Productivity and the Trump administration

September 29, 2016

Only half of all Americans tell public opinion pollsters that they expect today’s young people to have a better life than that of their parents, a remarkably pessimistic view given the economic and technological progress that the United States has enjoyed over the past several generations.[1] Will those pessimists be proven correct? The answer depends on how fast we can increase productivity, the amount of goods and services produced for each hour of work.

Editor's note:

The following brief is part of Brookings Big Ideas for America–an institution-wide initiative in which Brookings scholars have identified the biggest issues facing the country and provide ideas for how to address them.

Buy the book- Brookings Big Ideas for America

Productivity growth—making more stuff with less effort—is the magic elixir of rising living standards. It’s the reason why wages rise over time. It’s the reason why we have more and better goods and services than our grandparents did, even though we work fewer hours on average. Lately, productivity growth in the United States has been distressingly slow. If this persists, it’s bad news for everyone.

Although economists find it hard to ascertain which policies will pay productivity dividends in the future, they do offer a menu of policies that stand a good chance of making a difference, including taking steps to increase competition across the U.S. economy, encouraging more private investment, increasing public investment, and aggressively improving the quality of education and training. President Donald J. Trump should put some of these policies toward the top of his economic to-do list and set a goal of doubling the American standard of living by 2050.

First, a bit of economic history. From World War II to 1973, labor productivity—output per hour of work in the business sector—grew at a brisk 3.3 percent a year, and living standards (measured in GDP per capita) nearly doubled in a quarter of a century, or roughly a single generation. Around 1973 productivity growth slowed, a drought that lasted for about two decades and was accompanied by a lot of angst about the capacity of the U.S. economy to deliver for the American middle class. But then around 1995, productivity growth perked up, rising to almost post–World War II levels, a boom tied in part to the spectacular drop in the price of computing power, which also helped spread the benefits of the Internet throughout the economy. That spurt lasted for about a decade and then abated. Since 2004 labor productivity growth has been growing at only 1.3 percent a year on average, and even more slowly lately (see figure 10.1). At the pace that Federal Reserve policymakers expect the U.S. economy to grow over the next decade or so, it would take more than 70 years—close to three generations—for per capita GDP to double.


Decomposing changes in productivity growth, 1948-2015

The causes of the recent slowdown aren’t clear. Some analysts blame much of it on flawed statistics, the failure of official government data to properly account for the explosion of free Internet services (like Google Maps) and such innovations as mobile phones that have evolved to replace still and video cameras, tape recorders, radios, compasses, and flashlights. Yet it is hard to believe that mismeasurement, although undoubtedly substantial, has increased so much in recent years that it explains the entirety of the recent productivity slowdown.

So if the problem is not mostly mismeasurement, then what is it? A couple of culprits emerge from a quick glance at the data. Economics textbooks teach that productivity of workers rises when they have more capital with which to work—more and better tools, machines, computers, software. Growth in business investment spending has been substantially slower than in the past. At the same time, there appears to have been less kick from technological progress across the economy. Whether that is likely to persist is a matter of spirited debate. Northwestern University’s Robert Gordon argues that it will—that we should not expect a return to rapid, technology-driven productivity growth despite all the headlines about driverless cars and artificial intelligence. We cannot expect new technologies to match the power of electricity or even air conditioning to transform the economy, he argues. As the Wall Street Journal’s Tim Aeppel has reported, Gordon’s down-the-hall Northwestern colleague Joel Mokyr disagrees—and he is not the only one.

So if it’s not mostly mis-measurement, then what is it?

Other scholars find clues to the origins of the productivity slowdown from data on individual firms and industries. Economists at the Organization for Economic Cooperation and Development, for instance, find that firms at the technological frontier have been moving ahead but that other firms in the same industries are lagging behind. They find a large and widening gap both in manufacturing and service industries between the most and least productive firms.

A team of economists from the University of Maryland, the Census Bureau, and the Federal Reserve observe that productivity growth depends on new successful and innovative firms pushing out older, less productive firms. They point to a decay in the dynamism of the U.S. economy, particularly a decline in the number of new firms being born and old firms dying, as a potential cause of the productivity slowdown. A related line of work sees diminishing competition in the U.S. economy as a contributor to the slowdown in productivity growth. When shielded from competition, companies and industries tend to stagnate and simply count their profits, whereas competition forces old, established companies to change and helps spread productivity-enhancing innovations throughout the economy. There is evidence that there’s less competition in the United States now than there was in the past, at least partly the result of explicit government policies (such as patent rules) or government inaction (such as on the antitrust front). Academic research suggests that more than three-quarters of U.S. industries have become more concentrated over the past two decades.

