Front Line of Defense

The limitations imposed by the role of the states in the U.S. unemployment insurance system are the reason why a majority of workers are not protected and why even insured workers receive inadequate protection. Steven Attewell writes: “Our reconstruction of the unemployment insurance system should start from three basic principles. First, unemployment is a national problem for our single, national economy, and requires a nation-wide system to respond to it. Second, in order to protect the entire workforce from the sudden shock of wage loss and the economy from the sudden shock of consumer spending collapse, all workers need to be inside the system, contributing and protected. Third, unemployment benefits should be set at a sufficient level to keep individuals and families from falling into poverty and should be automatically extended in periods of economic decline in job losses, when normal expectations that people can find new jobs no longer apply.”


One of the first casualties of the 2007 recession was in the realm of ideas. Prior to the recent economic collapse, Fed Chairman Ben Bernanke had confidently proclaimed an end to economic catastrophes. The advent of modern central banking and the liberalization of financial markets, he argued, had ensured that the kinds of systemic failures last seen in the Great Crash of 1929 would never happen again. Future recessions would be mild and short-lived, as central banks would spring into action to head off panicking markets, and the “automatic stabilizers” of the welfare states of the developed nations would kick in to prevent a panic from turning into a long-term slump in demand.

The last three years have taught us many lessons to the contrary. One of the things we have learned from the Great Recession is that America’s hybrid state/federal unemployment insurance (UI) system is essentially broken. The unusually sharp job losses – some eight million jobs lost from the outset of the recession in 2007 to the nadir of job losses in 2009 – caused a sharp decline in consumer spending of all kinds. For example, sales of motor vehicles and parts fell by an average of 11.1% during 2008 and 2009, while furnishings fell by an average of 4.4%. One of the purposes of our UI system is to prevent sudden downturns in employment in particular sectors, like construction or financial services, from turning into sudden declines of consumer demand that lead other sectors to start mass layoffs of their own.

And yet our UI system largely failed to counteract this slump. Most workers are not eligible, and many of those who are eligible do not get enough to keep themselves and their families out of poverty. States routinely exclude millions of workers from coverage and deliberately underfund UI systems that they are structurally incapable of operating in the midst of a recession.

If the U.S is to truly recover from our current Great Recession, and reform UI to prepare for future downturns, we need to take action now.

What is Wrong with UI?

When the architects of the Social Security Act designed our unemployment insurance system in the winter of 1934, they were hampered by structural constraints that do not exist today. They sought to cover a very different workforce, in which 10 million workers still worked in agriculture and 14 million or more workers flocked into ever-growing factories. In addition, the members of the Committee on Economic Security (CES) were vividly aware that they had to design a system that could pass review by a Supreme Court that had already struck down minimum wage and maximum hours laws as violations of “liberty of contract,” and was currently considering doing the same to the Agricultural Adjustment Act and the National Recovery Act. In the face of this expected hostility, CES staff members could not design a simple unemployment insurance (UI) system, financed by a single federal tax and managed solely by a national Social Security Board. Instead, they designed a complicated workaround in which a federal “regulatory” tax would be forgiven if states enacted their own unemployment insurance systems. This elaborate system was deemed acceptably federalist by the Supreme Court. The system that we know today – where 50 states operate their own UI systems, each setting a different payroll tax rate, eligibility standards, and benefit levels – was not a decision made on principle by all-knowing Founding Fathers. Rather it was a nakedly tactical choice made by Roosevelt administration officials who would have preferred to design their system along the same national lines as Old Age Insurance (Social Security). But whereas unemployment insurance had to design a complicated regulatory tax to get around existing state institutions (the Supreme Court's federalist tendencies made a separate federal system that might override or conflict with existing state systems constitutionally suspect - hence the adoption of a "forgiven" regulatory tax), in the field of Old Age Insurance, no state systems existed. Thus OAI had a clear legal footing for a national program - politically, it was made clear a national program was Roosevelt's line in the sand; Southern Democrats in Congress (once they were able to exclude agricultural and domestic workers) were willing to accede to the President's demands.

