Economic Effects of the Unemployment Insurance Benefit* BY SHIGERU FUJITA

he U.S. labor market has remained weak The unemployment insurance (UI) system constitutes one of T in recent years, even though the overall the major components of the economy itself has started to grow again after social security programs in the U.S.1 It provides income (and thus the deep recession. In response to the weak consumption) protection for those labor market conditions, the U.S. government has greatly who have lost their jobs involuntarily. During “normal” times, unemployment expanded the entitlement period of unemployment insurance benefits are provided insurance (UI) benefits. In this article, Shigeru Fujita through the regular unemployment compensation (UC) program, which reviews some of the academic literature on the economic is funded and administered at the effects of UI benefits. On the one hand, UI can improve state level. Regular benefits, which people’s well-being because it helps them avoid a large are paid weekly, replace 50 to 80 percent of pre-unemployment earnings drop in consumption in the face of job losses when job and last 26 weeks in the majority of 2 losers do not have enough savings. On the other hand, states. During economic downturns, however, the federal government there is a concern that it might produce an adverse often provides additional support by effect on the incentive to look for a job. The author extending UI benefits. Especially in the last few years, the U.S. government covers leading theoretical as well as empirical studies, has greatly extended the duration of which are useful in evaluating the recent expansion of benefits as a means to combat the surmounting joblessness. As of the unemployment insurance benefits. summer of 2010, unemployed workers who reside in states with a relatively high unemployment rate are entitled to receive UI benefits up to 99 weeks (26 The U.S. labor market has deep recession. For example, in the weeks of regular benefits and 73 weeks remained weak in recent years, even fourth quarter of 2009, the average of extended benefits). though the overall economy itself unemployment rate was at a double- Given the painful nature of job has started to grow again after the digit level, a level we have not seen losses, the merits of UI benefits are since the early 1980s, even though real often taken for granted in public policy GDP grew by more than 5 percent. discussions. In this article, I will review One of the main policy reactions to some of the academic literature on the painful developments in the labor Shigeru Fujita is market has been the expansion of a senior economist 1 in the Research unemployment insurance. The original framework of the unemployment compensation system is contained in the Social Department of Security Act, which was signed into law by the Philadelphia President Franklin Roosevelt in 1935. Fed. This article is 2 available free of *The views expressed here are those of the The benefit level is also subject to a cap. The charge at www. author and do not necessarily represent weekly maximum ranges from $200 to $600 philadelphiafed. the views of the Federal Reserve Bank of across states. Because of the cap, the average org/research-and-data/publications/. Philadelphia or the Federal Reserve System. replacement ratio is roughly 50 percent.

