SSN Memo

A Federal Job Guarantee: Anti-Poverty and Infrastructure Policy for a Better Future

Policy field

Connect with the author

Harvard University
The New School
Institute on Race, Power and Political Economy

Executive Summary

To combat the current health and economic crisis, and build national infrastructure in the public health, environmental and transportation domains, American government should directly hire millions of citizens in the coming two years, offering a federal job guarantee (FJG) that strengthens government at all levels, especially local and state.  Doing so would transcend the limits of current stimulus programs. Those programs confer money but not stable, dignified work.  This direct government employment (DGE) would rebuild sectors of our country that have withered – our public health clinics and agencies, our transportation network, our physical plant for education and services in both urban and rural settings.  DGE would also supply workers for the vital transition to a new, energy-efficient, reduced-carbon infrastructure. Unlike other relief programs, a federal job guarantee can eliminate involuntary unemployment, directly build the capacity of government to reduce the likelihood of future crises and  respond effectively to those crises that do arise. A FJG can complement other relief programs including social insurance and income support, but there is no substitute for its poverty-combatting, inequality-reducing, worker-empowering, nation-stabilizing, and infrastructure-building potential.

Scale matters. To meet the needs of the nation, a FJG would provide millions of new jobs, ranging from public health positions (at least 250,000 jobs), enhanced postal services, including postal banking (at least 100,000 jobs), construction, rehabilitation, retrofitting and staffing of schools, clinics, parks, senior centers and civic centers (at least 1 million jobs), new infrastructure, energy transition and conservation work, including solar installation (many millions of jobs), as well as investments in unemployment and social insurance and job training (hundreds of thousands of jobs).  And a FJG would rejuvenate America’s civil service at a moment of mass impending retirements, injecting greater diversity and youth into a system that sorely needs it.


The multidimensional crisis unfolding with the global coronavirus pandemic calls for a multidimensional solution. It demands bold thinking.  American public officials have done some of that bold thinking in greatly extending unemployment benefits, in offering loans and loan guarantees by the hundreds of billions of dollars, and by using rarely-invoked federal authority to incentivize and compel the production of medical necessities. The Federal Reserve has undertaken actions even beyond those it took in response to the financial crisis of 2008. 

Yet an important part of the policy toolkit has been neglected to date: direct government employment. Put simply, governments at all levels can and should employ millions more Americans in the years to come. Direct government hiring at all levels of the American political system will be an essential part of the response to the coronavirus pandemic. This paper lays out the case for large-scale direct government hiring—a federal job guarantee, and then explains the types of jobs and projects that could deliver value to the American people.

Part One: The Case  

There are at least three reasons why a federal job guarantee with direct federal government employment at the federal, state, and local levels has to be part of the portfolio of policy responses to the crisis.

1. The Need for Sustained and Meaningful Job Creation. This will not be a V-shaped recession. Public health restrictions are likely to remain in place for many months more and to resurface in fall 2020. Many small and medium-sized businesses that have closed or will close in the coming weeks and months will not re-open, and if they do, they will not likely hire their full pre-crisis workforce. This is in part due to the usual cycles of business downsizing and the costliness and delay of start-up after lay-up and scrapping, but part of it comes from the economic disruption of COVID-19. Until a coronavirus vaccine is widely available (not merely a matter of medical innovation but of mass administration of the treatment to hundreds of millions of Americans), American consumers will likely refrain from spending at anything like the rate they did before January 2020. Indeed, older Americans (who are on average wealthier than younger generations and with more disposable income) are the most likely to continue to curtail their spending behavior in the summer and fall of 2020 and into 2021, forgoing restaurants and bars, gyms, haircuts and other businesses where people gather or there is physical proximity. Add to this the well-known psychological effects of major recessions, and it is likely that the coming years will be characterized by greater savings behavior (which happened after the Great Depression, for instance). More robust savings would be a good thing in many respects, but in a heavily service-oriented economy dependent upon consumer spending, it means that millions of private sector jobs are not coming back.

