Volume 52, May/June 2017, Number 3 · pp. 142-148
What the EU Can Learn from the American Experience with
Suzanne Simonetta, US Department of Labor, Washington DC, USA.
This article is part of the Forum A Fiscal Stabilisation Function for the Eurozone.
The transatlantic ties between the United States and Europe are quite strong. While there are many political, cultural, economic and social differences, this historical connection remains deep. The foundation of the American system of government draws heavily from European experience and political theory. The shared values among these diverse countries have been the basis of lasting partnerships. This relationship has remained solid as Europe transformed itself through creation of the European Union. The growth and evolution of the EU has led to new challenges, which require serious thought about how they may be overcome. Although the parallels go only so far – the US is a genuine federal system of government and the EU is not – there are sufficient similarities between them that there is great value in examining the American approach when the EU considers its options.
One area that could benefit from a review of the US approach is unemployment insurance (UI). The impact of the financial crisis and Great Recession resulted in questions being revisited about whether there is a need for a European unemployment insurance system.1 A major study was commissioned to examine the feasibility and added value of a European unemployment benefit scheme.2 This paper aims to contribute to the ongoing dialogue and debate by providing an overview of the UI program in the US and identifying valuable lessons from the American experience.
Unemployment insurance in the US
To understand unemployment insurance in the US,3 it is important to start with the historical context during which it was created – the Great Depression. While efforts to establish the program had begun years earlier in many states, none succeeded. Interstate competition – the concern that states without a UI program would have a competitive advantage over states with a UI program – was a barrier to achievement of this social insurance program.4 The experience of the Great Depression, with unemployment rates reaching 25% in 1933, had a devastating impact at both the individual and societal level, and made it abundantly clear that the US needed to establish a UI program.5 It was during this crisis that the political will was finally found to enact enabling legislation for UI in the Social Security Act (SSA) of 1935.
The UI program provides partial, temporary income support to individuals who are unemployed through no fault of their own. It is a federal-state partnership based on federal law but administered by states under state law.6 This structure was not necessarily considered optimal when design options were being contemplated. However, in the 1930s the role of the federal government was much more limited than it is now, and there were concerns that a national UI program would be determined to be an unconstitutional expansion of federal authority. For this reason, it is technically voluntary under federal law for states to have UI programs. However, there are incredibly strong incentives for the states to operate UI programs consistent with federal law – grants from the federal government for states to administer their UI programs and large credits for employers through the Federal Unemployment Tax Act (FUTA) . Because of this design feature, the Supreme Court of the United States decided that the UI program was constitutional in Steward Machine Company v. Davis (301 U.S. 548 (1937)).
Unless there is an explicit requirement or prohibition in federal law as a condition of grant or tax credit receipt, states have great latitude to establish the parameters of their UI programs. For this reason, there is much variation among the states with respect to qualification and eligibility requirements, weekly benefit amounts, number of weeks of benefits, disqualification provisions, taxable wages, tax rates and many other key policy areas.7 Thus, the description below focuses on the general concepts underlying the various aspects of the program and the areas where there is national uniformity.
Types of programs
Given the nature of this federal-state framework, UI benefit payments at the beginning of an individual's spell of unemployment are typically made and funded by states under state law; only state administrative costs are paid by the federal government. Starting in 1970, the permanent Extended Benefits (EB) program was established to provide additional weeks of benefits during recessions. Although paid under state law, the federal government pays half of EB costs.8 To expand access to benefits for certain groups of people, several permanent federal UI programs were created that states administer on behalf of the federal government under agreements. (States follow federal operating instructions and get reimbursed for benefit and administrative costs.) These include programs for federal civilian employees, members of the military, individuals who became unemployed as a direct result of a disaster and who are not eligible for "regular" UI, additional weeks of benefits for certain individuals who become unemployed due to the adverse effect of trade, and special, temporary programs providing additional weeks of benefits during recessions to long-term unemployed workers.9 In general, eligibility requirements and benefit amounts for the federal programs are the same as they are for the state's own UI program.
