A study sponsored by the National Bureau of Economic Research claims that the elimination of the federal unemployment benefit program in 2014 resulted in 1.8 million new jobs that year, suggesting that offering unemployment benefits can slow job growth.
Not all economists agree with the methods and findings of economists Marcus Hagedorn, Iourii Manovskii, and Kurt Mitman's working paper released in January. But their study does shine a different light on the purposes and consequences of government economic intervention.
"The purpose of this study was simply to see what kind of effect unemployment benefits had on jobs," said Mitman, an economist at Stockholm University's Institute for International Economic Studies. "There is a benefit to having unemployment benefits for those without a job and that shouldnt be ignored. Its a trade-off between how much insurance we can provide and how many jobs we want to create."
The study
In 2008, Congress had approved funding to extend state unemployment benefits during the throes of the recession. Democrats sought to reauthorize the funding for 2014, arguing that it would not just assist the unemployed but encourage companies to invest in production to hire more people.
But Republicans countered that the economy was strong enough to discontinue the jobless benefits extension and refused to compromise, resulting in the program going unfunded in the 2014 budget.
The program gave financial assistance to applicants who had been jobless long enough to run out of state-level unemployment benefits. According to the study, the average period of unemployment benefit coverage dropped from 53 weeks of coverage to 25 weeks after the federal benefits program was cut.
While the unemployment benefit cuts had federal and state workforce administrators panicked as to how they would extend assistance for millions of unemployed Americans, it created a perfect environment for Mitman and his colleagues to learn more about the role of unemployment insurance.
Ive been interested for a long time in studying unemployment insurance in the United States, said Mitman. At the end of 2013, because of the fact that the program was not reauthorized, it kind of created this natural experiment because states at that time had very different levels of weeks. Some states had zero additional weeks all the way to 37 (weeks of unemployment benefits.)
The responsibility to continue the extended federal benefits fell entirely on the states. Some state unemployment benefit programs offered more assistance than others, allowing the economists to measure what sudden changes in unemployment benefits meant to job seekers.
Using the Local Area Unemployment Statistics dataset that reports county-level information, the researchers compared similar economies that offered different unemployment benefits.
For example, although the entire state of California has a very different economy from that of Oregon, California's Alturas County has a similar economy to neighboring Klamath County, Oregon. In other words, comparing bordering counties in different states removes factors that would otherwise skew their findings.
If the study properly accounted for all other contributing factors, then Republicans were correct to cut federal unemployment benefits from the budget. The U.S. Bureau of Labor Statistics indicates that more people continued to find jobs after the benefit cuts. From January to December, unemployment dropped from 8 percent to 5.6 percent by the end of 2014.
According to the study, the effect of cutting unemployment benefits is so large that it can account for almost the entire remarkable employment growth experienced by the U.S. in 2014. The study attributed 1.8 million new jobs during the year of 2014 to not funding the federal unemployment benefits.
"Average employment growth was about 25 percent higher in 2014 than in the best of several preceding years," says the study. "The employment-to-population ratio rose. The unemployment rate declined sharply. In contrast to typical predictions, the labor force participation rate suddenly halted its steady secular decline. The number of job vacancies that employers were trying to fill increased sharply."
Disagreements
But not all economists found the study's conclusion valid.
Jesse Rothstein, director of the Institute for Research on Labor and Employment and an associate professor of public policy and economics at the University of California Berkley, said the conclusion itself calls the validity of these findings into question.
The results imply that if we had just left unemployment benefits alone in 2013, then job growth would have collapsed in 2014, said Rothstein. I dont have any reason to believe thats true, but youd have to believe this in order to believe the rest of the effect of unemployment insurance on job growth.
The economy gained almost 2.2 million jobs in 2013 and almost 3 million jobs in 2014. According to the NBER working paper, 1.2 million of those 2014 jobs were gained for reasons other than the federal unemployment program cuts.
According to Rothstein, there was nothing remarkable about 2014s job growth compared to the job growth in previous years as the economy has slowly recovered since 2008. Because of this, he says, there is no reason to assume that job growth can be almost entirely attributed to unemployment benefit cuts.
The Center for Economic Policy and Researchs co-director, Dean Baker, performed a separate experiment using the same method as the study but using a different data set.
Baker used the BLS Current Employment Statistics survey because he said the size of the survey is a better representation of the country than the LAUS dataset, which the study used.
According to Baker's findings, there was a slight increase in unemployment in states with lower unemployment benefits than those with higher benefits. But these differences are not as large as the differences that the study has claimed.
My conclusion is that the earlier literature is correct that it (unemployment benefits) really doesnt have that much impact on unemployment, said Baker. Unemployment benefits can keep people looking for work but it doesnt keep people from getting jobs.
According to Baker, the general consensus is that unemployment benefits have a small impact on job growth but not nearly as pronounced as the NBER working paper implies.
Mitman disagreed with Bakers analysis, saying the topic has not been studied in-depth enough to claim a consensus.
Our study is the first to measure the impact of unemployment benefits on employment, said Mitman in an email. No one has done it before. I don't know exactly what Baker is basing his reasoning on. My conjecture is that he is misinterpreting research on the search effort of people receiving benefits and mistakenly trying to extrapolate that to an effect on employment.
Mitman also warned that people shouldnt take their one study and assume that cutting benefits is always the preferable public policy decision.
Personally I think that the recession benefits went on for too long based on the response that we saw, said Mitman. But if you think back in 2008 and 2009, there were probably a lot of people who depended on the benefits to make it through. I think people should just be cautious about jumping to a policy recommendation because its really a study about measurement and not about recommendation."
