy as a whole, or it may be the result of increased competition because of technological change or market integration. The impact of these external forces may not be what one would expect if one thought of exit primarily the result of the firm’s faults. For example, research by Tim Bresnahan and Dan Raff using US Census of Manufactures data from the 1930s demonstrates that it was not low productivity or the least efficient firms that exited during the Depression.² More recent work [End Page 711] by Yoonsoo Lee using US Census of Manufactures data from the 1980s and 1990s similarly shows that there is very little Schumpeterian “cleansing” during downturns.³ There is a lot of exit, but as in the 1930s, the principle determinant of firm survival is not higher productivity, not being better or smarter, but access to credit. We had a perfect example of this during the 2008–2009 financial crisis, when Chrysler and General Motors entered bankruptcy, and Ford did not. What distinguished Ford was not its cars, its engineering, or its labor or legacy costs. What distinguished Ford was that it had arranged for a line of credit before financial markets collapsed.⁴ And of course in the end, Chrysler and GM did survive, as did Citibank and Bank of America, but not many smaller banking competitors, because they were given access to credit from the state while others were not. So exit created by macroeconomic downturns and financial crises will reduce competition, but it does not necessarily increase productivity by culling the laggard firms. It reduces competition by eliminating those that are least well connected to sources of finance, either private or public. As Simon Johnson has emphasized, the exit process in the recent crisis reflected political connections and the ability of a few financial institutions to exert excessive influence over state policy.⁵ This allows well-connected firms to escape from competitive equilibrium at enormous costs to economic efficiency and political democracy.

The question of entry is often treated as a question of entrepreneurship, focusing on what leads an individual to enter. One of the things that Naomi Lamoreaux and I have found in our work is that the decision to form a firm, to become an entrepreneur, is certainly in part a reflection of individual predilection and family experience.⁶ Historical and contemporary studies show that having a parent who is self-employed is a very important predictor of the likelihood that someone will be self-employed themselves.⁷ But entrepreneurial activity is not just a reflection of the characteristics of the individual person or the individual firm. It is also a question of opportunity and the likelihood of success. Opportunity and success are much more likely where there is a constellation of institutions or organizations that provide the entrepreneur with the set of resources necessary to create a successful firm: these resources include new ideas, which the entrepreneur may produce or buy, access to capital, access to labor and resources, and perhaps as important as anything, access to markets. Where it is relatively easy for a person or a group of people to [End Page 712] assemble these resources, the tangible and well as intangible ones, entry is more likely. In our work on Cleveland, Ohio in the late nineteenth century, we find that inventors were much more likely to start their own firms, compared with other regions of the country with similar levels of inventive activity, where they were likely to sell off inventions or work as employees. In the Cleveland area, there were firms that were nodal in a network of inventors and local capitalists. These firms provided training for aspiring inventor-entrepreneurs, venues for developing new technology, advice about potential markets, and introductions to potential investors. These nodal firms escaped from equilibrium themselves by helping other firms enter, creating and reinforcing an innovative entrepreneurial community. The existence of these networks, and their openness to new arrivals in Cleveland, to young men from Midwestern farms, from New York and New England manufacturing cities and from western Europe—let us not kid ourselves; these networks were not exactly open to everyone—generated sustained economic growth in the region and the country, leading Cleveland to become the country’s fifth largest city by 1920.⁸ Successful fostering of entry, with older firms actively helping to create new firms, requires a relatively open business network. One of the key messages I always took from the work of Bill Parker was that it was this constellation of institutions into a relatively open network that distinguished Midwestern cities from Southern ones, encouraging both in-migration of potential entrepreneurs and the active creation of new innovative firms, resulting in sustained economic development.⁹

These networks were important for facilitating entry, because, despite the increasing integration of the US economy, many intangible assets, necessary for firm entry, were still acquired locally. Despite the increasing integration of product and financial markets in the second half of the nineteenth century, Cleveland inventors assigned the vast majority of their patents to local firms. And most Cleveland firms acquired their technology from local inventors. This was the case even when there was no employment relationship between the firm and the inventor.