From the beginning this design created a dangerous vulnerability within the UI system. Because 49 states have a constitutional requirement for balanced budgets and more than 26 states have constitutional limits on borrowing, states lack the federal government’s ability to resort to deficit spending in recessions. When recessions hit, states find themselves in an impossible bind. As consumption declines, sales tax revenue plummets; as businesses shed jobs and go out of business, corporate income tax revenue declines; as workers find themselves without steady paychecks, individual income tax receipts fall; and when millions of homes and business offices go into foreclosure and the real estate market collapses, so too does revenue from property taxes. While state governments stagger under the burden of sudden deficits, millions of previously-employed workers desperate for protection against total destitution increase the demands on unemployment insurance, thereby emptying state treasuries at a time when they are least able to respond.

What makes this worse is that the state-run nature of the UI system creates powerful incentives for states to underfund their UI Reserve Funds. Because states must fund their UI system with payroll taxes on local employers, reducing payroll taxes is often the first bargaining chip that states offer when trying to compete with one another to attract new firms from out of state. The lower the payroll tax rate, the cheaper it is to hire workers. Corporations respond to the leverage they have over furiously-bidding states to extract concessions so that they do not have to pay in at all, with the state covering their contributions. The result is a dangerous depletion of UI reserves. In the current recession, no less than 32 states have had to borrow billions of dollars from the U.S. Department of Labor because they let their UI funds run dry.

Many states have responded to their vulnerable position by restricting the number of workers who are eligible for UI. This reduces the drain on the system at the expense of workers who will have no protection when they lose their jobs. By keeping contributing requirements high, and by requiring longer “Base Periods” (the length of time one has previously worked in order to be eligible for UI), states are able to keep lower-paid workers and temporary workers from gaining eligibility for state funds. Similarly, by restricting eligibility to full-time direct employees, states exclude part-time workers and so-called “independent contractors” from coverage. To reduce labor costs, employers have responded to these changes by accelerating their project of replacing full-time workers by chopping full-time jobs into part-time jobs and by re-defining their workers as “independent contractors” even when they are working 100% of the time for one company. Looking to further reduce their costs in the current recession, employers have increasingly turned to professional UI claim-challenging firms. (See Chart 1)

In addition to restricting eligibility, one of the ways that states have attempted to limit their exposure to sudden increases in UI costs is to keep their benefits as low as possible, replicating the insidious patterns of pre-New Deal poor relief. At the moment, Mississippi and Alabama pay average benefits that are below the federal poverty line for a single individual, and every state except Hawaii pays an average benefit less than the poverty line for a family of four with two children. Such low levels of unemployment benefits are clearly unable to support previous levels of consumer spending, meaning that these states’ UI systems are failing as “automatic stabilizers” both for people who get benefits and people who do not. Liberal states like Massachusetts lack the fiscal capacity to do much more, paying an average benefit that falls just below the poverty line for a family of four.

The result of these dodges is a safety net that is mostly made up of holes. This means that a UI system that is supposed to protect everyone does not, which causes dramatic increases in recession-induced poverty, and a system that is meant to keep consumption from falling too far in recessions is fighting with one hand tied behind its back, which leads to dramatic slumps in consumer spending. When consumer spending falls faster than it is supposed to, businesses suffer and respond with frantic layoffs, which creates a downward spiral as newly unemployed workers cut back spending faster than employers can reduce labor costs.  

This is unacceptable from a policy standpoint. Even in the best of circumstances, our current UI system is incapable of keeping consumer spending level – let alone increasing consumption to stimulate the economy. If we want a system that can really protect American workers from job losses and our economy from sudden collapse of consumer spending, we are going to have to do a total refit of our unemployment insurance infrastructure.