20 Q4 2010 Business Review www.philadelphiafed.org economic effects of UI benefits. This is extension is triggered automatically with the extensions enacted in the useful for evaluating the expansions of whenever the state unemployment past. During most of the post-WWII the UI system in recent years. rate reaches a certain level. It provides recessions, Congress has implemented First, UI can improve people’s additional weeks of unemployment federal emergency programs, but these well-being because it helps them avoid benefits up to a maximum of either 13 programs typically provided benefits large drops in consumption in the weeks or 20 weeks, depending on the for a total of around 60 weeks.4 Given face of job losses: The government state. past experience, the duration of UI provides an insurance against job loss. The second type is a federal eligibility in the most recent downturn There is, however, a concern that program that Congress enacts (that is, a total of 99 weeks) is quite it might produce an adverse effect temporarily during downturns. generous. on the incentive to look for a job. The latest program of this type, Figure 1 plots the number of UI That is, UI benefits could cause job the Emergency Unemployment recipients since December 2007, the seekers to put less effort into searching Compensation program (EUC08), start of the Great Recession. This for a job, consequently raising the includes those who are covered under unemployment rate. Some researchers the regular state programs as well as have argued that this incentive effect Historically speaking, those covered by the federal extension is large, given the observation that the scale of the programs. As can be seen from the the rate of exit from unemployment at figure, the number of claimants has the time of expiration of UI benefits extensions during the increased steadily since the start of increases noticeably. An important current downturn is the recession. One noticeable trend issue here is that the increase in the very large compared is the increase in the number of those exit rate from unemployment can be covered under the emergency programs driven by the fact that the worker with the extensions — it has more than doubled since the is simply dropping out of the labor enacted in the past. beginning of 2009. Because workers force, thereby losing eligibility for can be covered by the emergency UI benefits. This phenomenon can represents the eighth time Congress programs only after state UI benefits complicate the interpretation of the has created such a program.3 EUC08 are exhausted, the increase in the incentive effect. Other researchers was signed into law in June 2008. number of federal UI recipients implies also point out the possibility that UI Initially, the maximum entitlement that long-term unemployment is benefits enhance a firm’s incentive to period under this program was 13 increasing. create more jobs. Below, I will lay out weeks, but it has been extended several Figure 2 confirms this trend these arguments in detail. times since then. As of July 2010, from a separate data series based Before getting into the detailed EUC08 provides extended benefits on the Current Population Survey. discussion, let’s first briefly review for up to 53 weeks. This means that, The figure presents the total number recent developments regarding UI combining the regular benefit and of unemployed and those who are benefits and the U.S. labor market. the two emergency programs, an unemployed for 27 weeks or longer. unemployed worker is entitled to From this figure, we can see that the UNEMPLOYMENT INSURANCE UI benefits for up to 99 weeks. (See proportion of long-term unemployment 5 DURING THE GREAT The Chronology of the Emergency is rising rapidly. RECESSION Unemployment Compensation Program As mentioned above, regular (EUC08) for details.). unemployment insurance benefits Historically speaking, the scale 4 Again, see the article by Julie Whittaker for details of the previous programs. typically last 26 weeks. However, of the extensions during the current the federal government often enacts downturn is very large compared 5 Comparing the total number of benefit recipi- extensions of UI benefits during ents and unemployment allows us to see that economic downturns. There are two a substantial number of unemployed workers do not receive UI benefits. The main reason is types of federal emergency programs. that some workers are not qualified to receive The first is called the extended benefit 3 Congress created federal programs in 1958, them: To be eligible, workers must have at least 1961, 1971, 1974, 1982, 1991, 2002, and 2008. 20 weeks of full-time insured employment or (EB) program, which is permanently See the article by Julie Whittaker for details of the equivalent amount of work at insured wages authorized, meaning that the these programs. during the previous 12-month period.

www.philadelphiafed.org Business Review Q4 2010 21 These empirical observations FIGURE 1 underscore the importance of reconsidering the effects of UI benefits Unemployment Insurance Claimants on current labor market conditions.

Claimants (Thousands) Now let’s move on to the economics of UI benefits. 16000

A SIMPLE SEARCH MODEL An economic model called a 12000 “search model,” pioneered by John McCall and others, is often used to analyze the decisions facing a job 8000 seeker. In this model, the worker State receives occasional random job offers. How often the worker receives an offer depends on how hard he looks for a 4000 job. Once the offer has arrived, the Federal worker decides whether to accept or reject it. 0 One of the key implications of Dec Apr Aug Dec Apr Aug Dec Apr Aug 07 08 08 08 09 09 09 10 10 this model is that higher UI benefits lead to a longer duration of job search. The reason is that the worker puts less Source: Department of Labor effort into searching for a job because higher benefits mean that he has less to lose from being unemployed. Furthermore, he may hold out for a FIGURE 2 higher-wage job before accepting an offer, since higher benefits lower the Long-Term Unemployment cost of being out of work. This means that the arrival of an acceptable offer becomes less likely (that is, the chance Persons (Thousands) that the worker rejects the job offer is 16000 higher), and thus, the waiting time in the unemployment pool is longer. In this simple model, the reduction of the 12000 search effort caused by the increased benefit level is often called the moral Total Unemployed hazard effect.6