2. The Need for Public Sector Capacity in America. Just as important as counter-recessionary policy, American government at all levels has weakened considerably over the past half-century. These frailties are visibly on display in the crisis. States that need to process unemployment benefits cannot do so. Public health departments lack the capacity to monitor the health of citizens. Medical stockpiles were depleted and depreciated. Federal, state, and local agencies that conduct scientific research, provide services to vulnerable populations, and construct and operate physical, technological, and transportation infrastructure, have the lowest workforce levels they have had in a half-century or more. In total, the federal government now employs about the same number of people that it did in 1967 — even as the total U.S. population grew from 200 million to around 330 million. Today the federal government employs considerably fewer people than in 1990. State government hiring has increased much more since the 1960s, from approximately 2 million full-time equivalent employees to 4.3 million, but most of this increase is due to growing community colleges, other colleges and universities, and prisons and corrections systems. Hiring in welfare, public hospitals, conservation, and even policing has stagnated in the last two decades, with all of these sectors declining on a per-capita basis.

3. Government Employment Has Distinct Advantages. Direct government hiring has advantages over other forms of counter-recessionary policy. In addition to direct government payments (stimulus checks, for example), government hiring provides jobs. One’s dignity is not limited to work, but jobs do entail work, and productive work with proper working conditions can confer dignity, meaningful exercise of mind and body, and benefits often in social settings. 

And combined with at least livable wages and salaries, a federal job guarantee would increase worker bargaining power in the private sector and reduce uncertainty over income streams. Unlike “gig economy” work, where take-home pay is often limited, and work provided by contractors, where pay may be very low for some and potentially egregiously high for others, government jobs are characterized by limited high-end pay and provide meaningful working-class to middle-class incomes. Direct government employment thus holds the potential for limiting income volatility and inequality, and the very real possibility that counter-recessionary policy will only exacerbate it. And finally, the spending that goes into direct government employment is spread diffusely among workers, limiting the gains from lobbying for contracts and grants, and thus blunting some of that political activity.

We’ve Started This Before, and We Can Finish It Today: The New Deal and Contemporary Proposals. The proposal here builds on and meshes with the ideas of others. The obvious precedent rests in many public hiring programs that exist in American and world history, most notably the New Deal. Especially in the first two terms of Franklin D. Roosevelt’s presidency, the American government launched a bold series of public hiring programs whose names are now etched in memory: the Works Progress Administration, the Civil Works Administration, the Civilian Conservation Corps, the Federal Emergency Relief Administration. These programs provided many grants to states (such as through the Public Roads Administration and Social Security Administration Public Assistance Program), but also sponsored direct government employment, such as the federal component of the Public Works Administration and the Civilian Conservation Corps. Whether in grants to states or directly, work relief programs mean a massive increase in public sector hiring.

Importantly, research by economic historians suggests that the New Deal’s public employment programs were robustly associated with a range of desirable public policy outcomes, including boosts in state income, consumption activity, and internal migration, and reductions in crime and mortality. The program of loans to banks and railroads under the Reconstruction Finance Corporation is not associated with these indicators of economic and social wellbeing. And as many historians and political scientists have shown, the effects of the New Deal were sharply limited by the racial segregation of its program structures. The New Deal thus provides positive and negative lessons for direct government employment programs today.

In more recent years, one of us (Darrick Hamilton, and his coauthors Mark Paul and William Darity) has along with other economists (such as Pavlina Tcherneva) called for a government jobs guarantee. This proposal advances that idea while also making the case for governing capacity to finance and deliver meaningful jobs. Given the possible magnitude of the emerging wave of unemployment, a government jobs guarantee might provide more jobs in a near-term environment than programs that focused only on augmenting government capacity alone. The second difference is one of emphasis: the aim of this program is both to provide jobs and to augment government capacity at all levels in the United States, especially in areas of dire need including public health infrastructure, forging a care economy that professionalizes child care, adult care and senior care, rural broadband and communications provision, energy and environmental transition programs, and transportation infrastructure. Poverty and inequality reduction alone are enough reason for policies like these, but the crisis lain bare by the pandemic goes beyond economic dimensions. Our body politic is rotting, and our republic is withering.

