February 9, 2012

February 9, 2012 | Testimony on Labor

Attempting to Bolster the Economy: The Effectiveness of Extending UI Benefits

Testimony before
The Joint Economic Committee
United States Congress

February 7, 2012

James Sherk
Senior Policy Analyst in Labor Economics
The Heritage Foundation

Chairman Casey, Vice Chairman Brady, and Members of the Joint Economic Committee, thank you for inviting me to testify before you today. My name is James Sherk. I am a Senior Policy Analyst in Labor Economics at The Heritage Foundation. The views I express in this testimony are my own, and should not be construed as representing any official position of The Heritage Foundation.

Extending the length of time during which workers can collect unemployment insurance (UI) benefits in a recession makes sense, but only for humanitarian reasons. Extended benefits moderately increase the unemployment rate; they do not bolster the economy. Sixty weeks of benefits represent a proportionate increase in UI duration that matches the increased difficulty of finding work. Two years of benefits are excessive and counterproductive.

Congress should also reform the UI system to place a greater focus on returning the unemployed to work. Currently, the system focuses almost entirely on distributing UI checks. Congress should both increase job-search requirements—which most unemployed workers already fulfill—and enable the states to experiment with innovative strategies to help the unemployed find jobs.

The Unemployment Insurance System

Congress created the unemployment insurance system in the 1930s as an insurance system to enable workers who lose their jobs to smooth consumption until they find new work. It is an unemployment insurance system: The program insures workers against the risk of a harmful event outside their control. Consequently, only involuntarily unemployed workers may collect benefits. Workers who voluntarily leave their jobs may not collect benefits, nor may workers who are not searching for new work. UI normally provides workers with six months (26 weeks) of benefits through a system largely financed by state-level taxes. Workers in states with higher unemployment levels may collect an additional three months (13 weeks) of benefits through the jointly funded state–federal Extended Benefits (EB) program.

Since the recession started, Congress has increased the length of time that workers on UI can collect benefits. Congress created the Emergency Unemployment Compensation (EUC) system with a maximum of 53 weeks of benefits, while taking over all the financing of the EB system and extending it to 20 weeks. As a result, workers in many states can collect up to 99 weeks of benefits—almost two years.

Effect on Job Search

Like most insurance programs, unemployment insurance suffers from moral hazard. UI payments make being unemployed less costly, causing UI recipients to take longer to find new work.

The fact that UI benefits cause workers to stay unemployed longer is one of the most conclusively established findings in labor economics.[1] Researchers of every political persuasion have come to this conclusion. Even Alan Krueger, chairman of President Barack Obama’s Council of Economic Advisers, agrees.[2]

Contrary to some stereotypes, however, this phenomenon does not primarily occur because most workers on UI enjoy receiving government handouts. While some individuals do abuse the system, UI benefits are not especially generous. They typically replace between 35 percent and 40 percent of an employee’s previous income. Relatively few unemployed people stay jobless on purpose in order to receive UI checks from the government.[3]

Instead, most of the effect comes from changing how the unemployed search for jobs. UI benefits reduce the urgency of finding a new job immediately. This enables the unemployed to focus their search on jobs they prefer to find. Often this means looking for jobs near the city, the occupation, and/or salary they had before. As benefits draw down, workers widen their search to jobs they would not previously have considered. Alan Krueger finds that the amount of time that workers who collect UI spend job hunting triples when their benefits start to run out.[4]

Moderately Higher Unemployment Rate

Extending unemployment benefits to two years has kept many workers unemployed longer than they otherwise would have been. Researchers from many institutions, including Federal Reserve Banks, have examined how this affects the unemployment rate. They concluded that extending benefits to 99 weeks has increased the unemployment rate by approximately 0.5 percentage point.[5] Extended unemployment benefits have had a nontrivial effect on the unemployment rate, but they are not the main reason it remains so high.

Appropriate Duration

Providing UI benefits for a longer time during a recession makes sense because job loss becomes more costly when it takes longer to find new work. Consequently providing the same insurance against the risk of unemployment requires extended benefits. The important question is “How long should the government provide benefits?”

To answer this question Congress needs to balance several factors. Additional benefits provide workers with more resources and flexibility. However, they also increase unemployment and can hurt those they are meant to help. Many of the jobs that unemployed workers would prefer to find do not exist and will not return.[6] To find work, many workers will have to take positions that are much less than ideal. Extending benefits for too long encourages the unemployed to search for jobs that they will not find. This can hurt them in the long run.

