Setting The Record Straight: Don’t Confuse Impressive Growth For Stagnating Competition
In the wake of Apple’s recent valuation as a $1 trillion company, we continue to see a number of mistakes in competition analysis. Below are three oft-repeated claims that miss the mark with further details below.
1. Yes, economy-wide, wages and business investment have been stagnant, but not in tech where wages are rising and R&D investment leads all industries. Further, economy-wide, labor shares of corporate profit have fallen, but not in tech/telecom.
2. Yes, concentration has increased modestly over the past few decades when measured at the national level, but this does not mean there is less competition, especially in tech.
3. Yes, there is some research suggesting mark-ups are rising, but the evidence is not compelling and—to the extent it’s valid—does not apply to tech.
Leading tech services are bright spots in an economy that has been otherwise plagued by stagnant innovation and low growth. Critical analysis of ‘techlash’ claims consistently show that the hype trumps fact.
1. Yes, labor shares have fallen and wage as well as business investment have been stagnant. BUT not in the tech sector.
As the Commerce Department noted, the digital economy has been a “bright spot” for the U.S. economy. “BEA’s initial estimates show that the digital economy has been a bright spot in the U.S. economy, growing at an average annual rate of 5.6 percent per year from 2006 to 2016 compared to 1.5 percent growth in the overall economy. The digital economy accounted for 6.5 percent ($1,209.2 billion) of current‐dollar GDP ($18,624.5 billion) in 2016. When compared with traditional U.S. industries or sectors, the digital economy ranked just below professional, scientific, and technical services, which accounted for 7.1 percent ($1,326.3 billion) of current‐dollar GDP, and just above wholesale trade, which accounted for 5.9 percent ($1,102.6 billion) of current‐dollar GDP (chart 1).” (Kevin Barefoot, Dave Curtis, William Jolliff, Jessica R. Nicholson, And Robert Omohundro, “Defining And Measuring The Digital Economy,” Bureau Of Economic Analysis, 3/15/18)
Despite the broader trend in declining labor share of income in the U.S., labor share in tech/telecom has been on an upward trend. Michael Mandel of the Progressive Policy Institute writes, “The fall in the labor share has been highly uneven. In particular, the manufacturing labor share has experienced a very deep plunge. The labor share of value-added in manufacturing has fallen from 61% in 1991 to 51% in 2006 to 46% in 2016. By contrast, the labor share in the tech/telecom sector has been on a mild upward trend, with the exception of a blip from the 2000 tech boom. The labor share in the tech/telecom sector was 45% in 1991, 48% in 2006, and 51% in 2016.” (Michael Mandel, “Manufacturing, Tech/Telecom And The Falling Labor Share,” Progressive Policy Institute, 2/20/18)
Today’s leading tech/telecom companies have created more jobs than leading companies of the past. As Michael Mandel notes, “When we compare today’s tech leaders with the employment leaders of the past at a similar stage of development, it turns out that the job creation performance of the tech sector looks quite good (see the methodology appendix for an explanation of how the start date was identified). As Table 1 shows, even Facebook, the poster child for companies with high market values and low employment, looks better in historical context. Facebook had 17,048 employees in 2016, its fifth year as a public company. That doesn’t seem like much, but General Motors had only 20,000 in its fifth year of being incorporated as GM. FedEx, one of the great job stories of all time, averaged 10,000 full-time equivalent employees in 1982, its fifth year as a public company.” (Michael Mandel, “An Analysis Of Job And Wage Growth In The Tech/Telecom Sector,” Progressive Policy Institute, 9/17)
Leading tech services invest increasing amounts in R&D. “But the cost to generate that growth is going upward as well—at a faster clip. Combined spending on research and development is expected to rise 24% in 2018, while capital expenditures for the five are expected to surge by 48% compared with last year. For Big Tech, these expenses reflect the rising costs of running their current businesses while also developing new ones to stay more competitive—in a world where their most significant source of competition is mostly each other.” (Dan Gallagher, “Big Tech’s Growth Comes With A Big Bill,” The Wall Street Journal, 7/17/18)
2. Yes, concentration has increased modestly at the national level, but this does not mean there is less competition, especially in tech.
