Over the past 35 years, the economy has been transformed by technological change. The computing revolution and the rise of the Internet have changed the nature of product competition, reduced communication costs dramatically, and allowed the digitization and codification of processes. These changes have affected the returns to different types of labor, the role of labor itself, and the nature of bargaining between workers and companies.

Accompanying this technological change and augmented by it, globalization has proceeded apace. Capital mobility across borders has continued to increase, both in terms of portfolio investments that cross borders and in terms of the reach of multinational corporations. International trade in goods and services has continued to expand, fueled by decreases in information costs, technological changes, and greater policy openness to trade in many countries. Migration, while more limited by national barriers, has also had an important impact in many places over this time period.

Over this period of technological change and globalization, the role of labor and capital has also changed in important ways throughout the global economy. A generation ago, students of economics were often taught that the labor share and capital share of all income generated in the economy should be expected to be roughly constant over long periods of time, with about 70 percent of national income accruing to labor and 30 percent to capital. Labor and capital are both inputs into the production process, but the income received by workers and capital-owners likely accrues to different economic classes of people, and so this constancy was reassuring to those who worry about workers’ evolving standards of living.

However, several different sources of U.S. government data indicate that this constancy has broken down in recent decades.

Researchers have confirmed this decline across countries, using careful methods and controlling for other influences that could affect these trends. Economists focusing on the corporate sector have found that the labor share of income declined by 8 percentage points over the period 1980-2012, from 65 to 57 percent. This phenomenon is not a distinctly U.S. trend, but rather one that has affected most economies, regardless of their initial income level. A recent estimate suggests that the labor share in G20 countries has fallen at a pace of 0.3 percentage points per year between the late 1980s and the late 2000s, and several independent analyses have confirmed similar trends. Throughout the world, owners of capital are receiving a greater share of national income, so growth in national income is no longer sufficient to ensure increasing worker living standards for labor as a whole.

These facts are startling, but at least two caveats are worth bearing in mind. First, for many types of income, it may be futile to separate labor and capital income. For example, many of those in the top of the income distribution have tremendous wealth that has been generated from streams of entrepreneurial income with both labor and capital elements. For example, Mark Zuckerberg’s wealth likely represents some combination of returns to his labor, his “human capital” (or knowledge), his risk-taking, and his ownership of capital (servers, office space, etc.) that have been invested in Facebook. Second, the labor share alone is not a measure of inequality. While capital income is far more concentrated than labor income, the level of inequality depends on the distribution of labor income, the distribution of capital income, and the labor share. In the United States, recent trends of increased income inequality involve all three elements.

Even so, the declining labor share is problematic because it is occurring in a context of rising income inequality and stagnant wage growth for many workers. In the United States, all data sources indicate increases in income inequality in recent decades, and the more capital income is included in the data source, the greater the increase. Treasury data indicate that the top 5 percent of tax units have increased their share of national income by 13 percentage points, from 24 percent to 37 percent, over the period 1986 to 2012. Dividends and capital gains income are particularly concentrated, with the top 5 percent of tax units reporting 68 percent of dividend income and 87 percent of long-term capital gains in 2012.

The problem of wage stagnation is widely recognized, and it has been particularly challenging in the period surrounding the financial crisis and the Great Recession (although the past few years have seen some improvements in the United States). For example, a recent report by the McKinsey Global Institute indicates that two-thirds of households in 25 advanced economies experienced flat or falling incomes between 2005 and 2014. This occurred during a period when aggregate economic growth remained positive, illustrating the importance of income distribution in determining economy-wide income patterns.

What Are the Consequences?

Such troubling trends in labor market outcomes have potentially grave consequences for society. Stagnant incomes are harmful in times of weak aggregate demand, since income gains that are concentrated at the top are less likely to fuel consumption and aggregate demand. But beyond that impact, increased inequality amidst wage stagnation creates social tension and discontent. When workers’ wages fall short of expectations, and there are large cohorts of workers who are not as well off as their parents, people are more likely to turn to populist solutions to express their dissatisfaction. Indeed, the rise of populist candidates like Bernie Sanders and Donald Trump are testament to the depth of dissatisfaction, even if some of the policy solutions on offer are infeasible, misdirected, or even harmful. The U.S. Congress continues its trend of ever-increasing polarization and dysfunction, and political polarization extends beyond the United States as both far-left and far-right parties are ascendant through Europe.

