Cato Online Forum

Toward Faster, More Inclusive Growth

By William A. Galston
November 2014

The United States has an economic growth problem, and it started before the Great Recession. Between 1949 and 2000, the annual rate of growth exceeded 3 percent 34 times — two years out of every three. During the eight Clinton years, growth reached 4 percent five times and fell below 3 percent only twice. Since 2000, by contrast, growth has reached 3 percent just two times and did not reach 4 percent at all, even during 2003-2007, the peak years of the recovery from the 2001 recession. Since the beginning of the recovery from the Great Recession of 2008-2009, annual growth has not reached 3 percent until this year (perhaps).  

The big picture puts these details in sharp relief. Between 1949 and 2000, the economy grew at an average rate of 3.6 percent annually. Since then, growth has averaged only 1.8 percent.

In the economic circumstances of recent decades, only a sustained period of robust growth has been enough to raise wages and household incomes. Unless the economy can resume the more robust growth of the second half of the 20th century, U.S. workers will be hard-pressed to regain the ground they have lost, let alone offer the prospects of something better for their children.

What follows are a number of suggestions for how to spur faster growth — and to ensure that the gains from growth are more broadly shared throughout the American workforce. Over the next decade, there is one overriding challenge — recreating an economy in which growth works for everyone, not just a favored few. If we solve that problem, we can sustain a generous social order at home and our role as the guarantor of peace and security abroad. If we fail, much that we have taken for granted since the end of World War II will be at risk.


I begin with some macroeconomic considerations. Among Democratic-leaning economists, there are two broad schools of thought about the recovery from the Great Recession. Some argue that although the 2009 stimulus should have been larger and lasted longer, the painfully slow recovery will eventually bring the U.S. back to full employment, and eventually the new normal won’t differ fundamentally from the old normal.

Led by former Treasury Secretary Lawrence Summers, other Democratic-leaning economists have adopted a dissenting view. They argue that the 2007-09 crisis inflicted structural damage that, left uncorrected, will permanently reduce growth, jobs, and incomes. This thesis restores the contemporary relevance of John Maynard Keynes’s “liquidity trap” and Alvin Hansen’s theory of “secular stagnation.” But more than that, it proposes a mechanism — hysteresis — to explain how this permanent effect occurs.

Hysteresis is a technical term for a simple idea: temporary conditions can have long-term effects. That’s why timing often matters. If you deal with certain problems promptly, you can solve them. If you wait too long, you can’t. You can get stains out of carpets if you work on them as soon as they happen. Wait too long and the stains “set,” rendering them much harder to remove.

Much the same is true of economies. If workers are unemployed or underemployed for too long, they can become disconnected from the labor market. The resulting economic slack lowers demand, leading businesses to invest less and setting a vicious circle in motion.

For example, recent research shows a link between long-term unemployment and obesity: being out of work for an extended period makes many people depressed; depression leads to decreased physical activity and increased consumption of comfort food. Obese individuals are less able to meet job-related physical qualifications, and potential employers often regard them as lacking in self-control and self-respect.

Policies that would be unwise in normal circumstances may be necessary when hysteresis occurs. Not surprisingly, secular stagnation theorists have offered boldly heterodox proposals. If we need an extended period of negative real rates to restore full output and employment, we can negate the zero bound only by raising the target rate of inflation. If we need to reduce the supply of savings, developed economies should institute higher retirement ages and reduce the policy uncertainties that induce precautionary savings. If the demographics of the developed world contribute to economic stagnation, advanced economies should revitalize their workforces by opening the gates to immigration — the reverse of what their current populations favor.

In a paper kicking off the Center on Budget and Policy Priorities’ program on full employment, a powerhouse trio of Lawrence Summers, Brad DeLong (University of California, Berkeley) and Laurence Ball (Johns Hopkins University) argue that when specific conditions are satisfied, tax cuts can pay for themselves by boosting potential output through their effects on investment, labor force participation and other key determinants of long-term growth, which leads to a long-term increase in tax revenues. If the authors are right, there can be a fiscal-policy free lunch when things are bad enough — the liberal version of a claim that many supply-siders have advanced since the late 1970s. Short-term stimulus is consistent with long-term fiscal balance, perhaps more so than austerity.