8000 An important thing to remem- ber here is that this simple model is designed to focus on the incentives to search for a job, omitting from consid- 4000 eration many issues that are relevant in Unemployed 27 Weeks or More reality. In particular, workers who have no savings at the time of job loss may 0 Dec Apr Aug Dec Apr Aug Dec Apr Aug 07 08 08 08 09 09 09 10 10 6 In more elaborate models, it can be misleading to label the decline in the effort level as moral Source: Bureau of Labor Statistics, Current Population Survey hazard. I will discuss those cases below.

22 Q4 2010 Business Review www.philadelphiafed.org The Chronology of the Emergency Unemployment Compensation Program (EUC08)

s mentioned in the main text, the benefit for those receiving benefits under the EUC08. EUC08 was originally signed into law in November 6, 2009. The duration of the EUC08 June 2008 but has been expanded several program was substantially expanded. Tier III and Tier A times since then. Below is the chronology IV were introduced. The Tier I benefit continues to be of EUC08. up to 20 weeks. The Tier II benefit was expanded to 14 June 30, 2008. The EUC08 program was intro- weeks from 13 weeks and no longer depended on a state’s duced. The maximum duration of the extended benefit unemployment rate. The new Tier III benefit provided under this program was 13 weeks. It was set to expire on up to 13 weeks to those workers in states with an average March 28, 2009. The expiration date is when the program unemployment rate of 6 percent or higher. The new Tier stops accepting new claimants. The existing claimants IV benefit may be provided up to an additional six weeks can continue receiving benefits until the entitlement if the state unemployment rate is at least 8.5 percent. period is over. The expiration date stayed the same as before (December November 21, 2008. The maximum entitlement 26, 2009). Again, after this date, the program would no period was extended from 13 weeks to 20 weeks. Tier II longer accept new claimants, and existing claimants in of benefits was introduced, providing up to an additional the lower tier cannot move to the next tier. 13 weeks of benefits for those who worked in states with December 19, 2009. The expiration date was ex- a total unemployment rate of at least 6 percent. It was set tended to February 28, 2010. to expire on March 28, 2009. After this date, the program March 2, 2010. The expiration date was extended to would no longer accept new claimants and existing claim- April 5, 2010. ants in Tier I cannot move to Tier II. April 15, 2010. The expiration date was extended to February 17, 2009. As part of the American Eco- June 2, 2010. nomic Recovery and Reinvestment Act, the expiration July 22, 2010. The expiration date was extended to date of EUC08 was extended to December 26, 2009. It November 30, 2010. also included a provision to pay an additional $25 weekly

experience a large drop in consump- Moffitt tests the implication of the period between 1978 through the first tion. Moreover, if the economy is not search model. He looks at how the quarter of 1983. Note that this is an- producing many jobs, it will be difficult unemployment exit rate (the rate at other period in which federally funded to exit unemployment by becoming which a worker exits from the unem- extended benefits were available. More employed rather than dropping out ployment pool) changes right before specifically, Congress enacted the of the labor force. In these cases, UI UI benefits are exhausted, exploiting Federal Supplementary Compensation benefits can improve the economy’s variations of maximum entitlement pe- (FSC) program in the fall of 1982, welfare, offsetting the negative incen- riods across states and across individu- which, combined with the regular tive effect. I will come back to these als within states. For example, imagine benefit and the benefit under the EB issues later. But for now, let’s take this that two workers who reside in two program, provided UI benefits for more simple model as a useful benchmark. different states have the same charac- than 60 weeks.8 teristics such as gender and education EMPIRICAL STUDIES FOR but have different unemployment exit TESTING THE MORAL rates. We can associate the difference 7 Similarly, there can be differences in the HAZARD EFFECT in the exit rates with the differences in generosity of the benefits even across workers within the same state. Is there empirical evidence that the generosity of UI benefits.7 Moffitt moral hazard is a serious problem of UI uses a high-quality data set collected 8 Since the FSC was enacted late in Moffitt’s benefits? A seminal study by Robert by state UI offices, which covers the data set, his analysis focuses on the workers who were receiving the benefits for at most 39 weeks.

