On August 17, the Federal Reserve Bank of Kansas City, the arm of the Fed that runs the annual Jackson Hole economic meeting at the end of August, announced the symposium topic for 2018. The announcement seems rather opaque to me, but perhaps that is intentional, in the best central banker and Chairman Greenspan tradition. In case you want to parse it for yourself (skip it if you just want the English language version), here it is:
This year’s symposium topic will explore dynamics that have contributed to shifts in productivity, growth and inflation that are of concern to central bankers. Within product markets, there has been a notable increase in economic activity associated with large multinational corporations along with increased market concentration in many industries. These developments suggest that large firms today may have greater market power than in the past, and this shift may result in a decrease in competition within many industries. These shifts should concern central bankers since they likely have important linkages to observed structural changes in the global economy, including lower capital investment, a declining labor share, slow productivity growth, slow wage growth and declining dynamism.
Promptly after the announcement, Sam Fleming of the FT wrote an article explaining the issues, as he saw them. The Fleming article is clear. Perhaps Fleming has an inside scoop. Here is a set of opposing graphs that Fleming used to illustrate the apparently adverse correlation between corporate power and worker pay.
Quite a few similar illustrations have been published in recent years. Probably I have a dozen stored up. The following graph (from MIT professor Andrew McAfee) is one of the most arresting illustrations of the potential problem:
These illustrations certainly alert the observer to a strong correlation and, therefore, to the possibility that the two phenomena may be causally related.
Mr. Fleming and a co-author teed up the problem in an earlier article in the FT this month:
America’s biggest companies are grabbing a swelling share of revenues while workers suffer from pedestrian wage growth, teeing up an intensifying debate over whether public policy needs to respond,” said Sam Fleming and Brooke Fox of the FT in a recent lead sentence. The suggestion, both in that lead sentence and in the article that followed, is that the large company share of revenue is causing the pedestrian wage growth. Acceptance of that causality, moreover, is leading many Democrats, both those in office and the rank and file, to demand that something be done to stem the tide of large company dominance. The FT authors continued:
The IMF published research in June focusing on a measure of corporate power — mark-ups measuring the gap between the prices charged and production costs. Among US publicly listed companies these have risen by a sales-weighted average of 42 per cent from 1980 to 2016, in a possible sign of weaker competition. Similar trends are visible in other countries. ‘We see evidence of rising market power and declining competition in the US,” said Daniel Leigh, deputy division chief in the western hemisphere department of the IMF. ‘This is coupled with signs that the [labour] share is going down.’
Democratic party politicians and progressive think-tanks have latched on to the phenomenon: Elizabeth Warren, the Massachusetts senator, has urged that antitrust authorities sharpen their teeth and claims that ‘competition is dying’ in America.
These kinds of observations lead to calls for closer supervision of the largest companies and for stepped up antitrust enforcement. They also lead to calls for more union-friendly labor laws and restrictions on corporate power over employees.
However, in my opinion, what has been happening to competition in America—and to a lesser extent worldwide—is both more complex and less amenable to earlier ideas about the roles of the firm and of employees. In this article I will try to answer the question: “Is the large-company share of revenue and increased profitability causing the stagnation of middle class incomes?”
The major phenomena driving the perceived changes
Four major phenomena have been changing the competitive marketplace over the last 30 years. The most dominant has been the increasing importance of technology, both in selling goods and services and in gaining well-paid employment. Second, the type of businesses that the newer technologies have enabled are “scalable”—that is, they are capable of serving an almost infinite number of customers at little marginal cost. And third, all businesses have learned to utilize acquisitions as a way both to capture new technology and to reduce competition. I will discuss these three phenomena and how they may relate to wage stagnation in reverse order. I will not separately discuss the fourth major phenomenon—globalization—but its role will be apparent from the other discussions.