Concentration of incomes also creates disproportionate political power for affluent groups, who are able to hire lobbyists to influence the policy process as well as lawyers and accountants to best work around existing policies. Affluent groups also enjoy greater access to policymakers by frequenting the same elite institutions and circles. There are also important implications for the way the government is funded. The ideal progressivity of our tax system is affected by the fact that so much of the recent gains in national income have accrued to those at the top of the income distribution. Since most of the federal tax burden falls on labor income through the income and payroll taxes, we also rely on a smaller tax base as the share of labor income shrinks. These policy implications are discussed in more detail below.

Causes of Labor’s Decline

Six explanations have been offered for the declining labor share. First, the computing revolution has substantially lowered the price of investment goods (the plant, machinery, and equipment that make production possible). This has been a worldwide phenomenon. If companies respond to the declining price of investment goods by disproportionately increasing their use in production, then the capital share of income will rise. For example, as computing power becomes cheaper, companies may replace workers that used to perform data entry work with machines. A related explanation is that technological progress has been capital-augmenting. In this explanation, computing also plays a role, by increasing the productivity of capital investments. So capital investments are not only less expensive, but also more productive. In this example, firms increase their demand for computers, because computers process more data than ever before, and this also displaces demand for manual data entry by employees. Overall, the demand for capital increases, raising the capital share of income.

Second, international trade and international capital mobility also play a role in these developments. In high-income countries, increased competition from countries with large, inexpensive labor forces may lower wages and reduce the labor share of income. While many studies find that international trade is not the dominant influence on labor market outcomes of rich countries, there is still evidence that it is important, and there are relationships between international trade and technological change that are difficult to disentangle. For example, competition from low-wage countries may accelerate the pace of technological innovation in rich countries, as they compete through innovations that economize on labor. To the extent that the United States produces the same products as those in low-wage countries, it is often by producing them with far more capital-intensive processes. Agricultural goods are made with highly capital-intensive methods that rely on computer-guided farm machinery and technological sophistication in fertilizers and seeds. Shoes are made in the United States, but in highly mechanized factories that bear little resemblance to their less-developed country counterparts.

This raises the question of why the labor share is also observed to be declining in many less-developed countries. One possible explanation is that international trade increases the demand for high-skilled workers and capital in all economies, due in part to the disintegration of the production process. Products (or steps in the production process) that appear to be labor-intensive from the perspective of rich countries may actually be capital-intensive from the perspective of poor countries that are less well-endowed with capital. Thus, the growth of international trade may increase the demand for high-skilled labor and capital in both types of countries.

Third, the greater mobility of capital may reduce workers’ bargaining power more generally, as companies move abroad (or merely threaten to), restraining wage growth. Unionization rates have been declining in many countries, and the unionization rate in the United States has declined from 31 percent in 1960 to 11 percent today. These numbers mask an even greater fall in private-sector unionization rates, now at 7 percent. Reductions in unionization rates have occurred across developed countries; the average unionization rate for workers in OECD countries was 35 percent in 1960 and 17 percent in 2014. Other institutional considerations may also play a role in the declining labor share, including labor market regulations, minimum wage laws, and other features of the safety net.

Fourth, the so-called “Superstar Effect” attempts to explain why those at the top of the income distribution reap outsized rewards. Recent analyses have also emphasized how both globalization and technical change can affect the returns at the very top of the income distribution by boosting the real and relative earnings of “superstars” and capital. The combination of larger world markets and the ease of digitizing and distributing information creates outsized returns to the most productive talents in society relative to others that merely face increased competition.