These informal networks were also important in helping new firms identify their market and develop relationships with customers. One of the most striking things to me in working through Herbert Dow’s papers was how the technological innovations that led to the creation of the firm so quickly outpaced the ability of the firm to find customers. Dow knew that there was a market for magnesium, and he knew [End Page 713] he could make a lot of it cheaply, but he could not find who it was that bought the stuff.¹⁰ That distribution is important was a key insight of Chandler.¹¹ So much of business history today fashions itself “post-Chandlerian” but a crucial point made by Chandler was that manufacturing was not enough, that production was not enough. In modern capitalism, firms had to do more than just produce and dump the output onto some mythical market. They had to sell. In fact, in Jerry Davis’s virtual drone firm, it is really all about selling: the ability to create profits comes from being able to coordinate and connect each stage of the supply chain and the final consumer.¹² Chandler argued that firms increase profits by creating economies of speed, speeding up the coordination of production and distribution or what Marx called the rate of turnover of the circuit of capital.¹³ That first BHC I attended more than twenty years ago was quintessentially Chandlerian in its theme of Manufactures and Marketing.¹⁴ As Jerry’s presentation on Thursday emphasized, creating economies of speed does not necessarily require vertical integration. In fact, recent research by Chad Syverson demonstrates that the vast [End Page 714] majority of “vertically integrated” firms today do not actually transfer their output from the upstream to the downstream parts of the firm.¹⁵ But to the extent that anyone can do what Jerry’s virtual drone maker did, enter or exit with no sunk costs, and purchase services that are available to anyone, the drone maker cannot make profits in the long run. If anyone can do it, there will be competition and profits will go away. But if you control distribution, if you control access to customers, if you know who they are and what they want, you can limit competition and maintain profits. This is why the information you provide with your frequent shopping cards is so valuable. This is what Miguel Lopez-Morell’s paper yesterday on the mercury cartel emphasized.¹⁶ By controlling distribution, they control entry. They allow their producers to escape from equilibrium and maintain positive profits.

One challenge for a strategy of economic growth that depends on new entry, as was the case in late nineteenth and early twentieth-century Cleveland, is that it requires sources of external start-up finance. As a result, it is much more vulnerable to financial shocks. Incumbent firms can have retained earnings and finance growth and innovative and inventive activities out of those retained earnings.¹⁷ They can also weather downturns in demand, repurposing slack resources Penrosian style to innovation or the development of new markets.¹⁸ These network structures permit entry and increase opportunity and dynamism in an economy, but perhaps at the cost of increased vulnerability to macroeconomic fluctuations.

Naomi Lamoreaux and I have been studying firms that try, sometimes successfully, sometimes not, to create profits by creating new technologies, new products and new processes, building firms to profit from those innovations. And continuing to find new profits by innovating within the firm, but also by using their expertise to identify, encourage, finance, and develop other innovators. It is not quite the same as Bob Allen’s collective invention, because we are studying patented innovations, not open sourcing.¹⁹ But it is cooperative behavior among entrepreneurs that allows them to escape from equilibrium. Which is not to say that these firms do not compete. They do. They compete with other inventors. They compete with other producers. They escape from equilibrium by cooperating with new entrepreneurs who give them access to new technologies and new profits.

Recent research by John Haltiwanger and Ron Jarmin find that a surprisingly large proportion of both job creation and productivity [End Page 715] growth comes from new firms entering.²⁰ Of course, most new entrants are not what we think of as Schumpeterian “innovators.” They are what Lazonick calls “adaptive firms” or what recent research by Ben Puglsey from the University of Chicago argues are really people who are choosing self-employment over a boss, not choosing to build a firm.²¹ These folks may well be happy to stay in the economists’ long-run equilibrium, earning zero profits. But not any Schumpeterian, entrepreneurial firm.²² Not any MBA I know. They want to make profits. Not normal profits, real profits, economic profits.

So innovative firms are continually looking for new ways to produce and distribute goods and services. And they are continually looking for new markets. By doing so, they increase competition for other firms. When canals and then railroads lowered transportation costs, firms searching for profits took advantage and entered new geographic markets. This increased competition for other firms that had previously been protected. More recently, European integration and globalization more generally have increased competition. How do firms respond to competition? Sometimes they innovate. Sometimes they look for cheaper inputs, as when New England textile producers moved south, or when Southern textile firms moved to Mexico and then China and now to Vietnam. But the other thing that they do, in virtually every industry, is that they try not to compete. In the model and in my story, there is relentless competition. But relentless competition means zero profits. Nobody wants that. So the almost universal response to competition is collusion.