www.philadelphiafed.org Business Review Q4 2010 23 The key finding is that there interpreted as moral hazard? In other Is it realistic to think that workers is a large spike in the exit rate from words, the presence of a spike in the are actually dropping out of the labor unemployment at the time UI benefits exit rate is consistent with the moral force once their benefits are exhaust- expire. Using a statistical technique hazard story, but there may be other ed? To appreciate this possibility, con- called a regression analysis, Moffitt stories consistent with the empirical sider the following example: A worker translates this large spike as indicating observation. One alternative story initially tried very hard to find a job, that, on average, a one-week extension is based on the so-called “reporting but after a series of unsuccessful job of benefits leads to an increase in the effect” of UI. searches, he became very discouraged. duration of unemployment of 0.15 However, to be qualified for UI bene- week. Using the same administrative REPORTING EFFECT OF UI fits, he is required to be “unemployed.” data set, a study by Lawrence Katz and To understand the reporting This means that he needs to fill out Bruce Meyer and one by Meyer extend effect, note first that the earlier claim forms periodically and may even Moffitt’s work and find a similar spike literature looks at the effects of UI on need to report to the local UI claims in the exit rate at the time benefits are the “exit rate.” However, “exiting from office to show that he is “actively exhausted. unemployment” does not necessarily looking for a job.” Once the benefit is Figure 3 presents the median mean finding a job. In other words, exhausted, these requirements cease duration of unemployment in recent it is possible that workers are simply to exist, and consequently, he officially years. It increased dramatically from dropping out of the labor force when exits from the unemployment pool. the pre-recession level of around eight their benefits expire. Because the data This appears to be a plausible possibil- weeks to around 20 weeks at the end set used in the aforementioned studies ity. Note that the reporting effect story of 2009. This has occurred in tandem is based on UI records, it does not tell involves little change in a worker’s with the increases in the number the labor market status of workers, that decision around the expiration date, of benefit claimants (see Figure 1). is, whether the worker found a job or yet it induces a large change in the There is no doubt that the recession simply dropped out of the labor force unemployment exit rate. In this sense, was the cause of the longer duration after exiting from the UI system. it is misleading to infer the extent of of unemployment. However, the literature suggests that at least part of the increase in the duration was FIGURE 3 actually caused by the extensions of UI benefits. Estimating “how much” Median Duration of Unemployment is beyond the scope of this article, but The Effect of the Extension of UI Weeks Benefits on the Unemployment Rate: 30 An Illustrative Example presents an example in which I calculate the effect of doubling the maximum benefit on 25 the observed unemployment rate using Moffitt’s result. The exact magnitude 20 of the effect aside, it seems plausible to say that the extensions played at 15 least some part in raising the duration of unemployment and thus the 10 unemployment rate. While this calculation as an accounting exercise is useful for 5 inferring the effect of the extended benefits on the unemployment rate, 0 Jan Jul Jan Jul Jan Jul Jan Jul Jan Jul Jan Jul there is good reason to be somewhat 05 05 06 06 07 07 08 08 09 09 10 10 careful about its interpretation. In particular, should it really be Source: Bureau of Labor Statistics, Current Population Survey