More vigorous antitrust enforcement could address the third phenomenon
The acquisition phenomenon can be addressed through more vigorous antitrust enforcement—though perhaps the usual means of doing so, Section 7 of the Clayton Act, last amended in 1950, needs to be rewritten for 21st century reality in order for such enforcement to succeed in the courts.
The pharmaceutical industry, mining, and many consumer products industries, as examples, have used consolidation to reduce competition (sometimes claiming that market power is needed to combat market power). That can be prevented, and prices to consumers probably would benefit. Some additional jobs also might result, as one knows well from the merger announcements almost all trumpet that a certain number of jobs will be eliminated due to efficiencies gained by the merger. To investors, the reduction in workforce is a benefit. To the employees, it is not.
Should antitrust policy care about whether a merger produces greater efficiency in the form of reduced head count—one way or the other? Classical antitrust theory does not deal with that issue. Either a merger reduces competition and tends toward monopoly or it does not. Period.
I like that classical formulation because of what I was taught (and believe) about courts and their competencies. That is, courts are good at deciding that something is either a this or a that. Appointed judges are not so good at weighing whether the benefits of a particular action outweigh its detriments. Such judgments are necessarily subjective. Usually, we let markets decide those kinds of questions—and sometimes elected officials intervene when there is an apparent consensus.
Markets are not perfect decision-makers, but at least they represent the cumulation of consumer decision-making. And that is what market economies should rely on, so long as market participants are informed.
Markets and Merger decisions
But do markets make merger/acquisition decisions? These merger decisions are made most obviously by senior managements and boards of directors, subject in some cases to stockholder approval. Is that the market speaking? “Hardly,” I can hear many of you saying. And I agree, those panjandrums are not the market. But what motivates the panjandrums, if it is not to reduce competition? (Remember, we are assuming a case that does not violate the classical antitrust laws but that society might want to prevent because the loss of jobs is worse than the accompanying gain of efficiency.) The panjandrums are motivated mostly by creating efficiency. Why would they care about efficiency? So the corporation can make more money. How would the corporation do that? It would sell more product at a higher markup. How would it do that? By becoming more efficient. Who determines whether the strategy succeeds? The market does. Ergo . . .
Well, maybe not quite “ergo”, but at least maybe we should have a presumption that the market’s judgment will be better for society than the judgment of some alternative group of panjandrums who have a different (usually more opaque and often more dishonest) set of agendas. At least I know the corporate panjandrums will have to face two kinds of markets: The consumers who buy the products or services and the stockholders who hope to benefit from the efficiencies. The panjandrums know ahead of time that it is by those markets that their decisions will be judged.
Today’s managements and boards of directors are (or should be) aware that their success depends on how their policies affect their various constituencies. As an example, writing about Wells Fargo’s serial missteps in dealing with its customers, Ben McLananhan recently wrote in the FT:
“Rebuilding trust with our team members, customers, communities, shareholders, and regulators remains our top priority,” says Ancel Martinez, a spokesperson [for Wells]. “As part of our work to transform Wells Fargo, we are committed to closely examining every part of our company, fixing the issues we find, and making things right for all of our stakeholders.”
But once burnt, the stakeholders are hard to convince, McLanahan reports. “These companies are all running ads . . . saying how much they’ve changed. But consumers are saying, ‘OK, we don’t really believe you’,” says Mr Byrne [a social media analyst]. Saying sorry is one thing. But scandals are only over when customers say they are over.”
In theory at least, acquisition decisions are subject to the same kinds of market scrutiny as the other kinds of strategic decisions that got Wells and its board and management into trouble. The market decides whether businesses succeed.
Scalability of modern tech businesses—natural monopolies or oligopolies?
Phenomenon number two that I described above—the scalability of modern tech businesses—has changed many facets of business and employment. Tech-based businesses now can grow larger more quickly than old-fashioned manufacturing or sales businesses. Individual tech companies can grab high market shares that are difficult to dislodge. And on the employment side, these businesses can operate with relatively few employees compared with their revenues, and those employees are disproportionately high-skilled technology, design and marketing workers. In some cases, such as Apple (AAPL), their employee count is low partially because they subcontract manufacturing to foreign companies that employ foreign workers. From a U.S. data point of view, therefore, Apple looks more productive per employee than it would if it manufactured everything itself.
