A Primer on Unemployment Insurance

And why its terrible

Unemployment Insurance is a failed program in desperate need of reform. Despite being the primary tool for sustaining workers during periods of unemployment, it has been marked by:

• running out of money

• low benefit amounts that have not kept up with inflation

• short benefit durations

• low recipiency

You can read about these issues in a summary report published by The Century Foundation, which examined its performace during the pandemic.

So how is it that the program that serves as a critical macroeconomic stabilizer during recessions is so bad? If its so important, why isn’t it getting better?

The short answers: taxes and incentives, respectively.

Taxes.

Unemployment Insurance was created in 1935 by the Social Security Act as a joint state-federal program. Like its counterpart also created by the Social Security Act, the Old Age Insurance that we commonly refer to as Social Security, Unemployment Insurance is a social insurance program. Workers earn base eligibility by working in jobs that are “covered” by a payroll tax. They then apply for a benefit when an insured event occurs.

But unlike Old Age Insurance, Unemployment Insurance’s payroll tax has three distinct design flaws that doom the program for failure.

First flaw, who pays the tax. In Old Age Insurance, the payroll tax is shared by the workers and the employer. For Unemployment Insurance, its paid by the employer only. This has a lot of design implications:

• Maintenance. An employer-only tax separates out the people who finance the program (employers) from the people who benefit from the program (workers). This skews the incentives for keeping the program in good financial health with adequate benefits.

• Coverage. Only workers who have an employer are covered by unemployment. Contractors and the self-employed are completely excluded. Those workers have to pay Social Security tax (employer and worker side) when they file their federal income taxes, basically requiring them to opt into Social Security. Not so for unemployment.

• Knowledge. Without seeing the deduction for taxes on their paycheck, many workers do not realize that they are gaining coverage for unemployment, or that their employer has been paying taxes on their behalf.

Knowledge matters. A recent survey of the unemployed published by the Bureau of Labor Statistics (read gold standard for quality) found that of those who did not apply for benefits, more than half said it was because they thought they were ineligible. Workers coming into unemployment from a job with a union were likely to receive unemployment, because their union told them they can and should apply. And the state with the highest rate of recipiency—where the most unemployed workers receive unemployment—is New Jersey, which has a worker tax.

Second flaw, who sets the tax. Old Age Insurance is a federal program, with tax rates and benefits determined by Congress. Unemployment Insurance is actually 53 separate programs, run by the states, District of Columbia, and two territories. Each state sets the tax rate, tax base, eligibility test, and benefit amount. Each state’s tax collections are deposited into a state-specific trust fund.

This state variation is how workers, who have worked the same amount of time at the same job with the same pay, can end up with drastically different benefit amounts. That raises equity issues, not least because the states that pay the least in benefits tend to have the most Black people. It also sets off a race-to-the-bottom among states trying to attract employers as part of being “business friendly.” Higher payroll taxes means higher costs of business.

Third flaw, and most important, what increases the tax. The Unemployment Insurance payroll tax is “experience rated.” The more workers a firm sends into the program, the higher tax it pays on all remaining and future employees. This creates a terrible incentive for employers: if they lay off workers, their tax increases *IF* those workers successfully file an unemployment claim; if the worker doesn’t get a benefit, they don’t get a tax penalty.

This could be, at the least, why employers may not be rushing to help their laid off workers file for benefits. It could also be why they use other companies, like Equifax Workforce Solutions, to handle all of their unemployment claims. Through delaying tactics, which reduce the number of workers who claim or the length of their claim, Equifax can prevent employers from getting a high experience rating.

(If that sounds gross, you should read this 2004 article from the New York Times about Talx, which was bought by Equifax in 2007 and renamed Equifax Workforce Solutions. It wasn’t purchased for $1.4 billion because it costs employers money on their unemployment claims, and the only way to save money on unemployment taxes is to reduce benefits claimed.)

There’s a state issue here too. If lots of workers file for benefits, and the state’s trust fund runs out of money, the state has to borrow from the federal government to finance benefits, and then pay back the loan. The mechanism to do that is for many states is an automatic tax increase that stays high until the trust fund is solvent. To keep taxes low, states need to keep the trust fund from being drawn down. So employers and states (or at least, states that don’t want higher payroll taxes on businesses) both have the incentive to keep claims low.