Former Obama DOJ Economist Carl Shapiro finds even national market analyses show modest increases in concentration well below worrisome levels. “The Economist summarized their findings, stating: “The weighted average share of the top four firms in each sector has risen from 26% to 32%.” See the “All Sectors” bar in the chart. What does the structure of a market with a CR4 of 32% look like? As an illustration, think about a market with a CR4 of 32% in which the top four firms have shares of 10%, 8%, 8% and 6%. There must be at least 11 more firms, since the largest any of these other firms can be is 6%, and they comprise 68% of the market. The HHI is this market is between 300 and 700. Industrial organization economists would generally describe this market as being unconcentrated.” (Carl Shapiro, “Antitrust In A Time Of Populism,” SSRN, 10/24/17)
The decline in publicly listed firms is not due to concentration but changes in U.S. firms, who focus on intangible assets and thus are less likely to be publicly listed. “Participating in public markets is not as beneficial for firms that invest in intangibles as it is for firms that invest in fixed assets, especially when these firms are small and young. If a firm builds a recognizable product and requires capital to expand its production, it is relatively straightforward for it to explain to potential investors how their money will be put to use. As the firm explains its needs, it does not endanger its ownership of its assets. It is rather difficult to steal a firm’s plants. If a firm invests in intangibles, it is much more difficult for its management to convince investors that it will make good use of its money. If the firms give too much detail, which they could be forced to do by disclosure laws if public, their competitors can use the information. If they give too little detail, investors will pay little for their shares. It is not surprising, therefore, that for such firms, participation in public markets with their disclosure requirements is likely to be onerous.” (René M. Stulz, “The Shrinking Universe of Public Firms: Facts, Causes, and Consequences,” NBER, 2018)
A modest increase in concentration does not mean decreased competition when one accounts for economies of scale. As former DOJ antitrust attorney Carl Shapiro noted, “The bigger question is what do we make of the increases in concentration that we observe. There are two very different interpretations. One interpretation is that when a market gets more concentrated, that means it’s less competitive, so we have a problem. That is not a new view; it was a fairly popular view in the ’50s and ’60s. And many people seem to be taking that view without even realizing that there is a perfectly coherent alternative view. The alternative view attributes increases in concentration to growing economies of scale, which means that the larger companies tend to be more efficient than the smaller ones.” (Walter Frick, “As More People Worry About Monopolies, an Economist Explains What Antitrust Can and Can’t Do,” Harvard Business School, 11/01/17)
Unlike the largest firms in the past, large internet firms compete against each other across a range of products and services. As economist David Evans writes, “Unlike the largest firms at previous points in time, these large Internet firms compete with each other across a range of products and services, despite each having gotten a toehold in the digital economy doing completely different things from one another. They compete in the near term (what economists call static competition): Amazon, Facebook, Google, and Microsoft, for example, all compete for advertising and promotional dollars. The largest Internet firms also compete long-term (what economists call dynamic competition): Amazon, Apple, Google, and Microsoft are all engaged in an intense race involving voice-activated platforms.” (David Evans, “Why The Dynamics Of Competition For Online Platforms Leads To Sleepless Nights, But Not Sleepy Monopolies,” SSRN, 7/23/17)
3. Yes, there is some research suggesting mark-ups—price divided by marginal cost—are rising, but the evidence is not compelling and, to the extent it’s valid, points to smaller, not larger, companies. Tech firms, in particular, have driven down prices and costs.
Assumptions in the oft-cited mark-up study do not pass the sniff test. The oft-cited De Loecker and Eeckhout model assumes that a 1-percent change in labor input produces the same percentage change in output as it did 50 years ago. This assumption in their model does not pass the sniff test. The reality is that technological gains mean that for every unit of labor inputs, output has increased, reducing labor share of revenue.
Further, even assuming the analysis is correct, mark-ups are rising due to small businesses, not larger businesses, in their story. “If the increase in markups was being driven by the dominance of large, very profitable firms, then we would expect the red line to be substantially above the black. That is, we wouldn’t expect small independent businesses to be increasing their markups. If anything they would face stronger competition and be forced to cut prices. Yet, small independent business greatly outnumber large mega-firms. By sheer numbers they dominate the unweighted average. The unweighted average, however, is where we see the largest increase. The story this chart tells is one of intense competition among large multinational firms, but ever increasing market power by small business. (“Markups And Market Power,” Karl Smith, Niskanen Center, 8/23/17)