Customer loyalty (coupled in many cases with patent protection) also works together with their low marginal costs to allow tech companies to achieve higher profit margins—both gross and net—than more traditional large companies. These are the phenomena that the Fleming and Fox FT article identified as alarming Democrats and the IMF—and as perhaps calling for governmental action.
The following Doug Short graph illustrates what has been happening.
In the aggregate, corporate profits have been receiving the benefits of productivity gains while median household incomes have stagnated. The big questions are “Why has that occurred?” and “Should government do something about it?” I suggest that to discuss the second question requires an answer to the first question. Is something nefarious causing the disconnect that the graph shows? Or is it a natural phenomenon? If it is a natural phenomenon, that does not mean that nothing needs to be done. But that conclusion might suggest that the cure should not be to interdict the natural phenomenon, but, rather, to ameliorate its consequences for those who get left behind.
Causation is notably hard to prove. But there are some important correlations that lead me to believe that it is the changing nature of the most profitable businesses that has caused stockholders to appear to have gained so much at the expense of labor. Here is a graph that I have used a number of times. It shows how income per employee has changed over the years for the biggest companies by market cap compared with the relatively static income per employee at the companies that employ the largest numbers of people.
Income per employee
Here is the same data in numeric form:
Income per employee in dollars deflated to 1970 using GDP deflator
Top 50 by market cap
Top 20 employers
An employee of the technology companies that dominate today’s market cap leaders now accounts for almost 10 times the income that an employee of a big employer such Kroger accounts for. The tech company employees are paid very well, but because there are so few of them, compared with the number of people employed by the big employers, the national data show employees falling behind in the aggregate.
Another way to look at the same phenomenon is by market capitalization per employee for the top 50 firms by market cap. The change from 1970 is stark. But the change from 200o to 2014 is the shocker. The change was not enormous from 1970 to 1990, but in the 1990s the big upward movement began, and the in next 14 years the change was so great that it takes two bar charts to see both phenomena. Here are the first 30 years and the whole 44-year period side by side. The red bars show the first four data points. The blue bars show the five data points together.
Market Cap per employee for Top 50 firms by market cap
1970 (1), 1980 (2), 1990 (3) , 2000 (4) and 2014 (5)
in 1970 dollars, using GDP deflator
(Data from S&P IQ, computations and graphs by the author)
These data indicate the changing way the stock market has seen companies with many employees versus companies that have other valuable attributes, such as highly skilled employees, patents, trademarks, and customer loyalty. The companies that the market valued in the 1970s were manufacturing companies and oil companies, for the most part. By the late 1990s, that had changed significantly, and in the 2000s the top companies are largely technology companies that, to the extent that they manufacture products, do so mostly overseas where costs are lower. Thus, again, the data suggest that technology and globalization (significantly, China’s admission to the WTO at the end of 2001) have caused the changing relationship between business value and employment—and thus the relative stagnation of middle class incomes.
Globalization also has changed the nature of the profits of the big-cap corporations in that a far larger share has come from international sales. That has increased the corporate profits without correspondingly increasing the aggregate incomes of American workers.
The nature of productivity
It is not surprising that high-productivity companies have fewer employees. Think about the way productivity works: A company becomes more productive by using fewer people to make more revenue. And that is how in 2007 and again in post-recovery 2018 American manufacturing has made about three times the amount of goods as it did in 1972 with only about two-thirds the number of employees, as the following graph derived from FRED numbers illustrates.