It probably will take a decade or more for scholars to unravel the productivity riddle. But the new president and members of Congress need not—and should not—wait for a consensus diagnosis to take measures to treat the productivity slowdown. Suppose Mr. Trump declared that a national aspiration was to double the American standard of living over the next 35 years by pursuing policies that could nudge productivity growth up. What policies might plausibly improve the chances of reaching that goal?

Before we answer that question, there is one very important caveat to consider. Faster economic growth by itself will not solve all of America’s economic problems. The forces of globalization and technology are widening the gap between economic winners and losers. Recent history demonstrates that faster growth hasn’t always produced faster-growing incomes for working- and middle-class Americans. Recent research by Stanford’s Raj Chetty and co-authors finds only half of the children born in 1980 were at age 30 in households with inflation-adjusted incomes greater than their parents’. In contrast, 90 percent of the children born in the 1940s were making more at age 30 than their parents earned. If we are to achieve widely shared prosperity, we cannot rely exclusively on faster growth or the unfettered market. Progress will require public policies—on taxes, education, health care, labor market rules, and the like—that will counteract market forces driving wider inequality. We should pursue such policies, but that’s a subject for another essay.

Faster economic growth by itself won’t solve all of America’s economic problems. The forces of globalization and technology are widening the gap between economic winners and losers.

So if we were summoned to the Oval Office and asked what policies might help double American living standards by 2050, broadly speaking, where should we look?

Promoting competition

If absence of competition restrains productivity growth by allowing players with market power to make big profits without innovating, to squelch upstarts, and to restrict the adoption and spread of new ideas, then the obvious answer lies in government policies to foster more competition. Vigorous, thoughtful antitrust enforcement—making sure that companies do not collude and that mergers are not anticompetitive, while also avoiding counterproductive interventions in the market—is an obvious avenue. Picking the right people for key politically appointed positions at the Justice Department and Federal Trade Commission is crucial. But there are other, perhaps less immediately obvious, ways the government might stimulate competition.

A thorough rethinking of the patent system would be a good place to start. We should be sure that the United States is striking the right balance between, on the one hand, creating incentives for inventions and innovation—“to promote the progress of science and the useful arts,” as the U.S. Constitution puts it—and, on the other hand, granting papers that create monopolies that last longer than necessary, raise prices, and block technologies from widespread adoption. Some learned observers say that it’s simply too easy to get a patent. “Most of those who look closely at [the patent system] . . . think that the requirement for novelty is a little too low,” says Bronwyn Hall of the University of California at Berkeley. Others question the length of time that patents protect inventors in some fast-moving industries, and wonder if the maze of patent laws and patent litigation is doing more for law firms’ profit than for the overall economy.

Curtailing and perhaps rolling back occupational licensing is another promising option. More than a quarter of U.S. workers now require a license to do their jobs, according to the White House Council of Economic Advisers. The share of workers licensed at the state level has risen fivefold since the 1950s, and about two-thirds of this increase stems from an increase in the number of occupations that require a license. Licensing can and should protect consumers and maintain health and safety standards, but it can (and in many instances does) serve to protect professionals by limiting newcomers. As Brookings’s Martin Baily puts it: “You need a license to be a florist [in some states]; you need a license to do all kinds of things. We certainly know in health care there are a lot of restrictive practices that make it difficult for nurses to do some of the things that they probably should be able to do, technical stuff. You can’t have X-rays read in India. It’s hard for the Mayo Clinic to enter the Texas medical market because the doctors don’t want the competition. So there are a lot of these restrictive practices that are around that maybe we need to get rid of.” Excessive occupational licensing can have several unwelcome consequences, one of which is hurting economy-wide productivity growth by making it difficult for workers to move to careers or places where they could be more productive.

Of course, it’s also important to avoid policies that, though politically popular, can restrain productivity growth and end up hurting people they were intended to help. “If you think about the ongoing political campaign,” says Robert Barro, a Harvard University economist, “the policies that have been given a lot of attention include things like restrictions on trade and immigration. . . . These are not the kinds of policies you want to think about if you’re going to think about enhancing productivity.” Free trade has acquired a bad reputation in the United States, at least in part because the rhetoric about aiding those whose livelihood has been hit hard by foreign competition hasn’t been matched by effective effort. But there is ample evidence that trade—both the much-maligned competition from imports and investment by foreign companies in the United States—prods American companies to become more efficient and boosts U.S. productivity.

Promoting More Private Investment

Not all private investments pay off, of course, but the more businesses invest, the better the chances that the pace of productivity growth will quicken. Economists and governments have been trying to figure out what drives business executives to invest for decades. The great 20th-century economist John Maynard Keynes, writing in his General Theory, said that investment decisions involve something more than a straightforward cost-benefit calculation: “Most, probably, of our decisions to do something positive, the full consequences of which will be drawn out over many days to come, can only be taken as the result of animal spirits—a spontaneous urge to action rather than inaction, and not as the outcome of a weighted average of quantitative benefits multiplied by quantitative probabilities.”