An Opportunity to Diversify and Rejuvenate the Government Workforce. There is nothing well-timed about any crisis, especially one that carries thousands to their graves and leaves others physically and emotionally scarred. Yet the imperative for direct government hiring comes at a time when government workforces nationwide are losing vast numbers of workers. As Baby Boomers retire, government is experiencing what is sometimes called a silver tsunami, with mass retirements from civil service at the local, state, and federal levels. These workers are on average older, and more likely to be both white and male than the workers who will replace them. A direct government employment program thus offers a unique opportunity to rejuvenate American government workforces and to diversify them at the same time.

Why Not Contracting? Critics of the plan might argue that the government can hire by offering competitive contracts to private employers. This has been a growing mode of public spending in the last three decades in the United States, yet it comes with major drawbacks. First, there is no antipoverty guarantee in government contracting, as the distribution of income depends upon the structure of contracting firms. Government contractors emerge in those sectors where the profits are larger and, hence, the labor share of income is probably lower. This also explains why lobbying for government contracts comprises an industry in its own right, diverting energy and resources from more socially valuable uses. Nor is there consistent evidence that government contracting provides superior efficiency. Although comparisons between contractor and direct hiring are complicated by the kinds of work done by different kinds of agencies and firms, defense agencies have reported that civilian employment is cheaper than contractor provision for a wide range of services. And finally, the profit motive alone has not sparked adequate commercial activity to address our nation’s infrastructure needs.

Part Two: The Jobs

This section describes some nuts and bolts of a direct government hiring agenda, the government units to do the hiring, and the kinds of jobs that will get done to build a stronger, more resilient nation.

Secular Trends in Federal Employment. Without trying to recreate government as it existed in the 1960s and 1970s, it is worthwhile to note that expanding federal full-time employee hiring to the per-capita level it was in 1967 would require expanding the workforce from 2.9 million workers to 4.6 million, an increase of 1.7 million jobs.  

Government employment at all levels currently stands at 22.7 million, or 11 percent of the working-age population of 18- to 64-year-olds in the United States. A one percentage-point increase in this share would amount to over 2 million hires.  A New Deal style increase (the kind witnessed from around 1929 to 1940) would amount to four to six million hires. And this does not take into account the immense surge in federal government hiring in World War II.

In the current economic crisis, other countries have not shed jobs as quickly as America has lost them, in part because their countries directly support businesses, but also because their governments directly employ a greater share of workers. If America boosted its public sector employment rate (currently about 15 percent of the labor force) to the level of Great Britain (about 17 percent), we would be creating four million jobs. Augmenting to Canada’s level (19 percent) would bring another four million. Augmenting to the level of France (nearly 24 percent) would provide well over ten million jobs. In other words, highly feasible expansions of the civilian service at all levels of government would provide millions of jobs.   

This proposal argues that we can and should do better than our European colleagues. With a federal job guarantee fully implemented, naturally public sector employment would be substantially higher. Assuming a modest unemployment rate (U1) of five percent, and underemployment rate (U6) of 10 percent, Paul, et. al. (2018) estimate that a federal job guarantee would require about 13 million jobs to absorb all the workers looking for work or inadequately employed. At the peak unemployment resulting from the 2007 Great Recession, the estimate would be about 24 million jobs. 

What would a federal job guarantee program look like? The program would be administered by the Department of Labor and overseen by the Secretary of Labor. The Secretary and Department of Labor will be charged with working in conjunction with other agencies, and states and localities to identify an inventory of public infrastructure tasks and associated jobs.