The financing scheme—employer only, varying by state, increasing with claims— was, and is, the root of most of the program's problems. But that wasn't immediately obvious when it paid out its first benefits in 1938.

The tail end of the Depression was an incredibly fortuitous time to start an unemployment program, because the years leading up to and including World War II were characterized by record job creation, spiking wage growth, and little unemployment. States' unemployment trust funds swelled as they paid out virtually no benefits. The next 20 years—through the 1960s—also saw strong economic growth while the program expanded to cover more workers (and thereby collect more in taxes) beyond its original base of covered workers in manufacturing.

A windfall masks weaknesses of all kinds, including financing design. This leads to the second part, about why the program is not getting better.

Incentives.

Specifically, incentives to reform.

The cracks in the windfall (each state’s tax collections are held in a state trust fund for benefits, so it technically should be windfalls) started to show in the 1973 recession, the deepest and longest Unemployment Insurance had faced. Congress could see what was going wrong. The fixed duration of benefits meant that many workers would be cut off even as the recession dragged on and unemployment stayed high. As they had in the two recessions before 1973, Congress stepped in to extend the benefit period. But there was a larger concern that UI's finances were weak and that it wasn't keeping up with changes in the economy. So Congress created the National Commission on Unemployment Compensation (the NCUC) in 1976 to design comprehensive reform.

And here’s where the incentives start to go awry.

The Commission's final report came out in 1980. None of its recommended measures were adopted. Ten years later, the 1990 recession hit, and the story repeated: UI performed poorly, Congress stepped in to help extend the benefit length, but determined that the program needed reform. The Government Accountability Office raised the alarm that Unemployment Insurance’s financing issues, and states’ low-tax, low-benefit strategy was compromising the program's ability to help workers.

In the aftermath of the recession, Congress created the Advisory Commission on Unemployment Compensation (ACUC). The ACUC released its final report in 1996, echoing the issues raised in 1980 and making sweeping recommendations. Again, none were adopted.

When the economy went into recession again in 2001, Congress again expanded benefit duration. No committee this time, no rececommendations for reform ignored, but no reform either.

(Though Congress did act to restrict “SUTA dumping” in 2004 through the SUTA Dumping Prevention Act. SUTA is short for state unemployment tax. Employers with a high SUTA rate would re-incorporate or use a series of shell transactions to reset their tax rate to lower levels. Experience rating matters.)

And then came the Great Recession. I’ve said it before and I’ll say it whenever appropriate: this recession earned its title. The Great Recession entailed an incredibly lengthy period of job loss followed by a brutally slow recovery. Unemployment benefits were again lengthened by Congress. But the real Unemployment Insurance story in this recession was the trust funds. State after state became deeply insolvent, taking out loans that they had to pay back just as their weak economies were starting to recover. They made payments on these loans for a decade or more.

In the wake of the Great Recession, states had two (or three options). They could shore up trust funds to get their programs to a good fiscal place, as the Government Accountability Office implored them to do. Or, they could cut benefits and shorten durations to avoid having to pay out money. States like North Carolina and Florida chose to cut; ten in total offered less than the usual 26 weeks of benefits.

(They could also do nothing, which many did.)

When the COVID-19 pandemic arrived in March of 2020, throwing tens of millions out of work, Unemployment Insurance reform was five decades overdue. In every recession since 1973, Congress propped up UI, mostly through extending the duration of benefits' availability. But Congress had to do that and much, much more with the CARES Act. It didn’t just extend duration, it increased benefit amounts and created new programs for workers who had been previously excluded from Unemployment Insurance.

And it did for all states, whether their trust fund was healthy or nearly insolvent, whether their benefits were low or high, whether their tax rate was high or low. Every state Unemployment Insurance program got the CARES bailout.

Because Unemployment Insurance is a state program, governed by state decisions, funded through state taxes, paid out in state benefits—until there’s a recession, when its about workers in trouble and the economy writ large, and then it becomes Congress’s problem.

The finances guarantee states will never reform Unemployment Insurance to the level the national economy requires. Since Congress steps in during recessions, in larger and larger ways, they don’t really have to.

What should unemployment insurance look like? What reforms does it need? Next time.