What Should the Next President and Congress Do?

Remarks for the Harvard Alumni Association
Dean Douglas Elmendorf
December 3, 2016

Thank you. I am delighted to have this chance to talk with you today.

We have just lived through an extraordinary election campaign. The campaign was extraordinary in part because a complete political outsider rose from the bottom of the polls to a party nomination and ultimately to election as president: One observer wrote that this process amounted to a takeover of a party by an independent candidate. The campaign was also extraordinary because of the vitriol and rancor it generated: To give just one example, in the United States in modern times, we are unused to threats to jail opponents. The campaign was extraordinary in part because of the influence of new media: Twitter, Facebook, and WikiLeaks mattered in ways they never mattered before. And the campaign was extraordinary in part because of the cleavages it revealed in our society: People with and without college degrees, and people living in urban and rural areas, voted more differently in this election than in any election for decades.

Out of this extraordinary election, we have ended up with a new president and a partly new Congress. Control of the federal government is united in Republican hands—but that unity is somewhat illusory. To start, control of the federal government has bounced back and forth between the parties recently, with the Republicans having control of the White House, Senate, and House of Representatives after the 2004 election, the Democrats doing the same after the 2008 election, and now the Republicans doing it again after the 2016 election. However, the country has roughly split its votes between Democrats and Republicans through all of those years, so in fact the populace is quite divided. In addition, the views of the President-elect on many issues are unclear, and on other issues, his views seem to differ sharply from the views of Republican Congressional leaders. Therefore, predicting the policy changes we will see in the next few years is difficult.

I have taken an easier assignment today—to offer my own recommendations of what the next president and Congress should do. My recommendations are based on my diagnosis of what ails our country and my assessment of what changes could best heal those ailments. So, I will describe the ailments first and then offer my prescription.

To preview, I think the most important priority for the next president and Congress is to restore confidence in the American system for millions of Americans whose lack of confidence has become very clear in this election. The lack of confidence that many Americans have in our system arises partly from legitimate frustration that the country’s policies are not serving them well. It also arises from a political process that weakens rather than strengthens the relationship between our elected leaders and the people who elected them. After I have offered my perspective on these issues, I will stop and give you a chance to ask questions and to agree or disagree with what I have said.


Let me begin with my diagnosis of what ails our country. A colleague of mine frequently quotes Nietzsche as having said: “To forget one’s purpose is the commonest form of stupidity.” To avoid that sort of stupidity, recommendations for changes should begin with a statement of the problems one is trying to solve. I will highlight three fundamental challenges that our country faces—two relating to our economy and one relating to our political process.

One challenge is that overall economic growth has been slow in recent years. Since the trough of the last recession, real GDP has increased at an average annual rate roughly half that in the expansion of the 1990s. Some of that slowdown reflects the retirement of the leading edge of the baby boom generation; because we do not have a comparable surge of new entrants to the labor force, net labor force growth has slowed considerably. Some of that slowdown reflects a lingering shadow of the financial crisis and severe recession. Some of that slowdown probably results from measurement error, because an increasing share of our output is difficult to measure, like free digital products available through the Internet. But much of the slowdown stems from weak growth in productivity, which is the ultimate source of improvement in standards of living.

A second, and in some ways more significant, challenge is that many people have benefited little from the overall economic growth we have seen in the past few decades. According to official statistics, between 1979 and 2013, real GDP per person increased 72 percent, while household incomes before accounting for taxes and benefits rose 18 percent in the bottom quintile, 12 percent in the middle quintile, and 85 percent in the top quintile. Those specific numbers understate the true increases in people’s incomes because the data do not fully capture quality improvements or the introduction of new goods and services, but certainly people across most of the income distribution have gained much less than people at the top.

A shortage of economic opportunity for many Americans can be seen in other indicators as well. For example, over the past 50 years, the rate of labor force participation for men between the ages of 25 and 54—so-called “prime-age” men—has slipped from about 98 percent to about 94 percent for men with college degrees but has dropped from about 97 percent to about 83 percent for men with only a high school degree or less. Various factors are at work here, but one important factor is a shortage of desirable jobs for less-skilled men. The social consequences of this job shortage are becoming clearer as well, in the frightening rise of opioid use, the growing divergence in out-of-wedlock birth rates by education level, and the sharply different trends in life expectancy by education level.