1. Jobs to support state and local governments. State and local governments account for most of the government employment in the United States, and they comprise an important part of our federalist system. In a range of policy domains – public health, education and job training, infrastructure, health provision and conservation – it is state and local governments that have the legal authorities and historical capacities to act. Yet state governments also have the weakest capacity to respond to an economic crisis. First, state governments have nothing like the Federal Reserve to back their debt (this is one among several reasons that the Federal Reserve has recently been looking into mass purchases of municipal bonds). They cannot print money. Second, state governments lack the expansive revenue-generating capacity of the federal government. Beyond this, state governments have in recent decades constrained their powers with balanced budget amendments, while a number of states (Illinois and Connecticut are two examples) have badly underfunded pension systems, and still others have tethered their income to commodity (especially petroleum) sales, which are at historically low prices and whose price declines have damaged state revenues.

Workers and some capital costs for projects at the local, county, or state levels can be funded in a way to provide regional full employment and infrastructure needs. The projects and areas with the greatest needs could be cued to receive the federal intervention sooner. The Department of Labor would work closely with local governments. Local and state governments would, in conjunction with community organizations, develop project proposals to address local infrastructure needs, similar to the local, state, and federal collaborations in the design of direct government employment programs of the New Deal. Localities are likely most aware of the infrastructure needs to offer the greatest benefit to their communities.

In order to minimize inequality across states and reduce the burden of discrimination, the workers under this program would be federally employed but would be tasked or apportioned (perhaps by formula, to reduce near-term political discretion) to states according to their economic, public health, and infrastructure needs. Even if the work occurs at the state and local level, the federal government needs to step in to provide the jobs, and there are a range of tools it has to do so. The traditional tool consists of program-specific block grants that have been an important feature of American government since the New Deal and especially the Nixon era. The problem with these grants is that they often permit state and local officials to withhold important resources from populations in need, as occurred with so many states’ refusal to take Medicaid expansions under the Affordable Care Act. Our proposal would build upon the strengths of grants by making the jobs available to states and local governments, but would address the inequality-related weakness of these programs by making the employees federal workers who, as such, are made available to state and local governments.  

In order to avoid potentially unconstitutional imposition of requirements upon state governments (which were declared unconstitutional in NFIB v. Sebelius (2012)), the federal government could offer these jobs as an option, as it did with Medicaid expansion in the Affordable Care Act. Another alternative is to allow state and localities to develop a bank of proposals for physical and human infrastructure from which a federal authority (operating under formulaic or discretionary procedures) can prioritize with regards to urgent and useful needs. The state and localities can even administer the program with federal oversight, however, the workers would be ultimately employed by the federal government, as opposed to state or local employees by way of a federal grant.

2. Jobs to support federal agencies. Direct government hiring by federal agencies would be essential for many reasons. First, in a polarized political environment, it is quite possible that some states may obstruct cooperation with federal government employment. Having a robust program of federal hiring, including directly in uncooperative states, would broaden the distribution of benefits and mechanize the mandate for guaranteed employment. The federal government has a well-established personnel system that has expanded before: witness the mobilization of government employment after the September 11th, 2001 terrorist attacks, including the creation of a new Cabinet-level department and the launching of new agencies such as the Transportation Security Agency (TSA).

Congress should both create new federal programs and fund federal agencies to expand existing programs. The kinds of programs can and should be shaped by the imperative social, political, and economic needs of the American republic.

What kinds of programs and what kinds of jobs make the most sense?

Public health infrastructure. The news that we get on a daily basis on COVID-19 comes largely from the numbers collected by state health departments, agencies that are usually invisible but ever a bulwark against the most daunting health threats. Yet state and local health agencies have shed capacity since the 1980s. Then came the Great Recession. From an already weakened state, the American public health system has lost 55,000 jobs since 2008. Reliable estimates suggest that the deficit in American public health infrastructure amounts to at least 250,000 additional jobs. An assertive program of public hiring would include nurses, physicians, epidemiologists, health care aides, mental health professionals, social workers, health clinic administrators, workers to perform contact tracing for this and future pandemics, and building services workers like janitors.