Congress also has to consider fairness to taxpayers. The federal government is running record deficits. Current and future taxpayers should not be asked to pay for unreasonably extended benefits.

Even with the difficult job market, 99 weeks—almost two years—of benefits is excessive. In normal economic times, the average unemployed worker is jobless for four months.[7] During these times the government provides six months of unemployment insurance coverage—50 percent longer than the average duration of unemployment. In the current recession the average duration of unemployment has risen to 40 weeks (nine months).[8] Providing 60 weeks of UI payments would increase benefits proportionate to the deterioration of the labor market.

Congress should continue to offer UI benefits for periods that exceed six months—but not as long as two years. Congress should reduce the extended-benefit period to an appropriate level (around 60 weeks, down from the current 99 weeks). Especially in light of recent improvements in the labor market, taxpayers should not be required to pay for two years of UI benefits.

Keynesian Models

The arguments for extended benefits must rest on humanitarian grounds. Congress should not expect extended UI benefits to provide stimulus. While it would be nice if extended UI benefits also boosted the economy, they do not.

Some economic models, particularly those of Mark Zandi and some of the models at the Congressional Budget Office, do forecast that spending more on UI benefits boosts the economy. These models typically show that spending $1 on additional UI benefits increases economic output by between $1.50 and $1.90.[9] It is important to understand that such results are pre-programmed into these models.

A computer model is only as good as the assumptions built into it. The macroeconomic models that find that UI payments stimulate the economy are Keynesian models. They assume that government spending adds to total economic output, or that government spending adds value to the economy. It does so through a “multiplier” effect in the economy. That is, when the government spends a dollar, the recipients of that dollar spend it elsewhere. The recipients of those dollars then spend it elsewhere again, and so on, boosting demand and spurring production throughout the economy above what the private sector would produce. In this theory, government spending is the ultimate free lunch: Each dollar the government spends creates more than a dollar of economic output.

Keynesian models naturally show that extending UI benefits stimulates the economy. UI spending gives money to households, who the models assume spend it immediately, creating the multiplier effect and stimulating GDP. Given these assumptions it is virtually impossible for these models to come to any other conclusion.

No Economic Stimulus

Unfortunately, there is no such thing as a free lunch. Keynesian theories and models do not accurately describe how the economy operates. Many of these models assume that individuals consume almost every dollar of UI benefits they receive.[10] However, empirical research shows that receiving a dollar of UI benefits increases household consumption by just $0.55.[11]

Consumption does not rise by more because unemployment benefits change household behavior. For married men, each dollar of UI benefits reduces their wives’ earnings by between 36 cents and 73 cents.[12] The fall in spousal income partly offsets the increase in UI benefits. Workers also spend more of their savings if they do not have UI.[13] Extended UI benefits provide alternative financing for some consumption that would take place nonetheless. Assuming that households consume every dollar of benefits artificially inflates their modeled “multiplier effect” on the economy.

More fundamentally, the “multiplier effect” only looks at half the story. The resources that the government spends do not materialize out of nowhere. They are borrowed or taxed from elsewhere in the economy. This reduces spending and demand. Further, government borrowing redirects resources away from productive investments that produce economic value.

Empirical studies show that increasing government spending reduces private-sector output. Recent empirical work published by the National Bureau of Economic Research concludes that the multiplier is approximately 0.5. For each dollar the government spends, the private sector produces $0.50 less.[14] Other studies come to similar conclusions.[15] The economy does not operate the way Keynesian models assume it does. The multiplier is actually a divisor.

The models are wrong about how government spending generally affects the economy. Their forecasts about UI spending are similarly inaccurate. From 2008 through 2011, the government increased UI spending by $300 billion.[16] Congress has repeatedly heard testimony that extending UI benefits will bolster the economy. The White House predicted that if Congress passed the stimulus—which included UI extensions—unemployment would not rise above 8 percent. Congress passed the stimulus.[17] Unemployment rose to 10 percent, and Americans have suffered through the slowest recovery of the post-war era.

This should not have come as a surprise. Macroeconomic empirical studies have demonstrated that UI spending has, at most, a small effect in stabilizing the economy.[18] Empirical research into UI spending in individual states also finds it has negligible economic effects.[19]

The International Experience

International evidence reinforces this conclusion. If UI spending stimulated the economy, unemployment would rise less rapidly in countries with more generous benefits. Instead, the opposite happens. Unemployment rises faster in countries that provide more extensive benefits.[20] The disincentive effects of UI overwhelm any stimulative effects.