Moreover, underlying economic forces will probably make these problems worse over time. For example, there are 1.7 million truck drivers in this country and another 1.7 million drivers of taxis, buses, and delivery vehicles. Many of those drivers are men without college degrees, and many are earning enough to put them in the middle class. But the rise of self-driving vehicles will probably cut the number of those jobs substantially over the next couple of decades. Similarly, as more developing countries fully emerge into the global trading system—aided by trade agreements but propelled mostly by market forces and their own policies—the competition in production of tradable goods and services will increase further. That will put further downward pressure on compensation for less-skilled people in this country.

A third fundamental challenge that our country faces is distrust by many Americans of their elected leaders. The nature of this distrust is somewhat complicated. Ninety percent of the members of the House of Representatives ran for reelection this year, and 97 percent of them were reelected—so fully 87 percent of the members of the incoming House are unchanged from the outgoing House. Voters are not angry with all of their elected leaders.

Yet, many Americans are indeed angry with some of their leaders; they feel ignored or disrespected. Both Bernie Sanders and Donald Trump won substantial support from primary voters who were unhappy with traditional candidates for president, and at least some of the people who voted for Donald Trump in the general election did so as a rejection of the traditional leadership class. Why this rejection?

Partly economic concerns, as I have just described. Partly prejudice against people of different race, religion, or ethnicity in an increasingly diverse society. And partly a breakdown in our political process that includes leaders’ seizing short-term tactical advantages over their opponents in ways that reduce the long-term effectiveness of the system, well-intentioned changes in the political system that have backfired and made the system less effective, and a failure to manage the government effectively.


So, we face at least three fundamental challenges—slow growth in the overall economy, a shortage of economic opportunity for people with less education, and a wide distrust of our elected leaders. What should we do about those challenges? Unfortunately, we do not have policies that we can be sure will undo these worrisome developments. However, we do have policies that will help. Let me offer ten recommendations for the next president and Congress.

First, we should increase public investment in education and training, and we should focus that investment on people who do not have good access to education and training today. When the development of new tools and techniques in farming put agricultural employment on a downward trend in this country, we moved toward universal high school education. Now that manufacturing employment has fallen to just 10 percent of total employment, and middle-wage jobs in the services sector are under pressure, we should improve the education that people receive through high school and significantly increase the education and training they receive after high school.

We should focus that investment on people who do not receive good education and training today. That focus is important both because the overall economic return would generally be higher and because we would enhance economic opportunity for people whose opportunities are currently limited. This means more preschool education, more support for primary and secondary education in low-income areas, more apprenticeships, more slots at community colleges, and more mid-career training. Apprenticeships are crucial because they help people earn money today while building skills that are valuable in specific jobs and thus open doors for future employment. Indeed, when one thinks about community colleges and mid-career training, one should think not primarily about general education but about education that is linked to specific jobs and prepares people for those jobs, because that approach is usually more effective.

As a related matter, we should encourage less-educated men to enter professions they have traditionally been reluctant to enter but are of growing importance in the economy. For example, a friend recently sent me a speech where he said that we should “make work in the caring professions acceptable and desirable to … working class white males”; I think that is correct.

Second, we should develop mechanisms to help people with low-paid or temporary jobs build retirement savings, obtain health insurance, and receive other benefits.

Low-wage workers who are employed on a long-term basis by large businesses often participate in those firms’ pension plans, health insurance, and other non-wage compensation. And because those programs have been developed for the firms’ entire workforces, they generally provide substantial benefits. However, few low-wage workers receive such benefits today because of increased outsourcing of low-wage work from large firms and the growth of alternative work arrangements such as temporary help, on-call, and contract work.

Therefore, we need to develop alternative ways for people to accumulate sufficient retirement savings, to invest those savings effectively, to withdraw the savings at a sensible pace, to obtain health insurance, to pay for health insurance, and so on. For example, we can create saving vehicles that encourage saving for retirement and make savings portable as one changes jobs. We can create markets for health insurance in which people who are not long-term employees of large businesses can buy health insurance even if they present large health risks. And more.

Third, we should use monetary policy and fiscal policy to offset any future weakness in demand for goods and services in order to keep the economy close to full employment. The recovery in employment in the past few years has helped many people find jobs and receive higher pay. It has also kept in the labor force people who otherwise would have left out of discouragement: Relative to the trend decline in labor force participation, we have seen an increase in participation during the past 3 years amounting to roughly a million people. The return to full employment has been very important in both economic and social terms.

