Economies of Scale Versus Big Business

Last week, some very large employers – including Salesforce, Tesla, and Walmart – called for a corporate merger moratorium for hospitals and doctors groups. It’s unusual to have the paragons of big business assert the need for aggressive antitrust, but it speaks to how confused our current economic debates really are. My first blog post for the Commons was on the need for conservatives to think about corruption, to use bright line standards such as a merger moratorium to remove the temptation to concentrate power and misuse it. These very large institutions seem to agree, though of course, only where they are the victims of monopoly power and not the perpetrators of it.

This dynamic speaks to the problem of corruption in our governing institutions, a fear of concentrated power that has traditionally animated both the populist left and, at least rhetorically, the modern Republican Party. But this fear has waned. Samuel Hammond recently wrote an ode to Alexander Hamilton on this blog, as the godfather of industrial policy, fitting with the instincts of a right-of-center reform coalition to promote bigness, and to create national champions that can compete with China. Hammond notes it is not just the right but the socialist left who valorizes Alexander Hamilton, as a central planner, though they seek the Green New Deal and transformational industrial goals different from traditional conservatives goals. Regardless, both the statist left and planner right believe that economies of scale and scope require policies to support, or at least overlook, corporate concentration.

The planner model for addressing corporate concentration is not break-ups, but what John Kenneth Galbraith called countervailing power, or the use of big labor or big government to check big business, or big buyers like chain stores to check big sellers like packaged goods monopolies. The idea is that we must support big business because of scale economies, but find a way to check the concentrations of power they bring.

My sense, however, is that thinking about scale economics as necessarily connected to the legal form of a corporation or the specific regulatory model is a conceptual error, and leads to inefficiently substituting policies that enable the acquisition of market power by big slothful businesses over policies that actually support productive high technology industries.

Take what once used to be of the most popular examples of economies of scale, Boeing. In 2018, this corporation looked like it was on top of the world, with what looked to be excellent products, revenue of over $100 billion, and a limitless future. Boeing is a unionized and regulated company sustained by military contracts, a textbook countervailing power example if there were one. And yet its market power masked the deterioration of its underlying capacity to engineer aircraft. Today, the economies of scale Boeing exhibits is in efficiently making badly engineered civilian aircraft and the economies of scope it offers are in producing spacecraft that malfunction and military hardware that doesn’t work, all of which are at least partially result of a merger with McDonnell Douglass that antitrust enforcers should have stopped in the 1990s.

More to the point, there are obviously returns to scale in aerospace, but there are also obvious risks in consolidation and diseconomies of scale as well. If we hadn’t consolidated aerospace into one domestic corporation whose leaders were focused on sustaining market power, we’d have another corporation to fall back on. That is, for all the claimed benefits and checks of countervailing power, we put all our eggs in one basket, and so the mismanagement of Boeing not only ruined Boeing, but the entire domestic aerospace industry. Now we have no choice but to deal with Boeing as is, and may have to nationalize it, but that’s because we ruined a vibrant and competitive high-tech aerospace industry through consolidation.

The Boeing example speaks to a broader point. Yes there are important operational economies of scale in some business lines, but we have to recognize these arguments (1) are often overstated (2) are often used to justify specific unrelated business tactics designed to acquire market power (3) ignore diseconomies of scale and (4) ignore that such economies can often be organized through standard-setting bodies without being controlled by any particular corporation. Part of the problem, as I’ve noted previously, is confusion over what we mean by scale:

There is technical scale, which means massing men and capital in a way that produces more operational efficiency or innovation. There is legal scale. And too often we hear ‘economies of scale’ and confuse the two. Having a big factory can be operationally efficient. But holding company which owns a bunch of such factories and exploits market power can be slothful.

Take General Motors. In the early 1990s, its average plant produced 206,000 vehicles a year. Honda was about a third GM’s overall size, but its average Japanese plant time made 650,000 vehicles a year. GM was the least efficient producer, with its plants about 40% less productive than Honda’s. GM made arguments about scale, premised on the idea that policymakers wouldn’t see the difference between legal and technical scale.