Bringing UI Up to Scratch

Luckily, much of the machinery required for creating a functioning unemployment insurance system already exists: Title III of the Social Security Act, which structures the current UI system, can be used to enact changes through the regular amendment process. We already have a federal unemployment insurance payroll tax system in the form of FUTA (which is a 6.2% “regulatory" payroll tax that is offset by state UI taxes down to .8%). We have federal agencies that are familiar with the program and how it works. In this, we have an advantage over the creators of the Social Security system, who had to create this machinery ab nihilo.

Our reconstruction of the unemployment insurance system should start from three basic principles. First, unemployment is a national problem for our single, national economy, and requires a nation-wide system to respond to it. Second, in order to protect the entire workforce from the sudden shock of wage loss and the economy from the sudden shock of consumer spending collapse, all workers need to be inside the system, contributing and protected. Third, unemployment benefits should be set at a sufficient level to keep individuals and families from falling into poverty and should be automatically extended in periods of economic decline and job losses, when normal expectations that people can find new jobs no longer apply.

One National System

In public policy, everything begins with financing. In order to save the UI system from the inherent weaknesses of state-level government, the federal contribution to unemployment insurance should be gradually increased. As it stands, the federal government contributes the entire cost of running state UI agencies and employment bureaus, as well as 50% of the cost of extended UI benefits (the recent stimulus bill has temporarily increased the federal contribution to 100% of extended benefits). This federal contribution could be increased in a series of steps from 25% of the regular UI to 50%, 75%, and eventually 100%. All that is needed to do this is to reduce the share of FUTA that is offset by state UI taxes, while decreasing state UI taxes in step until we have a national FUTA tax of 6.2%. At the same time, we can make the FUTA more progressive by eliminating the current wage cap (FUTA is only paid on the first $7,000 in wages, which disproportionally burdens low-paid workers while giving the rich a much lower tax burden), and by varying the rate of taxation according to income level.[1]

By replacing 50 different state taxes with one national tax, UI would be greatly stabilized. Federalization would remove the incentive for states to compete for jobs by offering a lower UI tax rate and, by sourcing the financing of UI inside the federal government, would also give the system access to the federal government’s ability to engage in deficit spending in a time of crisis. Likewise, establishing a single national UI Reserve with clear reserve requirements to prevent the fund from running dry in an emergency would be an enormous improvement in the program’s stability. Ideally in good economic years the UI system would accumulate positive balances that could be spent in recession years, allowing for counter-cyclical spending without large deficits. 

It is important to note that making the UI system nationally-funded does not necessarily mean that the program has to be fully nationalized. A number of options are open: states could continue to administer a federally-financed and regulated UI program, UI programs could be run through joint federal/state agencies, or the program could become purely federally-administered. 

Universal Coverage of Workers

The extension of UI coverage to all workers would be the final culmination of a long historical process of inclusion. When the system was first established in 1935, the majority of workers were not eligible for unemployment insurance – notably, agricultural and domestic workers and seamen were not included. While these exclusions disproportionately affected women and African-American workers, it was also the case that a majority of white workers were not eligible for UI at the outset. Eligibility was gradually expanded through a series of Social Security Act amendments over the years. Most recently, the 2009 stimulus act included a provision known as the “Unemployment Insurance Modernization Act.” It offered $7 billion in incentive money for states that adopted “Alternate Base Periods” that allow workers to claim eligibility for UI by counting their most recent earnings as opposed to the previous year’s earnings, and agreed to cover part-time workers who are looking for full-time work and workers who leave work due to family emergencies. The ultimately-futile opposition of some conservative governors (not coincidentally from states with low UI taxes, underfunded UI systems, and low benefit levels) had as much to do with states trying to preserve a competitive advantage of low taxes and low labor costs as it did with presidential hopefuls looking to appeal to the conservative base through public defiance of President Obama.

This incremental process should be accelerated, with the ultimate goal being the inclusion of all hourly workers in the UI system, including full and part-time, temporary and permanent, and self-employed and “independent contractor” workers. At the very least, part-time and temporary workers should be able to claim benefits proportionate to what they would have been entitled to had they been working full time at their hourly rate.

