Unemployment Insurance Expansion Under the CARES Act: Too Much of a Good Thing?

With the coronavirus-induced stay-at-home orders, millions of citizens have stayed home from work. As a result, over 22 million jobless claims have been filed in the United States, breaking previous weekly records by almost ten-fold. In terms of percent presently unemployed, only the Great Depression rivals what we are currently experiencing. Until the spread of COVID-19 is more manageable, state governors are going to feel inclined to continue with the stay-at-home orders. To mitigate the economic fallout, Congress passed the Coronavirus Aid, Relief, and Economic Security Act (CARES Act [H.R. 748]) on March 27. The CARES Act, which entails a whopping $2.3 trillion in federal spending, represents an urgency that was not even felt during the Great Recession. I plan on covering more of the provisions of the CARES Act in the upcoming weeks, but what I would like to cover in further detail today is the unemployment insurance (UI) expansion under the CARES Act. The UI provisions include the following:

• An additional $600 per week in UI benefits for up to the next four months for those who are not working due to the stay-at-home orders
• Federal funding to state UI funds that will provide up to 39 weeks of UI benefits for those who typically are not covered under UI (e.g., self-employed, independent contractors) 
• Federal funding for up to 13 weeks of UI benefits until December 31, 2020 if the state's UI fund is depleted 

The idea behind the UI expansion is to stabilize households economically while people stay at home in order to significantly slow the spread of coronavirus. Plus, it will provide households liquidity to pay bills and buy essentials (e.g., food, medicine). On the one hand, we are in uncharted territory, both with the exogenous nature of the shock to the economy and the magnitude. This is not like the Great Depression or Great Recession when there was something fundamentally wrong with the economy. These are actions the government is taking in the near-term in the hopes that we can contain coronavirus, thereby being the least worst option. On the other hand, there is reason for skepticism that this will end up being worse.

When looking at the Great Recession, multiple Federal Reserve Banks found that unemployment insurance benefits prolonged the recession because of the disincentives to work that UI benefits create (Hagedorn et al., 2013; Mazumder, 2011; Valletta and Kuang, 2010). Another Federal Reserve paper points out that UI benefits also create moral hazard, substantial monitoring costs, and tax distortions (Fabre et al., 2014). And given that a) UI benefits are increasing, and b) only 41 percent of U.S. citizens have enough money in savings to cover a $1,000 emergency, we are seeing the moral hazard piece play out. As a matter of fact, a paper from the National Bureau of Economic Research shows that removing UI benefits after the Great Recession allowed for the creation of 2.1 million jobs  in 2014 (Hagedorn et al., 2015).

UI benefits prolonged high unemployment rates during the Great Recession because receiving UI benefits was preferable to working. If the UI benefits were smaller under the Great Recession and they caused such unemployment, it stands to reason that larger benefits would create larger incentives for unemployment. The higher the replacement rate (i.e., the ratio of the claimant's weekly benefit amount to average weekly wage), the greater the incentive to stay unemployed (e.g., Eugster, 2015).

Prior to the CARES Act, the average weekly UI benefit was $385 a week. Adding $600 a week means an average increase of 155.8 percent (Brookings Institution). The median weekly earnings for a full-time worker in Q4 2019 was $936, which is almost $50 a week less than the UI benefits under the CARES Act. While UI benefits vary from state to state, and while this provision is temporary, it still means that many workers would find it more profitable to stay at home than work.

May 5, 2020 Addendum: As a matter of fact, a Heritage Foundation analysis found that these expanded unemployment benefits will resulting in an extra 13 million people filing unemployment claims than would have otherwise filed absent the $600 a week.

The amount of the UI benefits is not the only concern. There is the matter of the length of the benefits, particularly with the extra $600 a week. The University of Washington's Institute for Health Metrics and Evaluation model for COVID-19, which is the model that the White House bases its projections on, predicts that a) we have already passed the peak, b) numbers will decline considerably, and c) there will be less than 100 COVID deaths a day in June until it peters out by the end of June. Do we really need for these benefits to go through July?