As an example, the best movie stars and athletes earn premium returns, since their talents are now accessible to their fans throughout the world, who can watch videos and sporting events remotely. This increases the size of the entertainment sector, but the actors and athletes that are slightly less talented may earn less, since there are increased costs associated with paying those that own the intellectual property of the superstars. These owners include the stars themselves as well as those investors and producers that control the rights to the worldwide distribution of their products.

Fifth, rents—that is, the excess profits that arise from entrepreneurial success—play an important role in this story of the declining labor share. Much of the inequality in both labor and capital income, as well as the explanation for the rising capital share, may be due to what Victor Fleischer terms “alpha” income, or income from entrepreneurship and the risky returns to human ingenuity. Such income may take the form of founder’s stock, carried interest and partnership equity, or simply outsized salaries.

Alongside entrepreneurial rents, there is substantial evidence of an increased importance of market power and corporate rents for large corporations. Researchers show that declining labor shares have been associated with a large, pervasive increase in corporate savings. Over the previous 30 years, corporate savings have increased corporations’ share of total global savings by about 20 percentage points. In the United States, Treasury economists calculate that the fraction of the corporate tax base that consists of these excess profits averaged 60 percent from 1992 to 2002, but has since increased to about 75 percent over the period 2003-2013.

Indeed, the recent era has been marked by growing firm concentration, with large firms earning the lion’s share of profits. McKinsey Global Institute calculates that the top 10 percent of the world’s public companies earn 80 percent of the profits, and firms with more than $1 billion in revenues account for 60 percent of all global revenues and 65 percent of market capitalization. In the United States, corporate profits in recent years are higher as a share of GDP than they have been at any point in the last 50 years, either in before-tax or after-tax terms. Since 1980, corporate profits after tax have increased 4 percentage points, from about 6 percent of GDP to about 10 percent.

Sixth and last, tax policy is a factor in labor’s declining share of income. Thomas Piketty, Emmanuel Saez, and their co-authors note that there is a tight relationship between increases in the share at the top of the income distribution and the evolution of tax policy. Over the past century, the top shares of the income distribution have followed a “U” pattern in several countries, including the United States, Australia, Canada, and the United Kingdom, where the share of income earned by the top 1 percent declined steadily until the 1970s, and then increased since the 1980s. At the same time, top marginal income tax rates moved in the opposite pattern, increasing in the first half of the twentieth century, and declining steeply since 1980.

One possible explanation for this correlation is that top incomes are more likely to be hidden in order to avoid taxation when tax rates are high, so high incomes are merely more visible in the presence of lower tax rates. A second explanation, and the one favored by these authors, is that changes in top marginal tax rates change the bargaining process between workers and managers. When top tax rates decline, this increases the incentive for highly compensated earners to bargain aggressively for pay, increasing incomes at the top of the income distribution. Simply put, high tax rates serve as a break on “surplus extraction” by high-income earners, but this break is relaxed just as the world economy changes in a way that increases demand for capital, the most highly skilled, and superstars, so compensation at the top surges accordingly. Still, the correspondence between tax rate cuts and surging income inequality could also reflect evolving social norms, or it could simply be a coincidence.

What Should Be Done?

As in medicine, perhaps the most important rule is simply to do no harm. While globalization and technological growth have played important roles in these dramatic changes in the economy, curbing trade or technology is likely to cause more harm than good.

Trade, like computers, creates both winners and losers. Unfortunately, the workers who would have made the goods now made by foreign workers (or domestic robots) are harmed by it. But workers in export industries benefit, and consumers benefit from price reductions on virtually every product they consume. Increased foreign competition prevents domestic firms from wielding undue market power. Economic growth abroad creates more stable societies and alleviates world poverty. Close, mutually beneficial economic ties between countries build peaceful relationships and reduce needless antagonism among nations. Addressing global policy problems like climate change will require an international community that is more interested in building bridges than walls.