In the 1980s, during his tenure on GM’s board of directors, Ross Perot talked about the problem of size and management: “At GM, if you see a snake, the first thing you do is to hire a consultant on snakes. Then you get a committee on snakes, and then you discuss it for a couple of years. The most likely course of action is-nothing. You figure, the snake hasn’t bitten anybody yet, so you just let him crawl around on the factory floor.”

Technical economies of scale are not necessarily related to the corporate structure for example, the TCP/IP protocol scaled by a factor of 10 million since it was created in the early 1970s by the government. Corporations preferred walled gardens (DECNet, WangNET, HPNet, Xerox OIS, etc) but highest scale network – the internet itself – happened outside of the legal form of the corporation.

As conservatives rethink their political economy framework, it makes sense to retain a fear of concentrated power and a recognition of the power of liberty, not just in a rhetorical sense, but because it actually is powerful. And this brings me back to the myth of Alexander Hamilton.

In many ways, Hamiltonian thinking served as the guiding ideology of Barack Obama, who gave an important speech at the Hamilton Project in 2006, which was then warehousing his coming administrative team, including the author of the paper Walmart: A Progressive Success Story. Obama saw himself as an heir to Teddy Roosevelt, and gave a speech at Osawatomie, Kansas in 2011, a little over one hundred years after Teddy Roosevelt had unveiled the New Nationalism in the same city.

This kind of thinking was not new. Herbert Croly, the founder of the New Republic, originated the idea of “Jeffersonian ends through Hamiltonian means,” which is what many imagine the New Deal, with its large administrative state, ultimately was. Croly helped create the idea of New Nationalism, which animated Teddy Roosevelt’s 1912 run. Roosevelt, who carried Hamilton’s flame, wanted domestic monopolies to battle foreign monopolies, as planners seek today. Croly was a strong advocate of big business, because he saw monopolies as inevitable and a sign of historical progress. Yet in his day, Croly had to deal with a serious political problem in praising monopolies and Hamilton, because Americans at the time understood that a key axis of debate between Hamilton’s Federalists and Jefferson’s Democratic-Republicans was over whether the U.S. would be a military dictatorship, not over economies of scale.

By contrast, Croly, with his technocratic progressivism, dabbled with authoritarianism in the 1920s, saying about Mussolini that “the Fascist route has its significant and even promising aspects,” substituting “movement for stagnation, purposive behavior for drifting and visions of a great future for collective pettiness and discouragement.” Croly never fully warmed to autocratic movements, but he was a proponent of big business, which he knew was not popular among the public. So he injected the Jeffersonian brand into his political economy vision so as to forestall the idea that big business would be considered autocratic in its structure. In other words, the idea of Jeffersonian ends by Hamiltonian means is a public relations strategy by liberal corporatist.

Thomas Jefferson’s Democratic-Republicans, and their populist heirs, were modern adopters of high technology. Indeed Phyllis Schafly routinely praised Jefferson’s creation of the patent office. In the 20th century, populists like Wright Patman and Estes Kefauver loved the ability of science to improve and strengthen democracy, helping to create the National Science Foundation. Patman even held the first hearings on automation, in the 1950s.

Economies of scale do exist, but how we choose to structure corporations is distinct from how to take advantage of network systems or declining production costs for mass produced items. Such an attitude, that humans have power over how we deploy scale, sounds fantastical to New Nationalists, who imagine anyone that discounts the importance of big business is simply akin to a flat-earther, because they just won’t admit empirical reality that bigness is what delivers modernity. But in fact, you are reading this on a system known as the “world wide web,” which is not owned by any specific corporation, but seems to be scalable and efficient nonetheless. Historian Margaret O’Mara wrote in her recent history of Silicon Valley that Hamilton Project founder, Clinton Treasury Secretary, and Goldman Sachs CEO Robert Rubin, the first time he was shown a browser, asked who owned the system he was looking at. “No one,” was the response, and he apparently could not believe it. And yet, this worldwide network of instantaneous connection, which until recently had few gatekeepers, exists. If it were up to big business or Alexander Hamilton, it probably wouldn’t.