In addition to the participation eligibility standards, the current requirement that workers must accept any offer of work to remain eligible for benefits needs to be altered to prevent two abuses. First, claimants should be entitled to refuse work that is substantially below their current income. This is needed to prevent abuses by which employers engage in mass layoffs, offer to hire back their workforce at reduced wages shortly afterwards, and then use the ensuring refusals as evidence that their former workers have voluntarily refused employment and thus the employer is no longer obliged to cover their UI costs.

Second, claimants should be entitled to refuse work if they are in a bona fide labor dispute. This is needed to prevent abuses by which employers deliberately incur strikes and then threaten strikers with the loss of their UI benefits if they refuse to work as scabs and break the picket line. Our unemployment insurance was meant to be a system that lifted up American workers, and gave them a sense of security so that they would not be completely vulnerable to unilateral decisions of their employers. To allow these abuses to continue would make a mockery of the spirit of the Social Security Act. 

A National Minimum Benefit

In order to prevent recession-induced poverty and to ameliorate recession-induced declines in consumer spending as much as possible, a minimum benefit should be established at the minimum wage rate of $290 a week ($13,920 a year). In 2009 for example, a UI system that provided all of the 15 million officially unemployed with a minimum benefit of $290 a week would have pumped in an additional $226 billion into consumers’ pockets. The existing UI system, which includes state benefits and federal support programs, only covers 9.8 million workers. Providing benefits for the 15 million people who are officially unemployed (to say nothing of the 15 million additional unofficially unemployed) would increase consumer spending and spur job growth, making recessions shorter and less steep, thereby reducing the number of workers who would spend extended periods on unemployment benefit and the ultimate drain on our UI system. In addition, eligibility beyond the initial 26 weeks of benefits should be automatically extended during recessions or periods of widespread unemployment.

A further improvement that could be studied is the creation of an additional, contributory benefit for families on UI with children, in order to make up the $170 difference between a minimum benefit that keeps a single individual out of poverty and a minimum benefit that keeps a family out of poverty. Such a program would provide a badly needed protection against families falling into poverty and requiring TANF assistance.


If these three principles are enacted, the U.S. will have a UI system that is up to the task of protecting workers from destitution and the economy from sudden shocks to consumer demand. However, we should be very realistic about what an “automatic stabilizer” can and cannot do – a revamped UI system will make recessions shorter and shallower, but it cannot prevent them; it will make layoffs less painful and should prevent some second-order job losses caused by falling consumer demand, but it cannot keep people from losing their jobs. 

And in an economy that is increasingly structurally disposed to shed jobs in recessions and rely on productivity increases, casualization, mechanization, and off-shoring rather than rehiring during recoveries, even a high-quality UI system will only be a tourniquet on an open wound. 

What would be needed for that larger problem is a system that can provide genuine economic security by creating incentives for employers to avoid or reduce mass layoffs. "Seventy-three years ago, FDR spoke of the hopeful belief that “the Nation, seeing and understanding the injustice” was “determined to make every American citizen the subject of his country’s interest and concern.” The same challenge that FDR posed to the America of his day – “whether we provide enough for those who have too little” – remains unanswered today, and it remains to be seen whether we share the determination to grapple with injustice. Thirty million Americans lack for work – and only ten million of them have any kind of protection. Unless we act to protect the other twenty million, and soon, we will have failed the “test of progress."

Steven Attewell is completing his PhD in history at the University of California at Santa Barbara. He co-authors a blog called “The Realignment Project,” which discusses current policy and politics through the lens of history.

[1] The ultimate outcome does not necessarily have to be a 100% federal contribution rate, although that would be the most successful at avoiding the structural limitations discussed above. UI could be reorganized similarly to Medicaid, with a 50/50 split, or at any other ratio.




Steven Attewell