Who colludes? I have been asked this question many times, and of course it is hard to answer because most collusion is secret.²³ But the simple answer is, anyone who faces competition. Nineteenth-century construction of canals has been shown to increase output and incomes in newly accessible areas.²⁴ But the canals were also accompanied, quite routinely, by attempts to collude. Pools follow in lock [End Page 716] step the locks of the Erie Canal. Sometimes these attempts were relatively informal, gentlemen’s agreements to limit sales to particular regions or customers. These agreements were frequently followed by more formal pooling or exclusive, joint sales agreements, like the salt or bromine pools, or the mercury cartel as we learned yesterday, in which merchants or distributors took control of all output, essentially restricting each firm’s output to the amount that the distributor agreed to buy, and in some cases “dead renting” firms, paying them not to produce for some specified length of time. In industry after industry, we see a dynamic in which market integration creates competition, to which firms respond by negotiating some sort of accommodation that reduces the intensity of competition, allows them to earn positive profits, only to have that arrangement disrupted by further market integration or by technological change.²⁵ This dynamic is intensified and typified by the railroads, which both integrated markets, increasing competition among the firms that used their services, and competed among themselves in increasingly intense, to the death battles, until 90 percent of the track in the USA was in bankruptcy.²⁶ For both the railroads and their customers, these bouts of competition would resolve themselves with attempts to restore profits through collusion and ever more complex and sophisticated collusive schemes. And when that was unsuccessful through merger or regulation.²⁷

Longstanding cultural norms helped to reinforce collusion. I had what Oprah would call an “aha moment” when reading the correspondence between Herbert Dow and J. H. Osborn, Dow’s dad’s friend and one of Dow’s most important early financial backers. Dow’s expansion of bromine output was limited by a pool’s control of the market. Dow, nothing if not a young, innovative entrepreneur, chafed at these restrictions, knowing in his heart that he could wipe out the older firms in the industry if just given the chance to ditch the restrictions of the pool. Osborn repeatedly tried to calm the young Dow down, urging him to cooperate with the incumbent producers and the pool, so that the firm in which he had just invested so much would actually make some money. And finally he quoted to Dow from the Bible reminding him that it was immoral to deprive another person of his livelihood. If competition that forces exit is wrong and eschewed for moral and ethical reasons, God will keep us from equilibrium and that zero profit condition. [End Page 717]

As far as I can tell, that was generally not sufficient to maintain collusion. Debra Spar’s work on the international diamond cartel suggests that the development of a cooperative ethic was important to the stability of that cartel.²⁸ And no one would question that the diamond cartel was both stable and successful, lasting for a century. Its success also reflected the cartel’s vertical structure and control of the distribution, as well as the production, of diamonds.

In many industries in the USA at least, the combination of continuing competitive pressure and increasing legal restrictions on explicit collusion led firms to turn to other mechanisms to control competition. The merger movement led to the consolidation of industries that had formerly been cartelized. Regulation limited entry or the intensity of competition in some industries. In others, forward integration and marketing allowed firms to differentiate their products and earn sustained profits.

The emergence of large, dominant firms with continuing identities that facilitate both product differentiation and tacit collusion even in well-integrated national markets took on air of inevitability. Not just Chandler, but Karl Marx and most nineteenth-century political economists, thought larger firms were inevitable, and that bigger was almost always better. As Herbert Dow wrote, in the midst of another price war in which he was losing his investors’ money because being clever and more productive is not enough when markets are controlled by incumbents,

The day of small manufacturing business in every line has either gone by or will soon be a thing of the past. A man can accomplish so much more when working at one thing all the while than he can when changing around from one job to another so that it gives a big concern where men do not change their jobs a big advantage. I think, however, the handling of things in a big way has more of an advantage from some other points than it has from the above mentioned, for example, one man can handle the blowing out apparatus and analyze the Bromin brine if our Bromin plant was ten times as big as any we have ever had. Again, the well man that is working all the while on well repair jobs gets so familiar with the different kinds of accidents that occur that he knows how to take care of them with facility, and he has all the tools on hand that are necessary for this work.

That sounds like Adam Smith: division of labor and specialization explain firm size. But that is not all Dow says: [End Page 718]

Looking at it from another stand point, if a little concern makes money they are bound to have an unlimited amount of competition sooner or later as there are lots of people in the country who could command enough money for a little business, and would take chances of litigation with a small concern who would not dare antagonize a large concern. For my own part I am very well satisfied that this Company is going to make a lot of money out of Bleaching Powder and allied things, but the present fight on Bleaching Powder, which is world wide, has prevented any of the Bleaching Powder makers from making very much money.