24 Q4 2010 Business Review www.philadelphiafed.org An Illustrative Example

n his study, Robert Moffitt estimates worker finds a job (that is, the job-finding rate) is 30 the effect of the extension of UI benefits percent per month, which implies that the duration on the duration of unemployment. He of unemployment of the average worker is 3.3 months I estimates that a one-week extension (approximately 13 weeks). These numbers are roughly of benefits results in an increase in the consistent with empirical observations. Also assume that duration of unemployment of 0.15 week, on average. employed workers are flowing into the unemployment Here, I take this estimate as given and calculate the pool at a rate of 2 percent per month. In the “steady effect on the unemployment rate when the benefit state,” where flows into and out of unemployment are entitlement period is doubled from 26 weeks to 52 weeks. equal to each other, the job-finding rate of 30 percent As mentioned in the main text and in the Chronology per month and the job-loss rate of 2 percent imply an of the Emergency Unemployment Compensation Program unemployment rate of 6.25 percent. (EUC08) on page 23, the maximum entitlement period Now assume that the maximum entitlement in the current downturn is 99 weeks. However, a worker period is increased from 26 weeks to 52 weeks. Moffitt’s may not have known at the time he lost his job that the estimate implies that the duration of unemployment entitlement period was 99 weeks because the extension goes up by 3.9 weeks. This translates into a decline announcement may have come after the initial job loss. in the job-finding rate from 30 percent per month to Furthermore, as explained in EUC08, after the expiration approximately 24 percent. I further assume that the job- date, workers can continue to be covered under the UI loss rate is unaffected by the extension. The steady-state program only up to the entitlement period of that tier. unemployment rate with the extended benefit entitlement Given these considerations, I only look at a simple case of period then becomes 7.7 percent. doubling the entitlement period. First, assume that the rate at which the average

moral hazard based on the size of the Chetty, and Andrea Weber did. people exit benefits by finding a new spike in the exit rate.9 One simple These authors analyzed this issue job. More specifically, Card and co- way to empirically distinguish them is using a rich data set from Austria. Ac- authors find a modest increase, roughly to examine whether the spike in the cording to the authors, the UI system 20 percent, in the re-employment unemployment exit rate is associated in Austria is similar to the one in the rate. They further point out that this with re-employment or dropping out U.S., although there are some insti- modest increase in the re-employment of the labor force. This is exactly what tutional differences. The data set is rate implies that less than 1 percent of a recent paper by David Card, Raj rich enough so that they can examine unemployment spells have an end date the effect of UI benefits on job finding that is manipulated to coincide with (not just exit from unemployment). the expiration of UI benefits.

9 Theoretically distinguishing the two stories When they focus on the unemploy- Several papers look at the effects requires extending the simple search model ment exit rate, they find a very large on re-employment rates using U.S. discussed above along several dimensions. For spike at the time of benefit exhaustion. data. A paper by Bruce Fallick, us- example, the simple search model does not incorporate the feature that workers’ skills can In their sample, the jump in the exit ing data from the Displaced Worker deteriorate while they are unemployed. In the rate amounts to 200 percent and is of Survey (DWS), finds that there is no model with such an extended feature, workers would reduce their search effort over time as similar magnitude to the one reported significant difference in the job-finding the value of work relative to being unemployed by Moffitt. However, when they con- rate after benefits have been exhaust- declines as their skills deteriorate. In such a sider only those who are re-employed, ed. On the contrary, Katz and Meyer model, the increase in the re-employment rate right before the expiration date can be much the spike almost disappears. In other argue that there is a significant spike smaller than that implied in the simpler search words, there is little evidence that in the re-employment rate associated model.