To be sure, major differences remain between left and right. Conservatives want tax cuts to be permanent; liberals argue that they should be in force only as long as needed to jump-start growth. Conservatives believe that tax cuts work by increasing incentives to innovate and invest, while liberals believe that boosting demand is the key to shocking lagging economies out of their doldrums. And of course, liberals believe that the right kind of direct government spending (on infrastructure, for example) is the best strategy for boosting long-term growth when demand is slack, while conservatives think that tax cuts are the preferred — perhaps only — means to that end.

Still, this unexpected left-right convergence raises intriguing political possibilities. Republicans who refuse to consider boosting growth through additional public spending should be open to a tax cut for this purpose — especially a cut that promises not to increase (at least in the long run) government debt as a share of GDP. The potential deal is this: liberals would lead renewed stimulus with tax cuts, and in return, conservatives would agree to the kind of tax cuts (conditions-based and progressive) that liberals want.

Here’s a simple, easily administered proposal along these lines: a five-year reduction in Social Security payroll rates — by 3 percentage points during the first three years, phasing down to 2 points in the fourth year and 1 point in the fifth. (General revenues would fill the gap in the Social Security trust fund, protecting current and future beneficiaries.) A Joint Economic Committee report on the effects of the 2012 payroll tax cut suggests that the proposed five-year reduction could increase growth by 0.75 percent during each of the first three years while increasing net jobs by 600,000 annually during that period.

This course of action would entail some risks, of course. If Summers, DeLong, and Ball turn out to be too optimistic, an increased national debt would yield only modest increments in growth and job creation. But staying on our current course is not risk-free either: a sluggish recovery could blight the lives of older workers and dash the dreams of young adults. A progressive and time-limited tax cut may be a risk worth taking.


To state the obvious, we also need comprehensive tax reform. Whatever a growth-optimizing tax code might look like, most analysts agree that we are far from it today. Broadening the base of the code while lowering the rates is more than an exercise in tax hygiene; it is the best way to improve the odds that investment and consumption decisions will be made on the basis of their economic value rather than their tax consequences.

Another tried and true principle is that we tend to get less of what we tax more. All things equal, a tax on labor is perverse, especially in an extended period of slack labor markets. I join many others in my belief that a carbon tax makes a lot more sense than does the payroll tax. We should therefore initiate a revenue-neutral swap: as a carbon tax is gradually phased in, the payroll tax would gradually phase out. By lowering employers’ total labor costs without cutting compensation to workers, this approach could stimulate the creation of millions of new jobs. And by using price signals rather than command and control regulations, it would allow users of fossil fuels and other natural resources to find the most efficient ways of reducing consumption. As they demonstrate the efficacy of their choices, government could gradually reduce the regulations designed to further the same ends.

Public investment

Along with other analysts located across the political spectrum, I believe that well-designed public investment — in core areas such as basic research, infrastructure, and education and training — can boost baseline growth rates. (Standard theories of public goods explain why the public sector plays an essential role in these areas.) But public budgets at every level have been neglecting these core functions.

In the past, the U.S. has been able to balance short-term needs with programs that helped build a better future for everyone. The long list includes land-grant colleges, interstate highways, and the postwar explosion of publicly funded scientific research that has transformed the world. But in recent decades, we have lost our balance. At the federal level, funding for basic research has plateaued and is projected to decline over the next decade, while the longstanding consensus over infrastructure funding seems to have collapsed. (Numerous expert panels have put underinvestment in infrastructure at about 1 percent of GDP per annum over the past generation, generating an aggregate repair and replacement backlog estimated at $2.5 trillion.)

We should set a national goal of boosting public investment. That would mean benchmarking basic research as a share of GDP to the global leaders, boosting the total infrastructure of our national product by 1 percentage point, from 2 to 3 percent, and making room for higher education in state budgets. To say the least, that would require a rearrangement of national priorities, through some combination of a tax hike dedicated to public investment, cuts in non-investment spending, and innovative policies that would free up public funds for investment while encouraging the use of private capital for public purposes.