(Herbert Dow to Henry Osborn, Post Street Archives, File Folder #030030)

Dow’s description of the gains from size sound remarkably like those of Adam Smith’s pin factory and reflect the widespread presumption that there are efficiency gains associated with size. But his assertion that “if a little concern makes money they are bound to have an unlimited amount of competition sooner or later” reflects the experience of his own small firm trying to enter a market controlled by incumbent firms with longstanding agreements to divide the market and restrain competition.²⁹ It was the restrictions on market access created by large firms that motivated Dow’s insistence that his firm would have to achieve similar size. That is, the imperative for size is about access to markets and the ability to prevent predation.

A generation of post-Chandlerian business historians has discovered small firms, especially networks of firms.³⁰ These firms did not just disappear as technological change created economies of scale. In different regions of the USA and around the world, firms that are not gargantuan have been shown to be viable and innovative, able to lead in the creation of new industries, new products, jobs, and productivity growth. But state policy and regulation often favor large firms, even unintentionally. Sometimes this is the result of straight-up regulatory capture, in which firms lobby to protect themselves from competition—as has been the case, for example, in an ongoing battle over building a new bridge between Detroit and Canada. The monopoly private owner of the existing Ambassador Bridge has lavished donations on state legislators to protect his monopoly, distorting not only transportation and tourism costs, but also the political process.³¹ There is a long history of firms using the state to regulate [End Page 719] entry and limit competition. State regulation, as Xaq Frolich illustrated so nicely during the Krooss session yesterday, can create and expand markets.³² But firms do not usually respond gratefully, leaving those markets open to all comers. As the owners of the Charles River Bridge demonstrated almost two hundred years ago, a state charter is good, but a state monopoly is better. The inter-state salt pools that arose after the construction of the Erie Canal used the state licensing and inspection power to facilitate collusion. When the state regulates firms, it is often the case that regulation is written by large firms themselves, even when those large firms have created the problems that regulation is being asked to solve. This was true following the 2008–2009 banking crisis.³³ This was true of the regulation of the stock market after the Great Depression, which favored the New York Stock Exchange and therefore firms with connections to New York City, to the detriment of local and regional stock markets that had sprung up across the country in the late nineteenth and early twentieth centuries. These local capital markets were created by local boosters, businessmen, and bankers working together to build their businesses and their regions.³⁴

The creation of large firms to escape competition creates what we might think of as negative externalities. As the size of one firm in the supply chain grows, there is an incentive for others to increase in size. For example, we have seen consolidation cascade through the automobile industry, as supplier firms attempt to achieve the size necessary to contract and bargain with original equipment manufacturers. A regulatory bias toward large firms emerges when there are underfunded health, safety, or environmental regulators who then rely on industry expertise to write regulations and industry capacity to self-regulate.³⁵

The structure of industry and the relationship between firms and the state that emerged after the Great Depression was one dominated by and favorable to large multi-product firms, firms with long-lived identities and vertically integrated into distribution. These were firms who could survive the shocks of the Depression, take advantage of broad federal engagement in the economy during the war, and control access to markets. [End Page 720]

If you had asked me twenty years ago, when I first started studying cartels, whether explicit collusion was an important way that firms managed to control competition in the second half of the twentieth century, I probably would have said no. It is an idiot test: if you are an idiot, you need to talk to other firms to control competition. If you are not an idiot, you can limit the intensity of competition, you can escape from equilibrium, with a million other tools: advertising, innovation, price leadership, or focal point pricing. Turns out I was wrong. (Either that, or there are a lot of idiots.)

If we look at the interwar period, the best estimates suggest that the percentage of world trade controlled by international cartels was in the order of 30–40 percent.³⁶ But presumably after the War it was different, especially since the occupying US forces imposed Sherman Act-like rules in Germany and Japan? It was a little different, for a little while. A paper that Valerie Suslow and I wrote about a decade ago found that international cartels that had been caught and convicted of illegal price fixing—which means we did not include any cartels like OPEC that have sovereign immunity or any that had successfully hidden their operations once the Europeans decided that cartels undermined rather than supported the integration of Europe—we found that those cartels affect about 5 percent of global trade.³⁷ Turning to domestic cartels, George Symeonidis’s wonderful book on collusion in the United Kingdom in the 1950s finds that 36 percent of industries self-reported having collusive agreements, and another 26 percent were ambiguous in their self-reports. Some of the industries classified as competitive undoubtedly were, like food and drink or clothing, but chemicals almost surely just did not report their collusive agreements, and others, like tobacco were so concentrated that collusion was redundant.³⁸ For the USA, where price fixing is illegal—it is a felony, I remind my students, and the USA throws people in jail, presumably nice jails where you get to hang out with Martha Stewart and get decorating tips, but still, if you are a Japanese or Swiss executive, it is probably something to be avoided—the extent of explicit collusion is harder to establish. From 1960 to 1987, there were about ten criminal cartel convictions by the US Department of Justice (DOJ) per year. On average, according to the US DOJ these cartels lasted a [End Page 721] little over five and a half years.³⁹ While there were some convictions of large, well-known firms, as in the famous GE-Westinghouse phases of the moon case, many of these firms were relatively small and in geographically well-defined markets—road paving and school milk. That is no longer the case. From 1992 to 2010, there were about 700 US DOJ cartel convictions, and the cartels lasted on average about 8 years.⁴⁰