The government does have a couple of levers. To some extent—economists and executives differ on exactly how much of an impact it has—uncertainty about public policies is the enemy of business investment. This argument can be overdone, but surely flirting with government shutdowns or debt default and perennially applying short-term fixes to long-term problems discourages business investment. Perhaps the most potent lever the government has is the U.S. business tax code, which is overly complicated, packed with perverse incentives, and incompatible with a world in which giant corporations spill over national borders.

Business tax reform is long overdue. Of course, any tax reform creates winners and losers, so there will be lots of political wrangling. Scrapping the entire U.S. tax code and starting from scratch, as appealing as that seems to some, is not likely to happen; the goal ought to be moving toward a business tax code that makes sense for the 21st century. This means a tax code that raises money to pay for government but also encourages investment without favoring politically connected industries, that makes the United States an attractive place to do business, and that reduces incentives to waste millions on tax avoidance while closing loopholes that companies have exploited to avoid paying taxes. No one said this was going to be easy.

Spending more on public investment

With private investment so weak and interest rates extraordinarily low, there is room for spending more on public investment today without crowding out private investment, particularly investments that the private sector will never make because it cannot capture the returns. Federal non-defense investment spending—defined by the White House budget office as physical infrastructure, research and development (R&D), and education and training—soon will fall to its lowest level in at least half a century, measured as a share of GDP. State and local spending on physical infrastructure has been muted despite very low municipal borrowing costs. Increased federal investment spending makes sense now.

Donald Trump has called for a substantial increase in federal spending on physical infrastructure. Politicians sometimes argue that this would create jobs immediately, recalling the “shovel ready” rhetoric of the 2009 Obama fiscal stimulus. But the stronger case for infrastructure and other federal investment spending is that it will pay off in the future, whether or not it creates a lot of jobs now. To be sure, it’s much easier to make the case for more federal investment spending in general than to identify ways to direct added public investment to those projects that are most likely to pay off. Bridges to nowhere—shorthand for public investment projects that have no value—will not boost productivity growth. The political pressure to spread spending across all 50 states to win votes in the U.S. Senate is enormous. This is always true with federal programs, of course, but there are ways to direct money to projects likely to have the highest returns. For instance, more than 80 percent of the National Institutes of Health funding is awarded through competitive grants. And the Obama administration’s “Race to the Top” education approach was crafted to encourage states to pursue reforms of their K-12 systems to get additional federal money. Some federal transportation funding is awarded in a similar fashion, but there’s room for more of that.

Federal R&D spending—especially on basic science and projects too risky for the private sector to undertake—can be an important stimulant to productivity growth but has been declining for decades relative to the size of the economy. That is short-sighted, and it should be reversed. It would take a 50 percent increase in annual federal non-defense R&D spending—that is, about $30 billion more a year—to return spending to the levels of the early 1980s, as a share of the economy. This is a good goal, even if it takes some time to reach it.

Improving the quality of labor

Investing in human capital is as important as investing in physical capital and infrastructure, particularly in an era when machines and computers are handling more and more chores that once only humans could do. The more capable American workers are, the more productive they are likely to be. That argues for spending more where necessary and spending more wisely on everything from pre-K to college and worker training. One good measure of the economic value of education is wages: More educated workers, on average, earn more than less educated workers. It may be hard to prove that education is the cause—as opposed to a proxy for underlying ability—but there is a strong case that sound investments in education increase a person’s productivity, the value of the goods and services he or she produces for each hour of work. This is about more than adding average years of schooling, though. Years of schooling is a familiar metric because its’s easy to track, but what really matters is harder to measure: the skills, flexibility, understanding, and habits of workers. These aspects are what we need more of—and we should pay particular attention to continuing to improve the quality of high school and high school graduation rates, and to strengthening community college and other training programs to lift the job prospects and wages of Americans who otherwise likely will end up in very low-wage jobs, if they find work at all.

When I was at the Wall Street Journal in the early 1990s, Randy Kehl, an idealistic Air Force major assigned to Vice President Dan Quayle’s Council on Competitiveness, came to see me and a colleague. The council had a well-deserved reputation as a toady for big business, as a vehicle for lobbyists to enlist vice presidential aides to derail proposed environmental or safety regulations. But Kehl wanted to “pick our brains” about a national crusade he had in mind. During the George H. W. Bush presidency, economic growth was so slow that it would have taken a century to double living standards. Why, he wondered, weren’t people outraged? Why did it have to take that long? And how would a White House initiative to double living standards more rapidly be received by the press?

We told him that journalists report on what policymakers do, rather than give the White House advice, and that we didn’t have much to give anyhow. But his questions stuck with us. He was asking the right questions, though the Bush administration never really sought to answer them. President Trump’s administration should.

Read more in the Brookings Big Ideas for America series »