Industrial production (blue line)
Manufacturing employment (red line)
(IPMAN vs MANEMP in FRED Data)
with 1972 indexed as 100
Based on these and other data (I am trying to be relatively brief here), I conclude that most of the change that we have seen in the Doug Short graph of profits, productivity, and household incomes and the McAfee graph above is due to the changing nature of business (and from globalization and from changes in American family structures as well), not to changes in governmental policies or heinous business practices (unless outsourcing is a heinous business practice). Indeed, I would argue that governmental policies have not changed enough—that is, they have not changed enough to enable a larger part of the population to participate in the bounty of the tech revolution.
Thus, although I am a Democrat, I do not see the way forward as being to hamper the great things that the tech revolution and globalization have brought us. The way forward has to lie with better educational policies and better safety net policies so that fewer people are left behind and those that are left behind nevertheless can live decent lives.
I recognize that I have not dealt with why the profit and income data in McAfee’s graph pretty much track from 1947 to 2001. I think we can see why the divergence took place around that date, but that does not explain the previous correlation. That will have to wait for anther day.
The rising importance of technology—the fundamental cause of the changing nature of work, and its relationship to profit and productivity
The importance of technological change is obvious. The fundamental question posed by Sam Fleming’s FT articles with which I led this article is whether it should be reined in somehow in the interests of workers who do not participate in its bounty.
Antitrust policy has not been designed to protect employees. It has been designed to protect competition in the belief that competition reduces prices and increases consumers’ range of choices. Moreover, the history of breaking up monopolies has not been very encouraging. Most monopolies are based on technology, and technological change has tended to make many breakups look irrelevant in hindsight.
It is popular to react to corporate power by calling for the breakup of large corporations. But I would allege that those who call for the breakup neither understand the potential value of the large aggregation nor the difficulty of breaking them up without destroying the benefits that they provide. I would rather address things like lobbying power in some other, more direct way.
If we look at contemporary prices and consumer choice in that light, I think we see prices declining and choice increasing, largely as a result of both globalization and the very technology companies that have become dominant in recent years. Amazon (AMZN) is perhaps the best example of how a company has used a combination of global sourcing and technology to increase choice and decrease prices. Amazon is huge and has a huge market share and market cap, though it does not have huge profits. But the consumer benefits of Amazon’s business have been enormous. Prices of everything Amazon sells are influenced by Amazon’s presence in the market. And the ease of buying on its platform has increased consumer choice. Thus although Amazon may appear entrenched and dominant, it has not acted like a monopolist.
Role of the stock market
The stock market has played an important role in enabling American technology to flourish. Not only does the market provide capital—even more importantly, it provides the exit and incentive to early investors. Without those benefits, our technologies would not flourish as they have. We should prize and protect the market. And as investors, we should understand how we can use our role most profitably in light of the market’s most important benefits. It is our support for younger companies and younger technologies that benefits both us and the economy the most.
Increased inequality needs to be addressed, but bashing the successful will not accomplish the goal of greater equality
At bottom, the politicians and pundits who decry the success of today’s great technology-based corporations want better earnings and a better lifestyle for ordinary Americans. I am with them on that goal—and I think the Board of Governors of the Federal Reserve System and their staff are, too. The question is how to achieve that goal, not how to confuse that goal with apparently politically attractive efforts to prevent success.
The answers have to lie in three spheres: (1) education, in the broadest sense, (2) income protection for those who lose their jobs, regardless of the cause, and (3) a better set of safety nets, including healthcare and other basic essentials of life. The goal should be to reduce stress without damaging incentives to work and to be productive citizens. There always will be tradeoffs between reducing stress and damaging incentives. That is inevitable. But American public policies should err on the side of reducing stress and improving lifestyles because we can afford to do that and because otherwise we will be forced into the alternative zone of class conflict and bringing down the successful. Let us use our strength to protect both the successful and the less successful. The successful may have to pay more taxes to achieve that result, but the costs will be worthwhile because they will create a happier, more cohesive nation.
Can monetary policy do any of that? I doubt it. Steady growth and fewer and shallower recessions help everyone. Let the Fed pay attention to that. It would be a great accomplishment.
Disclosure: I am/we are long AAPL. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.