But maintaining full employment will be difficult. In each of the past 3 recessions, the Federal Reserve has cut the federal funds rate by more than 5 percentage points. With equilibrium interest rates now exceptionally low by historical standards, the funds rate will almost certainly not have reached 5 percent when the next recession hits, so the Fed will not have as much room to cut. The president and Congress should not object when the Federal Reserve again uses nontraditional methods for achieving expansionary monetary policy, such as quantitative easing.

Even so, countercyclical fiscal policy will be even more important in the next downturn than it has been in the past. Policymakers should put in place stronger automatic stabilizers—that is, cuts in taxes and increases in spending that would take effect automatically when the economy falters—and should be ready to provide additional fiscal stimulus if necessary. A rush to normalize fiscal policy was the biggest policy error during this past economic recovery, and we should not let appropriate concern about long-term fiscal problems prevent appropriate short-term fiscal stimulus in the next recession.

Fourth, we should maintain the current scale of the government safety net of benefits and taxes. Some people assert that the safety net has become a generous hammock that has reduced work and been ineffective at reducing poverty, so cutting benefits might actually help low- and middle-income people. However, that assessment is not consistent with the evidence.

According to estimates by the Congressional Budget Office a few years ago, nonelderly households in the bottom quintile of the income distribution receive about 3 percent of total income before accounting for taxes and benefits and about 7 percent of total income including taxes and benefits. That move from 3 percent to 7 percent shows both that the safety net provides substantial help for low-income people and that low-income people are still notably worse off than people in other income groups.

It is true that cutting benefits would probably increase work by the recipients, but the evidence on labor supply responses implies that the increase would generally be small. In fact, the labor force participation rate of both prime-age men and prime-age women is lower in the United States than in many other developed countries, despite those other countries’ more expansive safety nets, so the safety net is not the primary determinant of work behavior.

It is true that the official poverty rate has declined only slightly over time, but that measure excludes SNAP (formerly known as food stamps), Medicaid, housing vouchers, the earned income tax credit, and other programs that constitute the War on Poverty for working-age families. Measures that include those benefits show a marked decline in the poverty rate during the past several decades.

It is true that the percentage of people whose pre-tax-and-benefit income is below the poverty threshold has not fallen notably over time, so the War on Poverty has failed to eliminate the causes of poverty. However, one cannot realistically expect that a fight against poverty in which most benefits have gone to retired people would overcome the effects on working-age people of technological change, globalization, and other forces that have depressed many people’s wages. Moreover, there is a growing body of evidence that children in lower-income families that receive certain benefits do better in the labor market when they grow up than children in families that do not receive those benefits. So, if we want children in lower-income families to have a stronger start in their lives, reducing benefits received by those families would be a mistake.

I am not arguing that the safety net programs are perfect as they stand. We can and should make changes to improve them. For example, a report released last summer by the American Enterprise Institute recommended a variety of interesting changes to benefit programs but did not expect those changes to reduce government spending notably. That approach is strikingly different from the budget plans of Congressional Republicans, which envision dramatic cuts in spending. The evidence shows that such cuts would be harmful to the people affected.

Fifth, we should increase federal investment in infrastructure and in research and development. Under the current caps on annual appropriations, federal investment in infrastructure and R&D—and in education and training, by the way—will soon be smaller relative to GDP than at any time in at least 50 years. That is not forward-looking, growth-oriented budget policy. To maintain the traditional level of investment as a percentage of the economy requires a substantial increase in the caps on appropriations.

Moreover, we should probably increase federal investment relative to its traditional level, because interest rates are so low and will probably stay low for an extended period. The yield on 10-year Treasury notes was 8 percent at the end of 1990 and 5 percent at the end of 2000, and it is under 2½ percent today even with the run-up of the past few weeks. That is a sea change in the economic backdrop for fiscal policy. The implications of low rates depend to some extent on the reasons why rates are low, and I have explored this issue at length in a paper with a co-author at Brookings. But the bottom line is this: With interest rates lower than in the past, the direct costs to the government of borrowing, and the indirect costs to the economy of government borrowing, are probably lower than we have been accustomed to for decades, so we should borrow more to finance federal investment than we would otherwise.

When we invest public dollars in infrastructure and R&D, we should, of course, spend the money wisely. For example, as we choose transportation projects, we should apply cost-benefit analysis consistently, and we should charge fees to users when we can. As we invest in R&D, we should focus on basic research rather than development because basic research is where private efforts tend to fall short of what is socially desirable, while development is often handled more efficiently by the private sector.