In the US, unlike many other countries, individuals must have had sufficient recent labor market attachment, measured by work experience, in order to qualify for UI benefits. Thus, eligibility for benefits is based on the state where individuals worked, not where they live.10 New entrants to the labor market, re-entrants after a withdrawal from the labor market, the self-employed and genuine independent contractors are not eligible for UI because they did not work in employment positions covered by the scheme. In general, applicants must have earned sufficient amounts working in covered employment positions during at least two calendar quarters in a 12-month period prior to becoming unemployed in order to qualify for benefits.11 Traditionally, states would examine earnings during the first four of the most recent five completed calendar quarters when making what is called a monetary determination. Recognizing that using this period of time as a basis for establishing UI eligibility (i.e. "base period") does not take into account up to the most recent six months of an individual's work history, many states have begun to use an alternative base period that,12 instead, examines earnings during the most recent four completed calendar quarters when making a monetary determination.13
Next, a determination must be made that the applicant was separated from employment (i.e. became unemployed) through no fault of their own. The classic example of this is when an employer lays someone off because work is no longer available. However, under certain circumstances, an individual may still be eligible for benefits even if they quit or are fired. Each state's UI law defines what constitutes good cause for quitting.14 While all states include good cause connected with work, many states also include certain personal reasons in their definition. Similarly, while there are many reasons why an employer may decide to fire an employee, in general, individuals would only be disqualified from receiving UI benefits if they were fired for work-related misconduct.
After the initial eligibility determination is made, applicants must demonstrate their continued attachment to the labor market by meeting a set of ongoing eligibility requirements for each week that they claim benefits. These include being able to work, being available for work and actively seeking work.
The UI program is funded primarily through federal and state taxes assessed on employers.15 All unemployment revenue is kept in the Unemployment Trust Fund (UTF), maintained by the US Treasury. In general, the FUTA effective tax rate is 0.6% on the first $7,000 of workers' earnings.16 The full FUTA tax is 6.0%, but employers get a credit of up to 5.4% if the state's UI law conforms to federal UI law and the state has no long-term outstanding federal advances (loans) to pay benefits. FUTA revenue is primarily used to pay for state costs administering the program, benefit costs for certain programs that extend (provide additional weeks of) benefits and for advances to states that run out of funds to pay UI benefits.
State unemployment tax revenue is used to pay for "regular" benefits – typically up to 26 weeks of benefits are payable to individuals when they become unemployed. Some states provide a uniform number of weeks of benefits to all jobless workers who qualify. Other states provide a variable number of weeks of benefits, whereby individuals with more earnings throughout the base period will be eligible for more weeks of benefits than individuals with lower earnings during only part of the base period. The unemployment tax rates and the amount of wages that are subject to state unemployment taxation vary significantly among the states. In addition, in all states, the state unemployment tax rate assigned in a given year varies among employers based on their experience with unemployment (i.e. experience rating). Employer accounts are "charged" for benefit payments made to their former employees, and these charges are factored in when determining employer tax rates in subsequent years.
The range of applicable state unemployment tax rates varies from year to year depending on the reserves the state has in its UTF account to pay future benefits. When the economy is strong, trust fund balances increase because there are more employers paying taxes on more employees' wages while fewer benefit payments are made. If the state's trust fund account balance exceeds certain levels, the range of applicable rates decreases in the following year because less revenue needs to be raised. When the economy declines, trust fund balances decrease; benefit payments go up as layoffs increase and tax revenue decreases as fewer employers pay unemployment taxes on fewer employees' wages. If the state's trust fund account balance drops below specified levels, the range of applicable rates increases in the next year so that more funds will be collected to pay for benefits. Thus, not only will state unemployment tax rates vary from employer to employer based on their experience with unemployment, they will also vary from year to year based on the state's reserves in the UTF.