Rebuild our crumbling cities. Urban infrastructure would be an important part of any American rebuilding effort, but key to rebuilding America’s cities would be to invest in the public hospitals, health clinics, schools, and libraries (including informatics and digital training centers) that would promote both civic engagement and human capital formation. Construction jobs alone to build the new buildings and renew the older ones required for first rate public health, education, and library systems in cities and towns could provide hundreds of thousands of jobs, with many more in buildings operations and maintenance. Funding these positions would also allow governments to target black and brown communities and lower resourced communities in general who have traditionally been passed by in government employment programs.

Expansion of rural programs through augmented postal services. The post office is the branch of government that Americans probably know best, with offices in essentially every zip code in the country. Until 1970, the Post Office Department provided banking services in its branches, providing an important tool of savings that dated from the Progressive Era. Over one hundred countries around the world offer a postal banking option. In addition to banking services, post offices could be augmented to provide rural broadband (not merely network extension but installation and repair consultations), health and wellness clinics, and claims processing for federal programs. Building upon our postal system would leverage existing infrastructure and provide rural benefits.

There are approximately 35,000 post offices in the United States, with just under 500,000 employees. Each office that adds banking services would add an employee or more. Transforming the post office into a local service provision agency would offer the opportunity for the creation of tens of thousands, if not hundreds of thousands, of jobs. 

Preschool, Pre-kindergarten and childcare for all. Americans have been confronting the reality that many jobs in our rapidly evolving society require an education beyond that provided by the K-12 system. Proposals for a vocationally-focused 13th and 14th grade have emerged, whether through reformed public school systems or through community college and vocational school expansion.  In the other direction, pre-kindergarten and pre-school programs have important precedents in Head Start and have been shown to bring a wide range of social, economic, and cultural benefits.  As the economist Randy Albelda has argued, programs like these, as well as expanded child-care and after-school options, can augment parents’ ability to participate in the labor force while also delivering educational benefits to their children as students. Important educational and non-economic by-products would include better student-to-teacher ratios, augmented tutoring, and increased student services.

Infrastructure and the Green New Deal. Important proposals for a Green New Deal – a series of programs that invest heavily in the renovation and transformation of our energy infrastructure and prepare the country for the energy revolution in wind and solar power – have been written about elsewhere. Yet direct government employment on the model of the Civil Works Administration, Works Progress Administration, and the Civilian Conservation Corps would efficiently and robustly serve the ends of environmental economic transformation in the United States. Jobs could include solar panel installation for existing buildings as well as new panel grid installation, windmill installation, white roof installation, and concomitant investments in energy storage. A system that complemented the repair of the federal highway system could install new networks of electric charging stations across the country, including in urban and rural areas.  

As the example of Vermont shows, “clean energy” and “green energy” investment has the potential to create thousands upon thousands of living-wage jobs that span a range of skills. Solar energy has begun to provide more jobs than fossil fuel companies in the United States. With the impending downturn in oil and gas hiring, and with the continuing downturn in coal-related employment, important investments in these areas would provide jobs in the energy industry just as they are drying up elsewhere.

Construction and conservation work, including the Great Lakes, rivers and oceans. Massive job losses are occurring in states and communities that depend heavily upon water, whether the Atlantic or Pacific Coasts, the Gulf of Mexico, the Great Lakes, or America’s many river systems that are crucial for transport. Jobs available include rivers and harbors construction and conservation, toxic waste mitigation, invasive species mitigation, and coastline construction to prepare for rising ocean levels.

Job training and education. Every kind of job imagined here will require labor supply adjustments. The federal government used to have expansive job training programs such as the Comprehensive Employment and Training Act created in 1973 and the Job Training Partnership Act that followed it. Studies of these programs were mixed on their efficacy, but Congress could create new programs or directly fund state universities and community colleges to do so. The key here is that completion of a job training program should come with the assurance of a productive job with at least a living wage at the end.

State elections administration and voter registration. Election administration in the United States rests at the level of state government, and this capacity, too, has withered. Direct government employment programs would be used to strengthen the administration of motor-voter laws, provide non-volunteer poll workers, supply workers to increase the registration of minority populations, make voting more accessible for the elderly as well as the young, and provide nonpartisan information to empower urban and rural citizens alike to vote. 