This may be why the argument that the government should spend heavily on UI to stimulate the economy is rarely heard internationally. In few other OECD countries do policymakers argue that UI boosts demand and employment.[21] Generous UI systems have not stimulated European nations out of persistently high unemployment. Just the opposite. Countries that reduced the generosity of UI benefits saw their unemployment rates fall.[22]

Policy Tradeoffs

This does not mean that Congress should return to six months of benefits. It means that Congress faces economic tradeoffs. The humanitarian benefits of extending UI come at a fiscal and economic cost. Congress can certainly conclude that the benefits outweigh the costs. But any extension of benefits should recognize these tradeoffs. If extending benefits is an important priority, they should be paid for by reducing spending on less important programs.

Additional Problems in the UI System

Changes to the UI system should extend beyond changing the weeks of benefits provided. The UI system currently places little emphasis on returning the unemployed to work. Instead, UI administration largely focuses on distributing benefits to covered workers. Job-search assistance is often a secondary concern.

The UI system also has few safeguards to prevent abuse. The federal government does not require workers who receive EUC or Extended Benefits to search for work. State laws do require claimants in the regular UI system to search for a job. However, states do little to verify that workers actually do so. In most states, claimants reapply for their benefits either online or by calling an automated hotline. They indicate they have contacted employers by clicking a box or pressing a button. Most states do little to follow up to ensure that workers were truthful.[23] This saves money on overhead and administration, but claimants who collect benefits without looking for work face few consequences.

Substantial evidence suggests that tightening requirements can reduce the time workers spend on UI. In the mid-1990s, the state of Maryland conducted a series of experiments. The state told some workers that the government would verify their employer contacts.[24] They required other workers to attend a four-day workshop on how to search for a new job. Other workers stayed on the regular program. The workers with the more stringent requirements spent 5 percent to 8 percent less time collecting UI than the workers in the regular system.[25] Interestingly, most of the increased job-finding by those assigned to the workshop occurred before the workshop began. It appears that the cost of spending four days in a workshop prompted UI recipients who had the ability to return to work to do so. Most other studies also find that penalizing inadequate job search reduces the time workers spend on UI.[26]

The vast majority of workers on UI try to find work. However, screening out the minority who do not would save a lot. Spending 5 percent less on UI extensions would save $2 billion a year.[27]

Reforming UI

The government should reform the UI system to address both these problems. The unemployment insurance system should focus on returning the unemployed to work. Those who can work should not be allowed to abuse the UI system.

The federal government should require workers claiming extended benefits to actively search for a job. However, the unemployment insurance system operates as a joint state–federal program. The federal government should not trample on the states as it reforms UI. Rather, the federal government should work with the states to improve the system. The provisions in the House-passed payroll tax cut and UI extension (H.R. 3630) provide a good starting point for such reforms. The House bill required workers who receive UI benefits to:

  1. Actively search for work, in such manner as states direct;
  2. Register with state re-employment services within 30 days;
  3. Post a resume or job application on a state database; and
  4. Participate in any re-employment services to which the state refers them.

The federal government should also allow the states to experiment with larger reforms to the UI system. Federal law tightly restricts how states can use UI funds. They may only use them to pay for UI benefits or administrative costs. This prevents states from enacting more innovative reforms focused on returning the unemployed to work.

Some analysts have proposed wage subsidies for employers that hire workers on UI. Others have suggested intensive job-search assistance services or employer-sponsored job-training programs. Others have pointed out that 99 weeks is enough time to earn an associate’s degree. States could require UI claimants to study for a degree. Online technology would allow states to do this at low cost while allowing workers to study from home—and not disrupting their job search.

Congress does not know which of these programs will succeed and which will not. Congress should give states the flexibility to experiment with UI reforms, such as through a waiver system. States could innovate and policymakers would learn what works and what does not. Congress should allow the states to innovate.

Congress should also give states more flexibility under the existing system. The stimulus funded a $25 increase in weekly UI payments. To prevent states from simply reducing their UI benefits by an equal amount, Congress passed a “non-reduction” rule. The rule prevents states from reducing their UI benefit amounts. Congress did not renew the supplemental federal payment, but the non-reduction rule remains on the books. This has handcuffed states as they try to close shortfalls in their UI trust funds. Several states have turned to cutting their weeks of benefits to reduce costs. Congress should once again let states determine the appropriate mix of benefit levels and duration.