UI is, as its name implies, a social insurance program paid as a matter of right to all individuals who meet its requirements. If a state runs out of funds to pay benefits, states may borrow from the federal government in order to continue to meet their obligations to all eligible unemployed workers.17 Federal advances accrue interest under certain circumstances. Since states may not use trust fund dollars to pay this interest, many states assess a separate tax on employers to cover this cost. Also, in general, should a state have outstanding federal advances at the beginning of two consecutive calendar years, its employers' FUTA tax credit begins to be reduced, with the additional revenue being used to pay back the federal debt. States may avoid or limit the credit reduction if certain requirements are met.18
Maintaining sufficient reserves of benefit funding is essential in order to mitigate the likelihood of large fluctuations in employers' state UI tax liability from year to year, and especially large increases as the economy recovers from a recession. Sustained insolvency results in marked increases in employers' total unemployment-related costs: the schedule of applicable state tax rates increases and/or a solvency add-on tax may be triggered, additional state taxes to pay interest may be assessed, and net federal unemployment taxes may increase to pay down the outstanding federal advances to the state to pay benefits.
Solvency is an important objective, but there are no requirements for it in federal law.19 Nonetheless, states have been encouraged to forward-fund their accounts in the UTF. The Average High Cost Multiple (AHCM) measures state solvency. Using data from the most recent three recessions to determine high benefit costs in a state, the AHCM measures how long the state could pay benefits when the payment levels are high, given the current balance in the state's account in the UTF. It is recommended that states maintain trust fund balances sufficient to pay benefits for one year at recessionary levels.20 However, since states have great latitude when designing their UI tax structure and the revenue it is expected to yield, some states have opted to follow more of a "pay as you go" model that keeps employer taxes low but does not generate enough revenue to build significant reserves for use during the next economic downturn. As explained above, this can result in greater volatility in the state and federal unemployment tax payments that employers make.
In recognition of the fact that regular state UI benefits provide insufficient income support for many unemployed workers during recessionary periods and federal involvement is needed,21 the federal-state EB program is intended to provide additional weeks of UI benefits when unemployment increases to recessionary levels.22 Benefit costs are shared equally by the state and federal government. States trigger EB when their unemployment rates exceed certain levels and are higher than they had been in recent years. All states must have an EB trigger based on the insured unemployment rate (IUR), which includes data concerning individuals who are currently receiving regular UI benefits. The 13-week IUR must be at least five per cent and be at least 120% of the rate for an equivalent 13-week period in each of the preceding two calendar years. Under the IUR trigger, individuals may receive up to 13 additional weeks of benefits. If a state uses an optional total unemployment rate (TUR) trigger23 – which uses household survey data from the Bureau of Labor Statistics about individuals who are not working and have looked for work during the last four weeks – individuals may receive up to 13 or even 20 weeks of additional benefits, depending on the state's TUR. Up to 13 weeks of benefits would be available if the state's three-month TUR is at least 6.5% and at least 110% of the rate for the corresponding three-month period in either of the two previous calendar years. A total of up to 20 weeks of benefits would be available if the state's three-month TUR was at least 8.0% and the rate met the 110% "lookback" requirement. Because these triggers are not very responsive to economic downturns and states historically have not triggered EB (or not done so soon enough) during recessions, especially via the IUR trigger, special federal programs have been created to provide additional weeks of benefits to unemployed workers.
While providing benefits to individuals unemployed through no fault of their own is the overall mission of the UI program, since its inception there has been explicit acknowledgement of the importance of helping individuals who receive UI benefits to become reemployed. While some workers maintain their attachment to a job (i.e. they are on a temporary layoff), most do not. It is for this reason that federal law requires UI payments to be made through public employment offices.24 As UI claims-taking moved out of local offices and was increasingly handled over the phone or via the internet, the connection to public employment offices weakened in some states. In order to strengthen this connection, several strategies were implemented, including the development of the UI Reemployment and Eligibility Assessment initiative. Since 2005 funds have been appropriated to the US Department of Labor to enable states to address the individual reemployment needs of UI claimants and to prevent and detect improper benefit payments. The results have been promising with respect to reducing the number of weeks claimed and compensated, the likelihood of exhausting UI benefits, and improper payments.25 Due to its early successes, funding has increased. In addition, the program was renamed Reemployment Services and Eligibility Assessments (RESEA), which reflects a narrower focus on individuals who are most likely to be long-term unemployed and on individuals who transitioned out of the military. Recognizing the need for increased reemployment services for these individuals, RESEA funding may now be used for this purpose.