Investing in Indian Country. An expansion of hiring by the Bureau of Indian Affairs could directly employ tribal members to work on Indian reservations throughout the United States, in positions ranging from health and education to conservation and infrastructure. Such an expansion would help the United States government meet the needs of a non-white population that is now fast-growing but which remains highly impoverished and often left aside from both economic innovation patterns and social welfare programs that target urban concentrations.

Legislative and regulatory capacity. Although it won’t be the source of millions of jobs, America’s Article One branch has seen its capacity wither in recent years, losing important analytic capacity with the downsizing of the Government Accountability Office, the Congressional Research Service, the Congressional Budget Office and the elimination of the Office of Technology Assessment. Committee staffs have also declined. As a result, a bipartisan, cross-spectrum network of observers has called for a reinvestment in congressional capacity. Such a reinvestment will be all the more important if any kind of direct government employment program comes to pass, as the dispensation of funds and working conditions must be overseen by the people’s representatives.

These congressional jobs need not be centered in Washington, D.C. Congressional staffers who oversee the work of the Veterans Administration hospital system, or who monitor the progress and employment conditions of infrastructure projects, or who facilitate and monitor state and local government use of federal workers, could be employed across the geography of the United States.    

Important federal agencies in the regulatory, scientific and health space have also shrunk. The Centers for Disease Control has been shedding jobs since the Bush Administration. The Food and Drug Administration has expanded, but much more heavily in staffing to approve drugs and much less to monitor imports and pharmaceutical and medical device safety. And as two public health leaders have recently proposed, the federal government could fund and structure a public service program to boost the supply of vital physicians by 20,000 medical students in the near-term of the pandemic.

The capacity of state legislatures, state agencies, and local governments is also worryingly weak. Indeed, critics such as Steven Teles and Brink Lindsey have pointed to the lack of personnel in decision-making agencies – and their increasing reliance on lobbyists – for decisions that frequently benefit special interests at the expense of the common good. Federal block grants to hire staff and increase the research capacity of state legislatures and agencies to analyze complex issues could have broad benefits for citizens through reducing regulatory capture and better decisions.


America seems stuck on “stimulus” as metaphor and model. Indeed, a federal job guarantee not only stimulates, it eliminates involuntary unemployment, the concept of working poverty, provides an automatic business-cycle stabilizer, and ensures a more resilient and secure public infrastructure.  

As millions file for unemployment amid the economic disruption cause by COVID-19, the primary response in Washington has been to rely ever more on the Federal Reserve as an instrument of social policy, to provide unemployment insurance for those out of work, and to provide further inducements to small businesses. While valuable, these policy moves do everything but directly provide jobs to those who need them. The limits of the American policy discourse will become apparent in the weeks and months to come, when corporate profits and asset values begin to rebound but millions of our fellow citizens remain without work and the meaning, structure, and benefits it provides. Americans can claim the mantle of the New Deal past while consciously avoiding the racism that limited its providence. And Americans can claim the mantle of the future by transforming our country into one ready for the energy revolution and the next pandemic, while investing in urban and rural communities that have been too often left behind during the past three decades, as economic gains have flowed overwhelmingly to a small percentage of Americans and corporate titans at the very top.

Unprecedented crises demand unprecedented action. The pandemic calls for restructuring, not just re-stimulation. Far from a temporary state of affairs, the COVID-19 pandemic crisis instead exposes the rot in our republic, the severe weakness of our society, and the frailty of a purportedly robust economy where, even with record-low official unemployment rates, four in ten Americans reported before the pandemic that they would be unable to meet a $400 emergency with their savings. 

We are confronting problems that trillions in central bank loans, hundreds of billions in small business support, and vast new extensions of unemployment insurance, while valuable, are simply no match for. Building a competent, effective government through an ambitious program of federal job guarantee will not solve all of our problems, but if it is not a sufficient answer to the crisis of our times, it is an absolutely necessary one.