Although two years of benefits is excessive, extending the duration of UI benefits in a recession is reasonable. However, the arguments for this policy should rest on humanitarian grounds. U.S. and international evidence shows that spending more on UI moderately increases unemployment. It has not, and will not, stimulate the economy.

About the Author

James Sherk Research Fellow, Labor Economics
Center for Data Analysis

Show references in this report

[1]David Card and Phillip B. Levine, “Extended Benefits and the Duration of UI Spells: Evidence from the New Jersey Extended Benefit Program,” Journal of Public Economics, Vol. 78, No. 1–2 (October 2000), pp. 107–38; Lawrence Katz and Bruce Meyer, “The Impact of the Potential Duration of Unemployment Benefits on the Duration of Unemployment,” Journal of Public Economics, Vol. 41, No. 1 (1990), pp. 45–72; Stepan Jurajda, “Estimating the Effect of Unemployment Insurance Compensation on the Labor Market Histories of Displaced Workers,” Journal of Econometrics, Vol. 108, No. 2 (2002), pp. 227–252; John T. Addison and Pedro Portugal, “How Does the Unemployment Insurance System Shape the Time Profile of Jobless Duration?” Economics Letters, Vol. 85, No. 2 (November 2004), pp. 229–34; and Rafael Lalive, Jan Van Ours, and Josef Zweimüller, “How Changes in Financial Incentives Affect the Duration of Unemployment,” Review of Economic Studies, Vol. 73, No. 4 (October 2006), pp. 1009–1038.

[2]Alan B. Krueger and Bruce D. Meyer, “Labor Supply Effects of Social Insurance,” in A. J. Auerbach and M. Feldstein (eds.), Handbook of Public Economics, First Edition, Vol. 4 (2002), pp. 2327–2392.

[3]Raj Chetty, “Moral Hazard versus Liquidity and Optimal Unemployment Insurance,” Journal of Political Economy, Vol. 116, No. 2 (2008), pp. 173–234.

[4]Alan Krueger and Andreas Mueller, “Job Search and Unemployment Insurance: New Evidence from Time Use Data,” Journal of Public Economics, Vol. 94, No. 3–4 (2010), pp. 298–307.

[5]Bhashkar Mazumder, “How Did Unemployment Insurance Extensions Affect the Unemployment Rate in 2008–10,” Federal Reserve Bank of Chicago Essays on Issues No. 285, April 2011; Jesse Rothstein, “Unemployment Insurance and Job Search in the Great Recession,” NBER Working Paper No. 17534, October 2011; and Rob Valletta and Katherine Kuang, “Extended Unemployment and UI Benefits,” Federal Reserve Bank of San Francisco Economic Letter No. 2010–12, April 19, 2010.

[6]For example, over half of net employment losses occurred in the manufacturing and construction sectors. Many workers in these sectors will need to switch industries.

[7]Department of Labor, Bureau of Labor Statistics, “The Employment Situation,” Table A-12 / Haver Analytics. The average duration of unemployment in non-recessionary periods is between 16 weeks and 18 weeks—four months.

[8] Ibid.

[9]See, for example, Mark M. Zandi, “Assessing President Bush’s Fiscal Policies,” Economy.com, July 2004, Table 4, at http://www.pbs.org/wsw/opinion/zandionbush.pdf (February 8, 2012) and Congressional Budget Office, “Options for Responding to Short-Term Economic Weakness,” January 2008, pp. 17, 22.

[10]See, for example, Wayne Vroman, “The Role of Unemployment Insurance as an Automatic Stabilizer During a Recession,” IMPAQ International, July 2010, p. 33, at http://wdr.doleta.gov/research/FullText_Documents/ETAOP2010-10.pdf (February 8, 2012). “These transfer payments are then almost entirely spent on consumption items in the same year.”

[11]Jonathan Gruber, “The Consumption Smoothing Benefits of Unemployment Insurance,” American Economic Review, Vol. 87 (March 1997), p. 195. Note that a 10 percent increase in the replacement rate (representing a 10 percent increase in individual income) reduces the fall in individual consumption by 2.65 percent. Footnote 9 of the Gruber paper notes that the average recipient obtains 48 cents of every additional dollar of which he or she is eligible because not all workers who are eligible for benefits receive them. So when UI raises incomes by 4.8 percent, consumption rises by 2.65 percent. Each dollar spent on UI raises consumption by approximately 55 cents.