Sixth, we should put greater emphasis on buffering the adverse effects of immigration and international trade on some people. Economists have known for a long time that greater immigration and trade generally raise a country’s average standard of living. We have also known for a long time that greater immigration and trade hurt some people’s standard of living. That second effect may or may not seem important if different people are being hurt at different times and if everyone’s income is rising briskly for other reasons. But it is certainly important if the same people are being hurt by immigration and trade over and over, and if those people’s incomes are also being hurt by technological change and other forces. That is what has been happening to less-educated workers in this country.

In addition, we should recognize that people value stability as well as growth. For many people, keeping a job with which they are familiar and in which they take pride may be more important than incremental improvements in their purchasing power even if they could ultimately find another job that suits them.

Further, we should recognize that immigration affects not just the economic environment but also the social environment in ways that some people find understandably unsettling. People born outside the United States are now a larger share of the population than at any time in roughly a hundred years. For some native-born Americans, the resulting increase in diversity of customs, languages, food, and so on is very appealing. But for other native-born Americans, the new social environment is not as desirable as the previous one, and they see their country changing in ways they do not like and do not control.

I do not think that we can or should roll back globalization. But as we set trade and immigration policies in the future, we should work harder to make those policies good not only for the country as a whole but also for people who feel they are being left behind economically and socially. That means providing much more transitional protection and support as individuals’ jobs and regions’ economic bases are disrupted. It may also mean foregoing or at least slowing changes in immigration and trade that would help the country as a whole but be hard on people whose economic and social circumstances are already shaky.

Seventh, and here I turn from economic policy to the political process, the president and Congress should respect facts and expertise. If we have entered a “post-factual” politics, as some observers assert, it is because many of our elected leaders and candidates for office are pandering to voters by avoiding hard truths and experts’ judgments. When our leaders duck facts and expertise, they sometimes say they favor “common sense” solutions—and who does not? But common sense cannot tell you whether the climate is changing, or how a particular regulation will ripple through the financial system, or what technical conditions are important in international nuclear agreements. That requires expert analysis of facts. Pretending that problems do not exist or have easy solutions may attract votes in the short term, but it leads to ineffective or counterproductive policies, and it ultimately increases voters’ distrust when outcomes fall short of promises.

I realize that facts can be difficult to ascertain and interpret, as suggested by the famous line about “three kinds of lies: lies, damn lies, and statistics.” I also realize that experts can be wrong in their assessments of the world, because of analytic errors, group-think, or conscious or unconscious bias. However, the solution to these problems is not to abandon facts and expertise—it is to collect more facts, examine data more carefully, and explore more deeply both the consensus and dissenting views of experts. I am not recommending that experts make decisions about public policy; our elected leaders should make those decisions. But our leaders should base their decisions on facts and expertise.

As one important example, the next president and Congress should have a serious debate about the best policies for responding to climate change. They should ground that debate in evidence provided by scientists, economists, and other experts.

Eighth, the president and Congress should reduce the vitriol and rancor in their rhetoric and look for opportunities to compromise. At one point in the 2008 presidential campaign, a woman made disparaging remarks to Senator McCain about then-Senator Obama, and Senator McCain replied: “No, ma’am, he’s a decent family man and citizen that I happen to have disagreements with on fundamental issues.” We need more leaders talking and acting like that. Instead, our leaders demonize their opponents and accuse them of advancing ideas for taxes, health care, and so on that are not just wrong but are un-American and illegitimate. Some rhetoric of that sort is always present, but we have seen much more in recent years than in preceding decades. Such rhetoric is not just unpleasant, it is damaging, because it stokes fear and anger, and it feeds a false perception that the essence of the country is at risk from different policy choices.

The ongoing debate about the Affordable Care Act (ACA) is an important example of both the need for facts and expertise and the corrosive effects of vitriol and rancor. Republicans have quite legitimate reasons for opposing the ACA: It significantly expanded the safety net and increased regulation of health insurance, both of which conservatives would naturally resist. And the debate about the act in 2009 and 2010 was no harder fought or more exaggerated than some previous debates about significant legislation. However, the continuous, intense attacks on the law since 2010 have been well beyond previous norms. The investigations, court cases, attempts to defund the administration of benefits to which people are entitled by law, prolonged government shutdown, threats to default on the debt unless the law was changed, and so on are not typical political behavior in this country. Those attacks have been damaging because they have generally not been based on expert analysis of the consequences of the law; have not brought a serious discussion of possible alternatives; have shown an unwillingness to accept the outcome of our political process; have created an environment of hostility, not just disagreement, over health policy; and have hindered the effective operation of the government in non-health areas.

