Analyzing Great Recession-Era Unemployment Benefits in Five Steps

From COVID-19 stimulus checks known virally as “stimmy” to expanded unemployment benefits, most Americans know that the government steps in with emergency programs to cushion the blow during economic crises. However, many of these programs involve frequently changing guidelines and a lack of data transparency, obscuring how they function in practice.

This step-by-step analysis clarifies how one such program, the Emergency Unemployment Compensation (EUC08) program, actually worked, including how it produced racially disparate outcomes. Understanding EUC08 provides insight into many other programs, as it is representative of how several emergency programs activate benefits.

EUC08 provided additional unemployment benefits to those who had exhausted their state-provided coverage during the Great Recession, when unemployment was rampant. Through this federal program, individuals received the same weekly payment as with traditional state unemployment benefits, just for additional weeks. During the program, $230B in federal benefits were distributed, and at certain times during the recovery, EUC08 provided the majority of all unemployment benefits received across the country.

Combining statistics employed by EUC08 with data² measuring the policy’s impact tells the story of when and why different levels of expanded benefits flowed to workers. In this article, I first show how a state’s unemployment rate triggered different levels of benefits. I then use data from Texas as an example to explore EUC08’s efficacy. Finally, I consider how the program’s use of unemployment triggers furthered racially disparate outcomes.

Total Unemployment Rates

EUC08 uses the Total Unemployment Rate (TUR)³ of a state, which is an average of a state’s topline unemployment rate from the preceding three months. Figure 1 shows each state’s TUR in the first and last month that EUC08 was active, while Figure 2 is a searchable database showing the TUR trendline in each state. Critically, TURs increased dramatically during the Great Recession and stayed elevated for several months past the crisis’ initial shock.

Tiers

EUC08 split benefits into four tiers, with higher TURs unlocking higher tiers with additional weeks of unemployment benefits to compensate for the higher unemployment rate. The benefits available under each tier changed frequently, as Congress updated the program 11 times. Figure 3 shows the number of additional weeks available under each tier of the program over time.

Importantly, the weeks available accumulate. Workers accessed benefits in their state’s current tier after receiving benefits in all lower tiers, and all workers had to exhaust existing benefits before qualifying again. For example, at one point in the program, a worker in a tier 4 state would have received:

14 weeks + (tier 1)
14 weeks + (tier 2)
9 weeks + (tier 3)
10 weeks + (tier 4)
------------------------
47 weeks total

Cutoffs

TURs and tiers come together in the program to form cutoffs – the TUR at which each tier was triggered. These cutoffs changed over time, and Figure 4 shows the cutoffs for each tier. Most notably, from 2010 until the middle of 2012, all states activated Tier 2 benefits, as the Tier 2 cutoff during this time was 0%.

Applying any state’s TUR on top of Figure 4 allows us to determine a state’s tier at any point during the program. Figure 5 applies Texas’ TUR. We can see that Texas triggered Tier 2 in April 2009, Tier 3 in December 2009, and Tier 4 in December 2011, before declining into Tier 2 by the end.

Weeks Claimed and Paid

In order to see the scope of benefits actually accessed by individuals, we can consider the number of weeks of benefits claimed and paid out in a given month, as shown for Texas in Figure 6. We can see that workers claimed hundreds of thousands of weeks of benefits in Texas each month. The number of weeks paid out in a given month exceeds the number of weeks claimed, as benefits claimed in a given month are paid weekly, continuing into subsequent months.

In theory, EUC08’s tier system increased the benefits available as the difficulty of finding a job increased, measured by a state’s unemployment rate. Figure 7 plots the average weeks claimed per claimant on top of the TUR in Texas. In general, as the TUR increased, the average number of weeks claimed increased as well.

Comparison to Black Unemployment Rate

While EUC08 employed TURs to determine available benefits, the program’s failure to include metrics accounting for the experience of marginalized groups limited the program’s efficacy. Notably, EUC08’s use of TUR triggers institutionalized inferior outcomes for Black workers, as I consider in my recent report. The data used to construct unemployment rates at the national and state levels undercounts Black workers significantly, and the program does not account for the fact that the Black unemployment rate is almost always nearly double the white and topline rates. For these reasons, EUC08 required relatively higher Black unemployment, compared to White unemployment, to unlock additional benefits. Similarly, it reduced benefits when the topline rate declined, even though the Black unemployment rate often remained elevated.

Figure 8 plots Texas’ quarterly Black Unemployment rate on top of Texas’ TUR and the EUC08 cutoffs. In absolute terms, Texas’ Black unemployment rate remained above the Tier 4 cutoff for the entire program. Further, at the start of the recession, Black unemployment increased much more rapidly than Texas’ TUR. The TUR also remained stagnant at several points when the Black unemployment rate increased. Most notably, at the end of 2013, a declining TUR moved Texas back into Tier 2, even though the Black unemployment rate spiked during these months. On the whole, Texas’ TUR is unrepresentative of the experience faced by Black workers during the Great Recession, and if EUC08 considered the relevant Black unemployment rate when triggering benefits, more coverage would have been provided to those still in need.

This analysis of EUC08 explores how the program's design impacted the benefits distributed in practice. Many past emergency programs structured benefits as EUC08 does, and future programs will likely do the same. Before deploying the next iteration of emergency unemployment benefits, policymakers must consider the critical nature of the measurements employed by the program and the data measuring its effect. Rather than using TURs, policymakers should use metrics accounting for those suffering most acutely from the crisis. Further, officials implementing the program should publish data in real time to allow policymakers and the public to understand the program’s efficacy and consider any deviations from the intended impact.

---

1: All views presented in this post are my own and do not represent the views of my past or present employers.

2: All data is seasonally adjusted where available.

3: I aggregated the TUR values presented by web-scraping weekly Department of Labor reports for each state given the absence of public TUR data sets.

4: The Insured Unemployment Rate (IUR) could also trigger Tiers 3 and 4 of EUC08, but the IUR cutoffs were too high to trigger tiers in practice, explaining their absence from this analysis.

5: The quarterly Black unemployment rate depicted in Figure 8 uses the Economic Policy Institute’s original methodology for disaggregating state-level unemployment indicators. This approach considers the state unemployment rate, national demographic data (CPS 6-month ratios), and state-level employment figures. Other approaches, including the Federal Reserve Bank of Dallas’ calculations using 3-month running averages from the CPS and EPI’s new approach emphasizing variance rather than a fixed sample size yield slightly different results.