Not all economists agree with the methods and findings of economists Marcus Hagedorn, Iourii Manovskii, and Kurt Mitman's working paper released in January. But their study does shine a different light on the purposes and consequences of government economic intervention.
"The purpose of this study was simply to see what kind of effect unemployment benefits had on jobs," said Mitman, an economist at Stockholm University's Institute for International Economic Studies. "There is a benefit to having unemployment benefits for those without a job and that shouldnt be ignored. Its a trade-off between how much insurance we can provide and how many jobs we want to create."
The study
In 2008, Congress had approved funding to extend state unemployment benefits during the throes of the recession. Democrats sought to reauthorize the funding for 2014, arguing that it would not just assist the unemployed but encourage companies to invest in production to hire more people.
But Republicans countered that the economy was strong enough to discontinue the jobless benefits extension and refused to compromise, resulting in the program going unfunded in the 2014 budget.
The program gave financial assistance to applicants who had been jobless long enough to run out of state-level unemployment benefits. According to the study, the average period of unemployment benefit coverage dropped from 53 weeks of coverage to 25 weeks after the federal benefits program was cut.
While the unemployment benefit cuts had federal and state workforce administrators panicked as to how they would extend assistance for millions of unemployed Americans, it created a perfect environment for Mitman and his colleagues to learn more about the role of unemployment insurance.
Ive been interested for a long time in studying unemployment insurance in the United States, said Mitman. At the end of 2013, because of the fact that the program was not reauthorized, it kind of created this natural experiment because states at that time had very different levels of weeks. Some states had zero additional weeks all the way to 37 (weeks of unemployment benefits.)
The responsibility to continue the extended federal benefits fell entirely on the states. Some state unemployment benefit programs offered more assistance than others, allowing the economists to measure what sudden changes in unemployment benefits meant to job seekers.
Using the Local Area Unemployment Statistics dataset that reports county-level information, the researchers compared similar economies that offered different unemployment benefits.
For example, although the entire state of California has a very different economy from that of Oregon, California's Alturas County has a similar economy to neighboring Klamath County, Oregon. In other words, comparing bordering counties in different states removes factors that would otherwise skew their findings.
If the study properly accounted for all other contributing factors, then Republicans were correct to cut federal unemployment benefits from the budget. The U.S. Bureau of Labor Statistics indicates that more people continued to find jobs after the benefit cuts. From January to December, unemployment dropped from 8 percent to 5.6 percent by the end of 2014.
According to the study, the effect of cutting unemployment benefits is so large that it can account for almost the entire remarkable employment growth experienced by the U.S. in 2014. The study attributed 1.8 million new jobs during the year of 2014 to not funding the federal unemployment benefits.
"Average employment growth was about 25 percent higher in 2014 than in the best of several preceding years," says the study. "The employment-to-population ratio rose. The unemployment rate declined sharply. In contrast to typical predictions, the labor force participation rate suddenly halted its steady secular decline. The number of job vacancies that employers were trying to fill increased sharply."
Disagreements
But not all economists found the study's conclusion valid.
Jesse Rothstein, director of the Institute for Research on Labor and Employment and an associate professor of public policy and economics at the University of California Berkley, said the conclusion itself calls the validity of these findings into question.
The results imply that if we had just left unemployment benefits alone in 2013, then job growth would have collapsed in 2014, said Rothstein. I dont have any reason to believe thats true, but youd have to believe this in order to believe the rest of the effect of unemployment insurance on job growth.
The economy gained almost 2.2 million jobs in 2013 and almost 3 million jobs in 2014. According to the NBER working paper, 1.2 million of those 2014 jobs were gained for reasons other than the federal unemployment program cuts.
According to Rothstein, there was nothing remarkable about 2014s job growth compared to the job growth in previous years as the economy has slowly recovered since 2008. Because of this, he says, there is no reason to assume that job growth can be almost entirely attributed to unemployment benefit cuts.
The Center for Economic Policy and Researchs co-director, Dean Baker, performed a separate experiment using the same method as the study but using a different data set.
Baker used the BLS Current Employment Statistics survey because he said the size of the survey is a better representation of the country than the LAUS dataset, which the study used.
According to Baker's findings, there was a slight increase in unemployment in states with lower unemployment benefits than those with higher benefits. But these differences are not as large as the differences that the study has claimed.
My conclusion is that the earlier literature is correct that it (unemployment benefits) really doesnt have that much impact on unemployment, said Baker. Unemployment benefits can keep people looking for work but it doesnt keep people from getting jobs.
According to Baker, the general consensus is that unemployment benefits have a small impact on job growth but not nearly as pronounced as the NBER working paper implies.
Mitman disagreed with Bakers analysis, saying the topic has not been studied in-depth enough to claim a consensus.
Our study is the first to measure the impact of unemployment benefits on employment, said Mitman in an email. No one has done it before. I don't know exactly what Baker is basing his reasoning on. My conjecture is that he is misinterpreting research on the search effort of people receiving benefits and mistakenly trying to extrapolate that to an effect on employment.
Mitman also warned that people shouldnt take their one study and assume that cutting benefits is always the preferable public policy decision.
Personally I think that the recession benefits went on for too long based on the response that we saw, said Mitman. But if you think back in 2008 and 2009, there were probably a lot of people who depended on the benefits to make it through. I think people should just be cautious about jumping to a policy recommendation because its really a study about measurement and not about recommendation."