Since enactment of the SSA in 1935, federal law requires, as a condition of grant receipt, that states have methods of administration "reasonably calculated to insure full payment of unemployment compensation when due".26 This is broad statutory authority that is the basis for many things, including federal performance measures (payment when due includes timeliness) and improper payment reduction and collection (payment when due also means no payment when not due). This also is the basis for federal monitoring and oversight of state UI programs. For example, in the Benefit Accuracy Measurement program a sample of weekly claims are selected and every aspect of eligibility is verified. Federal UI programs are also subject to review to ensure that states are administering them in a manner consistent with all operating instructions.
As briefly mentioned above, there were substantial political and legal challenges that had to be overcome in order to establish unemployment insurance in the US. Although benefit adequacy varies among the states, data indicate that UI provides substantial macroeconomic stabilization. UI benefit spending during the Great Recession generated a significant multiplier effect, with every benefit dollar generating about two dollars of economic activity. State benefits filled about one-tenth of the real gross domestic product shortfall, and federal extensions of benefits filled about one-twelfth of the shortfall.27 UI benefits are also important for poverty reduction, particularly during recessions. In 2011 the poverty rate for individuals in families in which someone received UI benefits was reduced by almost 40%.28
In the US, the federal role regarding UI benefit payments is largely limited to economic downturns, when the federal government serves as a funding source for states so they can continue their benefit payments and when it provides federal assistance to individuals later in their unemployment spell. Especially since many states are limited in their authority to take on debt, the ability to get advances from the federal government so they can meet all of their obligations has been extremely important. In addition, the fact that more than half of the states changed their laws to take advantage of temporary full federal funding of EB demonstrates the importance of an increased federal role during recessions; states were unable to meet this need on their own.
Some may suggest that the federal role should be larger, which could include some level of harmonization of benefits among the states and solvency standards; several states have reduced benefit levels due in large part to the lingering effects of their benefit obligations in recent years. However, in a federal system there is a delicate balance between maintaining state autonomy to determine the parameters of their programs and establishing greater uniformity through a stronger federal role. Within the constraints of such a system, it is reasonable to say, overall, that the US model is fairly effective.
A challenge that has become quite evident in recent years is the responsiveness of the EB program. While intended to provide additional weeks of benefits during economic downturns, EB generally does not become available soon enough (or at all). For this reason, special temporary federal programs have been created during recessions to assist long-term unemployed workers.29 Thus, reforming the EB program to make discretionary action unnecessary is an area of great interest.
Lessons for the EU
As options are evaluated when considering the development of a European unemployment insurance scheme, there is much to learn from the American experience. In addition to the insights from the overall assessment above, I will briefly highlight a few others. At a conceptual level, the fact that the American UI programs are premised on strong incentives (grants and tax credits) rather than actual requirements may be valuable, given the legal construct of the EU. In addition, genuine federal UI programs in the US are not administered by the federal government – the states operate them, which obviated the need to create a new, duplicative organizational infrastructure. This may be particularly appealing because it would avoid the need for a marked expansion of EU administrative functions.
The way American UI benefits are financed may also be noteworthy. While permanent transfers among European member states are of great concern,30 this issue is not much of a factor in the US. While there are likely many reasons for this, the fact that federal UI funding comes from employers, rather than from the states, may be a contributing factor. In addition, creation of a federal trust fund in which states are required to deposit all UI benefit funds and from which states may only withdraw funds to make UI benefit payments is valuable, because it helps guarantee that resources will be available when they are needed.31
The US program incorporates experience rating, which has been considered in Europe.32 Conceptually, this is consistent with the insurance principle underlying the program, because "premiums" increase when the insured event occurs more frequently. While in the American system, this rating is carried out at the employer level rather than the state level, some useful insights may be attained. For example, all US states have provisions allowing for "non-charging" when an employer is not at fault for the individual's unemployment. There are a wide variety of reasons, including if an individual voluntarily quit a job for compelling personal reasons, if the unemployment was due to a natural disaster, or – for long-term unemployment during recessions – if the state of the economy, rather than the employer's actions, caused the unemployment. Socialization of some benefit costs is essential; otherwise, UI would become a self-insurance scheme, which would not be likely to provide sufficient pooling of risk to ensure sustained availability of benefits. Thus, exceptions to experience rating would be worthy of consideration in the European context.