Two examples of such policies will suffice. If everybody were required either to purchase insurance against the possibility of nursing home care or to save for that eventuality, states could be relieved of the long-term care burdens for which they now pay in the Medicaid program. As for infrastructure, President Obama has been calling for increased investments since he was elected, without notable results. There’s a way of getting this done that enjoys broad bipartisan support. Representative John Delaney (the only former CEO of a publicly traded company now serving in the House) has introduced the Partnership to Build America Act, which would finance $750 billion in new infrastructure investment using no appropriated funds. A proposed American Infrastructure Fund would be funded by the sale of 50-year infrastructure bonds at a low fixed interest rate to corporations that would be allowed to repatriate a certain amount of their overseas earnings tax free for every dollar they invested in these bonds. Delaney’s bill has garnered broad bipartisan support, rare in these polarized times.


I need not tarry long on the topic of immigration reform; the self-destructive tendencies of our current policies are sadly apparent. If we wanted immigration to boost economic growth, we would impose no limits on applicants who score high on education and skills. Nor would we make it so difficult for employers of lower-skilled workers to get a reliable flow of employees to meet shift needs. Nor, of course, would we require foreign citizens who have received bachelors and advanced degrees in STEM subjects to repatriate after graduating from U.S. colleges and universities.


Recent studies have documented an alarming fall in business startups, long a key source of innovation and employment. Ian Hathaway and Robert Litan find a steady decline since the late 1970s. Since the Great Recession, new firms as a share of all firms have stagnated at the lowest level in four decades. Business deaths now exceed births for the first time since the beginning of these data series.

No one knows exactly why this is happening, only that it is a national rather than regional phenomenon, occurring at roughly the same rate in metro and non-metro areas. Conversations with entrepreneurs suggest that regulations have become steadily more burdensome and that startup capital has become increasingly difficult to acquire. In these circumstances, it would make sense to experiment. For example, what if capital gains from investments in new businesses were exempt from taxation?   

Narrowing the gap between Productivity and Compensation

Taking inflation into account, wages and benefits for workers have not budged since the end of the Great Recession. Nor are they likely to increase significantly anytime soon, according to recent survey conducted by the National Association for Business Economics. Some 82 percent of employers expect wages to rise between zero and 3 percent over the next three years. Only 11 percent expect faster growth.

All this is part of a larger story: compensation to workers has failed to keep pace with productivity gains for most of the past 40 years. Between 1973 and 2011, productivity rose by 80.4 percent while median hourly compensation rose by only 10.7 percent. But average hourly compensation rose by 39.2 percent, almost four times faster than the median. The reason is simple: the average has been dragged up by the top 1 percent, whose compensation rose twice as fast while everyone else fell behind. The gap between average and median hourly compensation has roughly doubled — from 40 percent to nearly 80 percent — since 1973.

Between the end of World War II and 1973, the rise of compensation in line with productivity fueled the creation of a middle-class society. Rising consumer purchasing power — which accounts for two-thirds of our GDP — kept growth robust. A growing, prosperous, self-confident middle class strengthened social mobility and anchored democratic trust. But over the past four decades, according to the Pew Research Center, the middle-class share of American households has declined from 61 percent to 51 cent. Not by accident, growth has slowed, mobility has stalled, political polarization has spiked, and confidence in government has collapsed.

Over the next generation, we face a stark choice: either rising compensation will yield steady income gains for average families, fueling faster growth, or the American dream will become a hollow promise. There is little chance that the continuation of current policies will yield the preferred outcome. So our policies must change.

Specifically, we should use public policy to reinforce the link between compensation and productivity. That connection must be accepted as a goal — and norm — across the economy. And to make it real, we should link the tax rates individual firms pay to the compensation strategies they adopt. The point is simple: Firms can either share productivity gains with their workers — or contribute to the public programs made necessary by their failure to do so.


Even the casual reader will have sensed that I have offered a potpourri of possible policies. That is no accident. Like most analysts who have tackled the problem of slow growth, I have concluded that the challenge is more than cyclical but goes to the basic structure of our economy. And like most analysts, I have some hunches about how to respond, but nothing approaching a clear new theory that would enable me to say “Do this but not that” with the requisite confidence.

I am convinced, however, that our current policies will not solve the problem. Unless we are satisfied with the status quo (or fear that efforts to change it will make things worse), we should be willing to experiment with measures that we would not have considered when the economy was growing much faster than today.

The opinions expressed here are solely those of the author and do not necessarily reflect the views of the Cato Institute. This essay was prepared as part of a special Cato online forum on reviving economic growth.

William A. Galston is a senior fellow in Governance Studies at the Brookings Institution.