www.philadelphiafed.org Business Review Q4 2010 25 with the exhaustion of benefits, sup- ship between higher UI benefits and finding rates for workers with little porting the moral hazard story.10 the duration of unemployment. That wealth, while they have no noticeable While these data sets derived from is, UI benefits work as a mechanism impact on job-finding rates for workers surveys include information on work- to relax the liquidity constraint of with greater wealth. He then estimates ers’ labor market status (employed, un- unemployed workers. To understand that 60 percent of the increase in the employed, and out of the labor force), the idea, note first that in the simple duration of unemployment from higher thus allowing the researchers to dis- search model, the wealth level of the UI benefits can be attributed to the tinguish between the re-employment worker has no implications for his liquidity effect. He further develops a rate and the exit rate, the information or her search behavior. More to the simple method of calculating the econ- in these surveys is necessarily less ac- point, it does not suppose a situation in omy’s welfare gains from UI. Using this curate, compared with the data that which an unemployed worker accepts method, he concludes that a UI system come from UI offices. For example, the a low-paying job simply because he in which benefits replace 50 percent worker-level information regarding his needs to put food on the table. Is the of pre-unemployment earnings for or her maximum entitlement period underlying assumption of the standard six months is optimal. Note that this and the actual benefit-collection pe- search model realistic? Probably not. “optimal” system is close to the cur- riod can be subject to serious measure- Actually, there is ample empirical evi- rent U.S. system during normal times. ment errors.11 dence that many unemployed individu- Presumably, a more generous benefit Given the limitations of these als do not have enough savings, and structure is desirable during economic survey data sets (DWS and PSID), we thus, their consumption is quite sensi- downturns,12 although answering the can only agree with Card, Chetty, and tive to cash on hand (see, for example, question of how much more generous Weber that “the size of the spike in the study by Gruber). When workers the benefits should be during reces- re-employment rates at exhaustion in are subject to the liquidity constraint, sions requires further research. the current U.S. labor market remains the wealth level does have an effect an open question.” This is unfortu- on search behavior. In particular, UI JOB-CREATION EFFECT nate, but the argument made by Card, benefits increase cash on hand held by The discussion so far has focused Chetty, and Weber at least gives us a unemployed workers to support their on workers’ job-search behavior. Daron reason to keep the reporting effect in consumption. Higher benefits then Acemoglu and Robert Shimer point mind when thinking about the positive reduce the pressure to take a low-pay- out another welfare-improving effect of relationship between unemployment ing job, leading to the longer duration UI, one that works through the feed- duration and UI benefits in recent of unemployment. At least for these back effect on job creation. The au- years. workers, UI benefits work literally as thors develop a model in which there insurance against job loss. are two types of jobs: high-productivity LIQUIDITY EFFECT OF Note that, as opposed to the and low-productivity jobs. The high- UI BENEFITS moral hazard effect, the liquidity effect productivity jobs are harder to find, A study by Jonathan Gruber and highlights the aspect of UI policy but they pay a higher wage. Similarly, one by Raj Chetty provide another beneficial to the overall economy. low-productivity jobs are easier to find, possible reason (other than the moral The liquidity constraint limits the but they pay a lower wage. hazard story) for the positive relation- worker’s ability to take an “optimal” To understand how Acemoglu and action, such as declining what may be Shimer’s model works, think of a job a poor job match, an action he might acceptance decision of a worker who have taken if he had enough savings. has been offered a low-productivity job. 10 The study by Katz and Meyer (as mentioned in the previous section) mainly focuses on Relaxing the liquidity constraint Note that the trade-off is whether to unemployment exit rates, but they supplement through UI is then desirable from a accept this low-paying offer or to bet their analyses by attempting to distinguish policy perspective. on getting an offer of a high-productiv- between re-employment and exit. They use the Panel Study of Income Dynamics (PSID) for Chetty empirically shows that the ity job in the future. The latter choice this purpose. liquidity effect is sizable. Using U.S. involves giving up the income from

11 Another issue is that these survey data con- labor market data from the Survey of tain relatively few observations. For example, Income and Program Participation in the Katz and Meyer study, which finds a (SIPP), he finds that higher UI benefits sharp spike in re-employment, there are only 26 12 For example, more workers may be liquidity observations at the spike. are associated with much lower job- constrained during economic downturns.

