Table of Contents Hide
On a sunny morning in December 2013, as Google employees boarded the bus that would take them on their daily commute from Oakland to the company’s Mountain View headquarters, protesters moved in. One unfurled a bright blue banner bearing the words “F— Off, Google.” Some handed out pamphlets explaining their anger: “While you guys live fat as hogs with your free 24/7 buffets, everyone else is scraping the bottom of their wallets, barely existing in this expensive world that you and your chums have helped create.” News outlets reported that people had thrown rocks and a bus window had been smashed.
Across the bay in San Francisco, Apple employees filing onto their own shuttle encountered a similar demonstration. There were several protests that winter, most attended by just a few dozen activists. Nonetheless, that was enough for Google to beef up security, and for one Googler to tweet from a bus surrounded by protestors that he and his coworkers were “under siege.” A protestor countered, “We’re here to send a message to the rich tech companies that their business has ramifications and consequences.”
The so-called Google bus protests failed to capture the national imagination the way Occupy Wall Street had two years earlier. The demonstrations were far smaller and seen as a mostly local phenomenon — one region’s reaction to rising rents and urban gentrification. But the episode highlighted an economic trend that the Occupy movement missed and that deserves wider attention. What the bus protesters understood — and what new research demonstrates — is that there’s more to income inequality than the 1% versus the 99%, CEOs versus typical workers, or the finance sector versus the rest of the economy. There’s more to it than skilled versus unskilled workers, although skills account for a lot. The real engine fueling rising income inequality is “firm inequality”: In an increasingly winner-take-all or at least winner-take-most economy, the best-educated and most-skilled employees cluster inside the most successful companies, their incomes rising dramatically compared with those of outsiders. This corporate segregation is accelerated by the relentless outsourcing and automation of noncore activities and by growing investment in technology. It’s no accident that a company like Google became a flashpoint for controversy; its employees fare much better than workers almost anywhere else.
Of course, some say that income inequality is not a problem, but even in the business world their ranks are shrinking. In a 2015 survey, 63% of Harvard Business School alumni said that reducing economic inequality should be a high or very high priority for American society; only 10% said it should not be a priority at all. The Brexit vote and the recent U.S. presidential election shined a spotlight on the ways in which income inequality is giving rise to a global populism that threatens to destabilize governments and economies around the world.
If we want to truly understand income inequality — if we want to mitigate it and its pernicious effects — we must look beyond CEO compensation and tax policy and consider the role played by firms and their hiring and compensation policies for ordinary, non-millionaire workers. This is not a simple morality play in which evil companies are pitted against the middle class. There is nothing nefarious about Google’s goal of being the global leader in software and machine learning, or in its hiring the best employees it can find. Yet the result of countless strategic decisions in pursuit of such goals by Google and other elite companies throughout the world — not just in tech — has been to raise the compensation of some workers far more than others.
It’s time to turn our attention from comparing individuals’ fortunes to considering differences between firms.
FIRST, INEQUALITY DATA
Let’s start with a review of what most people already know about the inequality debate, even if they haven’t read all 704 pages of Thomas Piketty’s runaway best seller, Capital in the Twenty-First Century.
Since 1980, income inequality has risen sharply in most developed economies, especially in the United States. Much of the public discourse has focused on the gap between the top 1% and everybody else. In 1980, the top 1% of adult earners in the U.S. made $420,000 a year, on average (before taxes and measured in 2014 dollars) — 27 times as much as the average for the bottom 50% of earners. Today the top 1% of earners make an average of $1.3 million a year — 81 times as much as the average for workers in the bottom half. (See the exhibit “Inequality Between Individuals Has Risen.”)
But it’s not just the top 1% who are pulling away. The gap between workers with a college education and ones with only a high school diploma has increased dramatically as well. In 1979, the average annual salary for American men with a college degree was $17,411 higher (after adjusting for inflation) than the average for men with a high school degree. By 2012, the gap had nearly doubled, to almost $35,000; the gap between women with college degrees and those with high school diplomas nearly doubled as well.
Meanwhile, the bottom half of earners in the U.S. have seen virtually no growth in earnings, before taxes and social-security transfers, despite a rise in the number of hours worked. The problem of stagnant incomes for this group is not sluggish GDP growth, as is often suggested; the U.S. economy produces far more each year than it did decades ago. What matters more, a study by Raj Chetty and colleagues demonstrates, is rising income inequality. Their research shows that only half the workers born in 1980 — today’s 36-year-olds — make as much money as their parents did at the same age. When the researchers ran simulations testing the effects of diminishing GDP growth versus rising income inequality on wage stagnation, the percentage of 36-year-olds who did better than their parents jumped to 80% when income inequality was held steady, but only 60% did better when GDP growth was restored to the older, faster rate.