And while there were still plenty of cases of small firms in a regional market colluding, there were also cases of large firms, firms we know well, the DuPonts and Mercks and Samsungs of the world. Firms that have antitrust compliance programs so you would think they would know better. The US DOJ has convicted over a hundred international cartels in the last twenty years and collected over $7 billion in fines from the conspirators.⁴¹ These cartels increase prices and restrict trade, the diffusion of technological innovation, and the development of new products.⁴² They do not do so in random industries. They do so in highly concentrated industries. And I do not mean a little, I mean a lot—an obscenely lot. In a recent paper that Valerie and I wrote, we said “The cartels in our sample occur predominantly in very highly concentrated industries. The mean industry four-firm concentration ratio (C4) is 75 percent.” Two-thirds of the cartels were in industries in which the largest four firms controlled over 75 percent of the global market.⁴³ Jerry emphasized to you on Thursday evening that there was a remarkable transformation of the corporation under way and that maybe it was disappearing. What he’s describing captures a very real phenomenon about how the supply chain is organized today. Merck, a leader of the Vitamin C cartel, may well get its Vitamin C today from Chinese factories that it does not own. But it controls their output, so it controls the market, and it prevents competition. It escapes from equilibrium and the erosion of its profits. Actually, a little back story: Chinese Vitamin C producers helped to destroy the Vitamin C cartel, producing and selling more cheaply than cartel members could—and selling a lower quality product in many cases, at least as they first entered the industry. They sold to Proctor and Gamble before they sold to Coca Cola, because [End Page 722] we drink Coke and we do not drink Tide. But after the demise of the cartel, the western Vitamin C producers contracted for the output of Chinese firms. They cannot buy them, so it is not a merger. It is not one big firm. But it is not competition. It is more like those old nineteenth-century salt pools.

We recently celebrated the hundredth anniversary of the breakup of Standard Oil; in the past twenty years, the component organizations of Standard Oil have recombined.⁴⁴ The component organizations of ATT have recombined. Our most important recent monopoly cases, Microsoft and Intel, eschewed structural remedies. Our famously fragmented banking industry has seen dramatic increases in concentration in the past decade, so that the C4 of the US banking industry increased from 34 to 53 percent between 2000 and 2010; the HHI increased from 497 to 877.⁴⁵ We know that in many specialized financial services, there were just a few firms operating, and they engaged in the most explicit and unacceptable behavior, such as paying their competitors to exit a market.⁴⁶ This pattern in banking is partly the result of regulatory capture: only the industry has the expertise to regulate itself, and the industry’s ability to offer employment opportunities to former regulators is unsurpassed. Recent legislation creates a new statistical agency, an Office of Financial Research, but the US government cannot pay anyone who has the expertise to run such an agency a salary that will compete with employment in the private sector.⁴⁷

The recent period has seen increased market integration, particularly in the European Union, but also globally with the establishment of the WTO and the proliferation of bilateral trade agreements. Market integration brings new competition, and new competition reduces profits. A wonderful paper by Espen Storli on Alcoa and the aluminum cartel yesterday asked, why did Alcoa continue to do this when it was so clearly illegal?⁴⁸ The answer is that competition erodes profits, and firms are determined to escape from that bad equilibrium. It is our job as business historians to illustrate the multitude of ways that firms do this. It is our job as citizens, as members of society, to assure that firms can choose to compete and to escape from [End Page 723] competition through innovations that improve our world, and that all of that creative energy is not directed toward undermining living standards, undermining competition, undermining the dynamism of our economy, or undermining democracy.