[12]J. B. Cullen and J. Gruber, “Spousal Labor Supply as Insurance: Does Unemployment Insurance Crowd Out the Added Worker Effect?” Journal of Labor Economics, Vol. 18, No. 3 (2000), pp. 546–572.

[13]Eric M. Engen and Jonathan Gruber, “Unemployment Insurance and Precautionary Saving,” Journal of Monetary Economics, Vol. 47 (June 2001), pp. 545–579.

[14]Valerie Ramey, “Government Spending and Private Activity,” National Bureau of Economics Working Paper No. 17787, January 2012.

[15]Robert Barro and Charles Redlick, “Macroeconomic Effects from Government Purchases and Taxes,” Quarterly Journal of Economics, Vol. 126, No. 1 (2011) pp. 51–102 and Robert Hall, “By How Much Does GDP Rise if the Government Buys More Output?” Brookings Papers on Economic Activity Fall 2009, pp. 183–236.

[16]Heritage calculations using data from the Bureau of Economic Analysis, “Personal Income and Outlays,” Table 1, UI income from 2008–2011 / Haver Analytics. From 2005 through 2007, the government spent an average of $31.6 billion a year on UI. From 2008 through 2011, the government spent $427 billion on UI benefits—$300 billion more than if UI benefits had remained at normal levels. These figures include both regular state UI programs and the extensions funded by the federal government.

[17]Christina Romer and Jared Bernstein, “The Job Impact of the American Recovery and Reinvestment Plan,” January 10, 2009, p. 2, at http://otrans.3cdn.net/45593e8ecbd339d074_l3m6bt1te.pdf (February 8, 2012).

[18]George M. Von Furstenberg, “Stabilization Characteristics of Unemployment Insurance,” Industrial and Labor Relations Review, Vol. 29, No. 3 (April 1976), pp. 363–376 and Alan J. Auerbach and Daniel Feenberg, “The Significance of Federal Taxes as Automatic Stabilizers,” Journal of Economic Perspectives, Vol. 14, No. 3 (2000), pp. 37–56.

[19]Kyung Won Lee, James R. Schmidt, and George E. Rejda, “Unemployment Insurance and State Economic Activity,” International Economic Journal, Vol. 13, No. 3 (Autumn 1999), pp. 77–95.

[20]Andrea Bassanini and Romain Duval, “Unemployment, Institutions, and Reform Complementarities: Re-assessing the Aggregate Evidence for OECD Countries,” Oxford Review of Economic Policy, Vol. 25, No. 1 (Spring 2009), pp. 40–59.

[21]David Grubb, “Assessing the Impact of Recent Unemployment Insurance Extensions in the United States,” Organisation for Economic Co-operation and Development, Working paper, June 2011, p. 49.

[22] Ibid., pp. 46, 55–59.

[23]Testimony of Douglas J. Holmes, President, UWC–Strategic Services on Unemployment & Workers’ Compensation, before the Subcommittee on Human Resources, Committee on Ways and Means, United States House of Representatives. Hearing on “Improving Efforts to Help Unemployed Americans Find Jobs,” February 10, 2011.

[24]Most states did not switch over to online or phone systems to reauthorize claims until the late 1990s or early 2000s. Before then, workers in many states filed paper forms listing the employers they had contacted.

[25]Jacob Benus and Terry Johnson, “Evaluation of the Maryland Unemployment Insurance Work Search Demonstration,” Maryland Department of Labor, November 1997, at http://wdr.doleta.gov/owsdrr/98-2 (February 8, 2012).

[26]Peter Fredriksson and Bertil Holmlund, “Improving Incentives in Unemployment Insurance: A Review of Recent Research,” Journal of Economic Surveys, Vol. 20, No. 3 (2006), pp. 357–386.

[27]The Congressional Budget Office estimates that maintaining the benefit extensions will cost $50 billion in 2012. Five percent of $50 billion is $2.5 billion. See Congressional Budget Office cost estimates, “Budgetary Effects of the Temporary Payroll Tax Cut Continuation Act of 2011,” December 22, 2011. Note that the CBO estimates the two-month cost of the extensions to be $8.4 billion. The 12-month cost is six times greater, $50 billion.