Institutional moral hazard is another concern in Europe.33 While this is not a significant issue in the US for UI, individual moral hazard is. (In this case, moral hazard refers to the effect benefit payments may have on institutions' and/or individuals' actions such that they may increase the likelihood of greater future receipt of benefit payments.) The American approach to addressing this concern may offer a partial solution to the EU: active labor market measures. The RESEA initiative described above is, in part, intended to ensure that individuals who receive UI benefits are still in the labor market and searching for work. By themselves, such requirements may not be adequate to ameliorate concerns about institutional moral hazard in Europe. However, adding rigorous monitoring and oversight, along with establishing a set of measured, incremental consequences for inadequate state compliance, may suffice. Note that in the US, programmatic performance reviews are not just conducted by federal employees; some are done by employees of other states in "peer reviews". Thus, a large cadre of federal staff is not necessary to accomplish this objective.
From this brief description of the UI program in the US, its strengths and its opportunities for improvement, it should be evident that there is a wealth of knowledge and experience to be drawn upon as Europe considers its own version of a federal-state partnership to address the needs of its unemployed workers and the economic well-being of its member countries. A multi-layered governing system is, by its nature, incredibly complicated, and it is not an easy task to undertake. However, it should be clear that this is an endeavor worth pursuing and that programmatic features can be designed to adequately address the wide array of policy concerns while avoiding unintended negative consequences.
The opinions expressed in this paper are the author's alone. They do not purport to reflect the official position or views of the US Department of Labor.
1 L. Andor: Towards shared unemployment insurance in the euro area, in: IZA Journal of European Labor Studies, Vol. 5, No. 1, 2016, pp. 1-15.
2 M. Beblavý, K. Lenaerts: Feasibility and Added Value of a European Unemployment Benefits Scheme, CEPS e-Book, 2017.
3 While this overview includes many key aspects of the American UI program, it is not comprehensive. It focuses only on the aspects of the program pertinent for this paper – consideration of a European unemployment insurance program.
4 The underlying issue was the tax burden on employers that would be needed to fund the program and the concern that employers would move to a state with a lower tax burden. This is a classic example of the prisoners' dilemma, in which a socially optimal outcome cannot be achieved because the parties are concerned that they will not all act in their collective best interest.
5 W. Haber, M.G. Murray: Unemployment Insurance in the American Economy, Homewood IL 1966, Richard D. Irwin.
6 US Department of Labor: Unemployment Compensation. Federal-State Partnership, April 2016, available at https://www.ows.doleta.gov/unemploy/pdf/partnership.pdf.
7 US Department of Labor: Significant Provisions of State UI Laws, January 2017, available at https://www.ows.doleta.gov/unemploy/content/sigpros/2010-2019/January2017.pdf.
8 During the Great Recession, federal law was amended to temporarily make EB benefits 100% federally funded.
9 The permanent programs are Unemployment Compensation for Federal Employees, Unemployment Compensation for Ex-Servicemembers, Disaster Unemployment Assistance and Trade Readjustment Assistance. The most recent special temporary federal extension program was Emergency Unemployment Compensation of 2008.
10 Federal law requires states, as a condition of FUTA tax credit receipt, to provide for benefit payments if an otherwise eligible individual does not live in the state and to "combine" wages earned in more than one state when establishing eligibility. States are charged a proportionate amount of the cost of benefits based on the individual's base period work in the state.
11 Washington State does not determine UI eligibility based on earning wages equal to or exceeding a specified amount. Instead, state law requires an individual to have at least 680 hours of base period employment.
12 A. Mastri, W. Vroman, K. Needels, W. Nicholson: Final Report, 2 March 2016, Mathematica Policy Research.
13 Following the enactment of the American Recovery and Reinvestment Act, 23 states enacted new or modified existing alternative base periods because funding was made available via UI Modernization incentive payments.