In fact, there are no alternatives to the ACA framework that would preserve the significant increase in insurance coverage we have seen at notably lower budgetary cost or with notably smaller changes in regulation. That is why Mitt Romney used the same framework in Massachusetts, why proposals for broad insurance coverage from conservative experts are more like the ACA than the pre-ACA system, and why no alternative to the ACA has been seriously pursued by Republicans in Congress during six years of controlling the House. I am not suggesting that we should avoid a vigorous debate about health policy in this country, including a vigorous debate about the ACA; I am suggesting that a debate with much less heat and much more light would be a big improvement.

More generally, our elected leaders should look for opportunities to compromise on the policies they put in place. Our Constitution requires compromise in many respects, and the rules of the House and Senate mean that compromise between the two parties is often required to enact legislation. Elected leaders—and voters—need to view the negotiations that generate compromise as effective leadership rather than weak leadership.

Ninth, we should strengthen political parties and the leaders of those parties relative to newcomers and outsiders. The reporter Jonathan Rauch had a long article in The Atlantic last summer arguing that well-meaning attempts over several decades to make the political system more open and responsive to voters had turned out badly. The presidential nominating process came to involve more primaries and open caucuses and a smaller role for party leaders. The amounts that parties could spend to support their candidates were limited in an effort to reduce the influence of money on politics. Legislative earmarks for specific projects were banned so that Congressional leaders could not gain support for legislation through private deals. And other changes were made as well.

Those changes did make our political system appear more open and responsive, but they have actually hurt the ability of our leaders to govern effectively. The leaders of the House and Senate and the leaders of political parties recognize the importance of maintaining working relationships with leaders on the other side of the aisle and the importance of maintaining some consistency in the operations of the government; by contrast, individual politicians can often gain more from unconstructive obstruction. In addition, leaders understand the demands of public offices and often know well many contenders for those offices, so they have useful judgment about who would be most effective in office; by contrast, most voters do not have as much information. So, weakening Congressional leaders and party leaders hinders effective governance. Michael Gerson, who was a speechwriter for George W. Bush, recently quoted the 10th Federalist paper as saying: “Men of fractious tempers, of local prejudices, or of sinister designs may, by intrigue, by corruption, or by other means, first obtain the suffrages, and then betray the interests, of the people. … [It is better to mediate public views through] a chosen body of citizens, whose wisdom may best discern the true interest of their country.”

We have weakened the leaders who could mediate public views. Reducing the ability of party leaders to spend in support of their candidates and reducing the ability of House and Senate leaders to gain support for legislation through earmarks has made the leaders’ jobs more difficult. I watched this problem up close in my years working for Congress. And political scientist Julia Azari recently wrote: “The defining characteristic of our moment is that parties are weak while partisanship is strong.”

I am not an expert at the mechanics of the political process, but various analysts have recommended changes that would increase the power of the political parties and their leaders, and we should explore those changes.

Tenth, and finally, we should make the government operate more effectively. Addressing nuts-and-bolts issues of public administration is less glamorous than developing grand strategies, but it is no less important to rebuilding citizens’ confidence in their government. Paul Volcker quoted to me in the spring a comment by Thomas Edison that “vision without execution is hallucination.” We need less hallucination in our government.

Let me give some examples. For Congress, this means confirming presidential nominees more quickly and no longer shutting down the government or threatening to default on the debt. For the president, this means demanding rigorous cost-benefit analysis of regulations and further developing the performance metrics now being posted at performance.gov. For the president and Congress together, it means pursuing evidence-based policies, for example through the work of the new Commission on Evidence-Based Policymaking that Speaker Ryan and Senator Murray designed. And for the president and Congress together, it means reforming the Civil Service so that bad performers can be fired more easily and good performers can be rewarded better; to have an effective government, we need to attract more of our most talented people to work for it.

That concludes my recommendations to the next president and Congress. These 10 changes in economic policy and the political process will not be easy to put into practice. They will require our leaders to stand up to some pressures and temptations they face. But I think that leaders who stand up in these ways ultimately may be rewarded by voters. Thank you.