26 Q4 2010 Business Review www.philadelphiafed.org the low-productivity job. Furthermore, However, it is possible to associate the the economics profession has not if the worker rejects the offer, he also model’s implications with a real-world accumulated enough research that tells faces the risk of not getting an offer at situation in which more generous UI us how large the extensions should be all in the near future. This acceptance benefits give workers some time to look during economic downturns. decision is based on balancing between for a high-paying job, which in turn Also, one important issue that the two competing effects. In this has some impact on firms’ decisions to has not been studied very much in situation, the higher benefit level shifts create such jobs. the literature on UI is the interaction the balance toward looking for a high- between the benefit level and human productivity job, turning down offers SUMMARY AND MISSING PIECE capital or skill depreciation. There is of low-productivity jobs. In this article, I have reviewed a long-standing empirical literature When the benefit level is raised, some of the key findings on the on earnings losses; those who are out firms have a higher incentive to create economic effects of UI benefits. It has of work for a long time tend to lose high-productivity jobs, knowing that sometimes been argued that extending human capital and thus earn much less workers are more likely to turn down UI benefits causes adverse incentives than they did pre-unemployment, even low-paying job offers (the job-creation for searching for a job. However, if one is lucky enough to find a job. effect). Through numerical exercises reporting effects complicate the Longer eligibility of UI may exacerbate using this model, Acemoglu and interpretation that moral hazard effects this effect. The academic research Shimer show that higher UI benefits predominantly account for the spike examining this interaction would raise the unemployment rate mainly in the exit rate from unemployment. also be valuable for policymakers and through the moral hazard effect, but Furthermore, the arguments based economists. BR aggregate output and welfare increase on the liquidity and job-creation as a result of the positive feedback effects justify the positive relationship between workers’ willingness to look between the level of UI benefits and for high-productivity jobs and the the duration of unemployment as creation of high-productivity jobs.13 socially desirable. 13Acemoglu and Shimer’s model does not feature the liquidity effect, and thus the higher benefit They do not assess the empirical The expansions of UI benefits causes workers to devote less effort to job significance of this job-creation effect. during the most recent recession may search, raising the unemployment rate. How- We thus do not know how significant be supported by the latter argument ever, its negative effect on output and welfare is more than offset by the positive job-creation the job-creation effect is in reality. at least qualitatively. Unfortunately, effect.

REFERENCES

Acemoglu, Daron, and Robert Shimer. Fallick, Bruce. “Unemployment Insurance Meyer, Bruce. “Unemployment Insurance “Productivity Gains from Unemployment and the Rate of Re-Employment of and Unemployment Spells,” Econometrica, Insurance,” European Economic Review, Displaced Workers," Review of Economics 58:4 (1990), pp. 757-82. 44:7 (2000), pp. 1195-1224. and Statistics, 73 (1991), pp. 228-35. Moffitt, Robert. “Unemployment Card, David, Raj Chetty, and Andrea Gruber, Jonathan. “The Consumption Insurance and the Distribution of Weber. “The Spike at Benefit Exhaustion: Smoothing Benefits of Unemployment Unemployment Spells,” Journal of Leaving the Unemployment System or Insurance,” American Economic Review, Econometrics, 28 (1985), pp. 85-101. Starting a New Job?” American Economic 87:1 (1997), pp. 192-205. Review Papers and Proceedings, 97 (2007), Whittaker, Julie. “Extending pp. 113-18. Katz, Lawrence, and Bruce Meyer. “The Unemployment Compensation Benefits Impact of the Potential Duration of During Recessions,” Congressional Chetty, Raj. “Moral Hazard vs. Liquidity Unemployment Benefits on the Duration Research Service Report RL34340 (2008). Constraint and Optimal Unemployment of Unemployment,” Journal of Public Insurance,” Journal of Political Economy, Economics, 41 (1990), pp. 45-72. 116:2 (2008), pp. 173-234. McCall, John. “Economics of Information and Job Search,” Quarterly Journal of Economics, 84:1 (1970), pp.113-26.

www.philadelphiafed.org Business Review Q4 2010 27