NEXT, LOOKING WITHIN AND BETWEEN FIRMS
But that’s not the whole story. Over the past several years, economists have begun to examine pay gaps between and within firms to see how company strategy and corporate trends affect the broader rise of inequality. The findings from this new area of study are striking and help explain why incomes have risen so much for some and not at all for others. They also explain why so many executives, managers, and other well-paid workers have failed to notice the growing disparity.
Companies can contribute to rising income inequality in two ways. As we’ve just discussed, pay gaps can increase within companies — between how much executives and administrative assistants are paid, for example. But studies now show that gaps between companies are the real drivers of income inequality. Research I conducted with Jae Song, David Price, Fatih Guvenen, and Till Von Wachter looked at U.S. employers and employees from 1978 to 2013. We found that the average wages at the firms employing individuals at the top of the income distribution have increased rapidly, while those at the firms employing people in the lower income percentiles have increased far less. (See exhibit “Inequality Between Companies Is Also Growing.”)
In other words, the increasing inequality we’ve seen for individuals is mirrored by increasing inequality between firms. But the wage gap is not increasing as much inside firms, our research shows. This may tend to make inequality less visible, because people do not see it rising in their own workplace.
This means that the rising gap in pay between firms accounts for the large majority of the increase in income inequality in the United States. It also accounts for at least a substantial part in other countries, as research conducted in the UK, Germany, and Sweden demonstrates.
WHY DID ALL THIS HAPPEN?
Gaps among companies are growing for a variety of reasons. A big one is that some industries have boosted salaries more than others — law firms have increased their pay more than retailers, for example. But the gap between high-paying firms and poorly paying ones in the same industry has risen considerably too.
Why are some firms paying better than others? It could be that some are just more generous — paying their workers higher amounts than other firms pay for the same work — although that would surprise economists, who assume that the “law of one price” ensures that similar workers are paid similarly. The more likely explanation, our research suggests, is that companies are paying more to get more: boosting salaries to recruit top talent or to add workers with sought-after skills. The result is that highly skilled and well-educated workers flock to companies that can afford to offer generous salaries, benefits, and perks — and further fuel their companies’ momentum. Employees in less-successful companies continue to be poorly paid and their companies fall further behind.
I believe that much of the rise of between-firm inequality, and therefore inequality in general, can be attributed to three factors: the rise of outsourcing, the adoption of IT, and the cumulative effects of winner-take-most competition.
Outsourcing. In considering the effect of outsourcing on inequality, it’s instructive to look at GE. In the 1960s, it employed manufacturing workers, line managers, executives, janitors, administrative staff, and many other types of workers. Over the past several decades, it has automated or outsourced a wide range of functions. Yet during that time its head count has stayed relatively constant, at about 300,000 employees. That means GE has hired more engineers and coders, doubling down on its core competency as the premier maker of high-tech industrial equipment and paying other firms to handle tasks outside its core.
Or consider Google. It aggressively recruits software engineers and data scientists, pays them generously, and lavishes them with perks, such as free transportation on buses like the ones protestors threw rocks at. But its bus drivers don’t necessarily see all of those benefits — they’re contractors, not employees.
Highly skilled workers flock to companies that can afford to offer generous salaries while employees in less-successful companies fall further behind.
In 1990, C.K. Prahalad and Gary Hamel published “The Core Competence of the Corporation” in Harvard Business Review, urging companies to focus strategy not on products or markets but on “the collective learning in the organization” — GE’s expertise in industrial equipment, for example, or Google’s in software. Companies seem to have listened. One study found that from 1980 to 2008, the share of workers in Germany employed by temp agencies or cleaning, logistics, or security firms more than tripled, from roughly 2% to 7%.
As companies focused on their core competences and outsourced noncore work, the corporate world began to divide between knowledge-intensive companies such as Apple, Goldman Sachs, and McKinsey and labor-intensive companies such as Sodexo, which provides food service and facilities management services. Workers with lots of education and desirable skills were hired in the knowledge sector, with high pay, perks, and benefits. Less-educated workers got jobs in labor-intensive firms, where pay was stagnant or even falling and benefits such as health insurance were hardly guaranteed. Employees from these two types of firms often work in the same building, but they’re no longer in the same orbit. And when it comes time for the holiday party, the struggling contractors are nowhere to be seen.