Also, what about long-term unemployment? What if after four months of not working, it takes a few months to find a new job? Long-term unemployment (normally defined as lasting for at least six months) not only makes it more difficult to find a new job, but it means less earnings and lower career potential once a job is found (Van Horn and Zurkin, 2014), emotional turmoil (Basbug and Sharone, 2017), lower health outcomes, less social mobility, and higher crime and violence on a societal level (Nichols et al., 2013). The effects of unemployment, even short-term unemployment, go well beyond the initial unemployment phase.

I also worry about how these lavish UI benefits will affect state UI benefit funds. According to the Right-leaning Tax Foundation, six states (including three large states: California, Texas, and New York) will not have enough funds to get them through the month of May. The Upjohn Institute for Employment Research found that most states have inadequate UI funds (O'Leary, 2020). This research also found that 36 states exhausted their UI funds after the Great Recession and had to borrow to pay benefits (also see Brookings Institute research here).

Unlike the federal government, states cannot run up deficits. As I pointed out in my analysis on balanced budget amendments, 46 states and the District of Columbia require balanced budgets. While states do have rainy day funds, most states do not have large rainy day funds. Given the scope of the pandemic, states will need to use the rainy day fund for more than UI benefits.

If incentivizing unemployment and weak UI funds were not enough to worry, here are some other facets to consider:

• The amendment to add a cap at 100 percent of workers' wages was rejected. Capping the UI benefits would have mitigated the unemployment effects. 
• The "self-certification" provision in Section 2102 neither specifies the necessary conditions of self-certification nor does it provide adequate oversight. All of these factors create strong incentives for individuals to stay unemployed.
• These strong UI benefits do not only create perverse incentives for employees, but also employers. Companies such as Equinox, Macy, and Steelcase have started laying off workers because they figure that their employees fare better because of the lavish amount of the enhanced UI benefits. 
• The replacement rate could be high enough where it creates the risk of incentivizing unaffected or essential workers to become unemployed and remain so (Birinci and See, 2019).
• Depending on how many people use the benefits and for how long, these enhanced UI benefits could cost anywhere between $187.6 billion and $319.9 billion. The Left-leaning Tax Policy Center believes it could end up costing more.

It is one thing to create a financial stopgap during a public health crisis. It has the potential to reduce the duration and magnitude of the economic downturn. Frederich Hayek, who was a renowned economist who defended libertarianism (classical liberalism), argued for a basic social safety net. An argument for UI benefits during these difficult times can be made. Conversely, these provisions under put strain on the federal and state economies and respective UI funds. The way that UI is set up under the CARES Act incentivizes people to be on UI benefits longer than they need to be. Policies such as these that create perverse incentives that have large swathes of the labor market be unemployed longer than otherwise necessary is going to hit this country hard in the upcoming months.


May 19, 2020 Addendum: A working paper from the University of Chicago confirms my concerns (Ganong et al., 2020). According to this paper, the median replacement rate is 134 percent. To quote the paper, "replacement rates over 100% create distributional issues and may hamper efficient labor allocation both now, and especially during an eventual recovery. That is, UI induces tradeoffs between consumption smoothing and moral hazard."

June 5, 2020 Addendum: The Congressional Budget Office (CBO) released a report this week on what extending the UI benefits would look like. One finding is that five out of six recipients would have a replacement rate exceeding 100 percent [p. 4]. It would lower incentives for low-income households to work more than high-income households [p. 6], which would reduce demand for goods and services [p. 7]. Employment would be lower in 2021 than it would be without the extension [p. 7].

January 4, 2021 Addendum: I figured that expanding unemployment insurance would lead to more unemployment. The latest research from the National Bureau of Economic Research confirms that notion (Holzer et al., 2021). Since 18 states left the program early, the researchers were able to present difference-in-difference estimation and event study estimates to compare. States that continued to participate in the additional unemployment benefit programs (i.e., FPUC, PUA) had 0.8 percent higher unemployment in July and 0.7 percent higher in August. Additionally, the states that terminated the programs early saw the flow of unemployed workers into employment increased by about two-thirds. Even better, there is no evidence to show that expanded unemployment benefits had an improvement on one's welfare.