14 US Department of Labor: 2017 Comparison of State Unemployment Insurance Laws, 2017, available at https://www.oui.doleta.gov/unemploy/pdf/uilawcompar/2017/nonmonetary.pdf.
15 Alaska, New Jersey and Pennsylvania levy nominal UI taxes on workers under certain limited circumstances. In Alaska, the tax rate is equal to 27% of the average benefit cost rate, but not less than 0.5% or more than 1.0%. In New Jersey, the tax rate is 0.3825%. Depending on the adequacy of the fund balance in a given year, Pennsylvania employees pay contributions ranging from 0.0% to 0.08% of total gross covered wages earned in employment.
16 See US Department of Labor: 2017 Comparison..., op. cit. Until June 2011, the FUTA tax was 6.2% and the effective FUTA tax rate was 0.8%. The 0.2% "surtax" was originally added in 1985 to help the federal accounts in the UTF pay back their advances from the General Fund of the US Treasury. Between advances to states and federal benefit spending during the Great Recession, the federal accounts in the UTF ran out of funds and had to borrow to meet all obligations.
17 During the Great Recession, 36 states borrowed from the federal government, and the peak amout of the advances was $47.2 billion. However, states may use other state funds or may borrow from other sources to pay UI benefits. Six states borrowed via bonding during the Great Recession; on 31 December 2013, the outstanding bond amount was $9.725 billion.
18 The state must apply for and be found eligible for relief from tax credit reduction in the form of avoidance or a cap on reduction. See section 3302, FUT; and section 901(d)(1), SSA.
19 Federal regulations have been modified with the intent of providing states with an incentive for attaining solvency by making it a condition of receiving interest-free short-term advances. See 20 CFR 606.32.
20 Advisory Council on Unemployment Compensation: Collected Findings and Recommendations 1994-1996, Washington DC 1996, U.S. Government Printing Office.
21 Some states have "additional" benefit programs to provide additional weeks of benefits to unemployed workers under certain circumstances, including economic downturns. However, most states do not.
22 The Federal-State Extended Unemployment Compensation Act of 1970 made EB payment a condition of FUTA tax credit receipt.
23 When EB law was temporarily amended during the Great Recession to provide for 100% federal funding of benefit costs, about half of the states temporarily amended their EB laws to include the TUR triggers so EB would be available to their unemployed workers without the state having to pay any of the benefit costs.
24 See section 3304(a)(1), FUTA, and section 303(a)(2), SSA.
25 J. Benus, E. Poe-Yamagata, Y. Wang, E. Blass: Reemployment and Eligibility Assessment (REA) Study, FY 2005 Initiative, Final Report, March 2008, IMPAQ International.
26 See section 303(a)(1), SSA.
27 W. Vroman, J.M. Benus: The Role of Unemployment Insurance As an Automatic Stabilizer During a Recession, July 2010, IMPAQ International.
28 T. Gabe, J.M. Whittaker: Antipoverty Effects of Unemployment Insurance, Report for Congress, Washington DC, October 2012, Congressional Research Service.
29 US Department of Labor: Temporary Federal Benefit Extension Programs, January 2013, available at https://ows.doleta.gov/unemploy/spec_ext_ben_table.asp.
30 J.-C. Juncker, D. Tusk, J. Dijsselbloem, M. Draghi, M. Schulz: Completing Europe's Economic and Monetary Union, Five Presidents' Report, European Commission, 2015.
31 There are very limited exceptions to this "withdrawal standard" in section 3304(a)(4), FUTA, and 303(a)(5), SSA.
32 J. Fath, C. Fuest: Experience Rating of Unemployment Insurance in the US: A Model for Europe?, CESifo DICE Report, Vol. 3, No. 2, January 2005, pp. 45-50.
33 F. Vandenbroucke, C. Luigjes: Institutional Moral Hazard in the Multi-Tiered Regulation of Unemployment and Social Assistance Benefits, CEPS Special Report No. 137, April 2016.