Outsourcing isn’t the only factor driving firm inequality, however. If it were, inequality would have risen only between firms in different industries: between Goldman Sachs and Sodexo, for instance. But over the past 35 years, top firms have been pulling away from average companies in their own sectors, in terms of both profits and productivity. (See the exhibit “Winner Take Most.”)
IT and automation. This dynamic appears to be driven largely by technology. My research and other studies suggest that between-firm pay inequality has grown faster in industries that spend more on IT. Investments in technology allow successful online firms to rapidly scale up and reap the benefits of network effects. In this way, leading companies such as Amazon and Facebook dominate their markets. Offline, improved enterprise software and automation of routine tasks make it far easier to manage and grow large businesses, from Shake Shack (burgers) to Xiaomi (smartphones).
Some experts, most notably Jason Furman, the chair of the Council of Economic Advisors during the last three years of the Obama administration, argue that the rise of these “superfirms” results from a lack of competition. Industries have become more concentrated, and the number of new businesses has declined. But industry concentration doesn’t necessarily imply a lack of competition. In some sectors, such as manufacturing, evidence suggests that increased global trade has strengthened the ferocity of competition in recent decades, leading to more domestic competition as fewer and fewer U.S. firms survive. In my view, it’s unclear whether competition has increased or decreased in aggregate over the past few decades.
Winner-take-most competition. What is clear is that over the past 35 years, firms have divided between winners and losers, and between those that rely heavily on knowledge workers and those that don’t. Employees inside winning companies enjoy rising incomes and interesting cognitive challenges. Workers outside this charmed circle experience something quite different. For example, contract janitors no longer receive the benefits or pay premium tied to a job at a big company. Their wages have been squeezed as their employers routinely bid to retain outsourcing contracts, a process ensuring that labor costs remain low or go ever lower. Their earnings have also come under pressure as the pool of less-skilled job seekers has expanded, due to automation, trade, and the Great Recession. In the process, work has begun to mirror neighborhoods — sharply segregated along economic and educational lines.
WHAT CAN BE DONE?
The picture of firm inequality I’ve sketched here doesn’t invalidate other theories of income inequality. On the contrary, it supports many of them. And to be clear, although firm inequality is a large part of the puzzle, it’s not everything. Notably, it doesn’t explain the rise of the 1%. That’s a separate and important trend that’s been well documented elsewhere. But in shifting the focus from individuals to companies, several unique recommendations emerge, for both policy makers and executives.
Focus on antitrust. For a start, a renewed focus on antitrust issues may be a good idea, to the extent that lack of competition can exacerbate the winner-take-most dynamic. But because that dynamic is global, more-robust antitrust policies alone will not solve the problem.
Reframe the policy debate. Second, established policies should be reassessed through the lens of firm inequality. Take the pay disclosure rules prescribed in the Dodd-Frank Act of 2010, which require companies to disclose the ratio of the CEO’s pay and the median worker’s. Intended to mitigate within-firm wage disparities, the measures may have limited value in light of the research showing that gaps within firms are not the main contributor to income inequality. They may even have unintended consequences: Savvy CEOs may very well decide that the easiest way to raise the salaries of median workers is to outsource more low-wage work, thus lowering the ratio of CEO-to-median pay. As an economist, I’m sympathetic to policies that make more corporate information publicly available, but if the Dodd-Frank disclosure rules are ever closely scrutinized, we may see that they could backfire.
Reframe corporate decision making and hiring practices. Executives at well-paying firms should recognize the extent to which their strategies and practices contribute to income inequality. You do not need to be a hedge fund manager to be on the winning side of some very profound economic divides. To be sure, companies should not start insourcing all services or stop automating tasks, but senior leaders and those responsible for hiring should understand the role their decisions play in the larger economy.
Invest in education. Perhaps the single biggest priority for policy makers and corporations should be education. Since between-firm inequality appears to be largely driven by workers being sorted according to education and skills, the best way to set people up for success is to ensure that they have the skills needed to compete in the 21st–century job market. It’s become fashionable lately to point out that more education for ordinary workers wouldn’t alter the extraordinarily high incomes of the top 1%. That’s true: Having more college graduates would do little to rein in the incomes of hedge fund managers and CEOs. But for the equally important inequality between the non-fabulously wealthy and the poorest — between the top 20% and the remaining 80% — education and skills training are clearly part of the solution.