Indeed, the country as a whole benefits from trade. When the international community wants to punish countries for wrongdoing, it sanctions them by reducing their ability to engage in international trade, often greatly harming their populations. Protectionism is like sanctioning ourselves, a form of economic self-harm. We should no more throw away the benefits from trade than throw away our computers.

Luckily, there are far better ways to address these troubling labor market developments. Targeted income redistribution, such as expansion of the earned- income tax credit, coupled with larger tax contributions from those that have most benefited from the economic growth of the previous decades, would be a sensible way to help U.S. workers. Investing in infrastructure and education would help buttress worker productivity and living standards. Labor laws and regulations can allow a flexible and dynamic economy, while also aiming to protect worker interests and buttress their hand in bargaining.

The Role of Tax Policy

The tax system is our most important instrument for affecting the distribution of income, and reforms of tax policy can help make sure that changes that benefit the whole economy, such as international trade and technological progress, also benefit all those within the economy. Such tax policy changes will involve smaller gains in after-tax income for those at the top of the income distribution, coupled by reductions in tax burdens for those whose incomes are stagnating.

At present, there are aspects of our tax system that work well and aspects of the tax system that are in desperate need of repair. On the labor income side, relatively small tweaks can help address these labor market trends, including a more generous earned-income tax credit as well as middle-class tax cuts.

Perhaps most problematic is our taxation of capital income. Here we have a weak and porous corporate tax system, coupled with little or no capital taxation at the level of the individual taxpayer. Capital income is highly tax-preferred at the individual level, taxed at top rates that are about half that of labor income, if it is taxed at all, which it often isn’t. Recent research suggests that as little as one quarter of U.S. corporate equity income is taxable through the income-tax system. Most capital income grows tax-free in tax-preferred retirement accounts, pensions, 529 accounts, or in nonprofit endowments.

This leaves the corporate tax as the dominant tax on capital income. Yet the U.S. corporate tax collects relatively little revenue compared to GDP, compared with peer nations, in spite of the fact that U.S. corporate profits have been increasing rapidly and stand at historically high levels. Several factors are responsible for this disconnect: Our tax base is narrow, there is a large tax preference for non-corporate income, and multinational corporations have become increasingly adept at shifting profits offshore, where they accumulate at very low tax rates.

These problems, as well as proposed policy solutions, are discussed in far more detail in my recent report for the Washington Center for Equitable Growth on “Strengthening the Indispensible U.S. Corporate Tax.” Protecting the corporate tax base is important for many reasons: Capital taxation has an important role in an efficient tax system, it is difficult to crisply distinguish capital and labor income for purposes of taxation, and capital income is increasingly taking the form of rents, or excess profits. Importantly, capital income is far more concentrated than labor income, and the corporate tax is likely to burden capital or shareholders far more than it burdens workers. This stands in contrast to most other federal taxes, where the burden falls entirely on labor.

Several key reforms would strengthen our corporate tax system. Helpful incremental steps would include a minimum tax of foreign income earned in low-tax countries, “earnings-stripping” rules that would make it more difficult to shift profits to low-tax countries, and anti-inversion rules that would limit the ability of firms to move their headquarters to low-tax jurisdictions for tax purposes. More fundamental reforms such as worldwide consolidation of U.S. corporate tax returns would be even more effective.

These corporate tax reforms should be coupled with changes that harmonize the treatment of different types of income, in order to minimize tax-gaming and to make our tax system more efficient. Debt-financed and equity-financed investments should be treated more evenly, corporate and non-corporate business income should be taxed more similarly, and capital and labor income should face the same tax rate. In addition, capital income taxation should be buttressed by fixing loopholes in estate taxation and eliminating the step-up in basis at death, which allows many capital gains to escape taxation entirely, disproportionately benefiting those at the very top of the income distribution receiving inherited wealth. Policymakers should also consider limiting the size of tax-free retirement accounts.

These tax reforms would update our tax system to make it more suited to the modern global economy, making sure that economic growth benefits most households. Such tax reforms address the troubling side of the changes in our economy over the previous 35 years, the diminishing share of labor income, and increasing income inequality.