How many people miss out on unemployment insurance?
Among U.S. benefits programs, researchers have spent a lot of time
trying to understand why eligible recipients don't actually take up the
benefits they are entitled to.
Low take-up is severe
for many programs:
26 percent
of potential recipients took up unemployment insurance (UI) in 2022.
This is not a new fact. Since the early 2000s, recipiency rates, what
they call who receives benefits of those who are unemployed, has not
changed all that much.
Why is this? Research has provided a lot of answers to plausible reasons
why individuals don't take up benefits - people may not be aware they
are eligible, applications are hard to complete, there's stigma against
taking government handouts, risk of paying back benefits if you
mistakenly or purposefully claimed benefits you weren't entitled to can
be severe. All of this and more contributes. But even just figuring out
who is entitled, granted as a right by the law, is a complicated
question.
Some history on these programs can be helpful. Take unemployment
insurance. Unemployment insurance was established
by the 1935 Social Security Act. Although UI is authorized by the federal government, each state
administers a separate UI program through a state workforce agency.
Unemployment insurance was originally established to partially replace
income for people who lose their jobs due to changes in the economy, not
their own fault. Over the past century, it has grown and changed. Some
core features have stayed the same though: the program is funded by
taxes on employers and paid for by the federal government; the United
States Department of Labor (USDOL) Employment and Training
Administration (ETA) oversees these programs and provides administrative
funding to states to administer it. In each state, individual recipients
of UI must meet eligibility criteria defined by each state workforce
agency. All recipients (i) must have earned a sufficient income from an
eligible employer in a specific time period and (ii) have become
involuntarily unemployed through no fault of their own. For example,
Nevada's monetary eligibility requirements
are described as "1.5 [times the] high-quarter wages in the base period
and $400 in the high quarter [or] of wages in 3 of the 4 quarters in the
base period and $400 in the high quarter [of the base year]."
Since each persons' eligibility is based on when they earned their
income over the past however long and the individual circumstances they
lost their job, it's hard to know if an unemployed person is actually
eligible. There's no centralized database with a standardized measure of
if a job separation is "for no fault of their own." There's also no
centralized database of the eligible unemployed. The US uses
survey-derived estimates
to measure unemployment. With unemployment insurance, the department of
labor assumes that about half of the unemployed are probably eligible.
This measure
undercounts
some people who may be eligible but only partially employed and
overcounts
people who might not be eligible.
Why don't we measure recipiency well?
Why do we not collect this information systematically? There's many
answers to that questions. One area is really important. Getting
unemployment insurance is actually a two-stage process, where applicants
have to file a claim and then claim the benefit each week. Here's what
applying for unemployment insurance might look like for someone who
recently got laid off. Let's call her Andrea:
Applications to UI are generally filed through an online application
these days. Andrea might would start by submitting an initial claim to
be processed on the web portal. She'd then not hear anything for a few
days or weeks while the state workforce agency determined monetary and
nonmonetary eligibility under state law. This often involves the agency
first verifying Andrea's income with state and federal tax information
and then verifying the context in which Andrea lost her job with her
last company.
After, she'd receive instructions to file her weekly (or biweekly,
depending on where she lives) certifications to validate that she was
eligible each week to receive payment. These certifications often
include attesting to eligibility criteria that recipients are able and
available to accept work, that they have accomplished state-specific
work requirements, and they do not have too much income during the
benefit year or certification period. All of these criteria have
specific legal definitions that may not be clear like being "able and
available" for work. Similar to the initial claim, determinations that
arise from questions of eligibility can lead to Andrea not receiving her
benefits for a weekly period.
While Andrea has started filing her weekly certifications, state
workforce agency staff are making a formal determination about
eligibility based on the initial claim information. If they don't have
enough information they're tasked with collecting facts ("fact-finding")
from relevant parties, like Andrea's old company. This determination
includes adjudicating between differing claims from individuals and
employers based on state criteria. Applicants, or claimants, can then
appeal any determination on the initial claim.
Once an initial claim is determined eligible, Andrea would receive two
letters, probably by mail, that said first she was "monetarily eligible"
and gave her the amount of unemployment insurance she would receive each
week. The second letter would tell her she was "non-monetarily eligible"
to get benefits because her layoff was due to no fault of her own.
Then, she'd start to get her benefits for each week she submitted a
valid weekly certification.
Claimants' recipiency changes on a week-by-week basis depending on how
long people are unemployed, how often they complete their weekly
certifications, and state law.
Unemployment insurance policy is often complex in different ways between
states. It's also different complexity in each state. Each state program
determines the specific meanings of those eligibility criteria. Similar
to eligibility criteria, benefit characteristics vary between states.
Each state writes, maintains, and determines what is in that initial
claim application. Each state has
different requirements
for applicants to be eligible and a different process to apply.
This leads to drastically different recipiency outcomes.
In the image above, each gray line is its own state. Although the
federal average hasn't changed all that much in the past few decades,
there's been a lot of changes in different states, some increasing, and
some decreasing.
Each state is making their own decisions
Benefits agencies have control over how policies are implemented and are
often evaluated on recipiency through federal and state oversight.
Although recent federal memoranda describe a need for benefits programs
to reduce administrative burden, state workforce agencies are often
limited in funding to make changes to programs. Determining what
investments increase social welfare is a challenge for agency
administrators, especially when balanced with alternative focuses on
reducing fraud. Beyond just changing the proportion of people that can
apply, policy and administration lead to drastically different amounts
of people that can apply.
In, 2023, XY states made up XY% of unemployment
claims. It's not hyperbolic to say that these states are the US
unemployment system for the most part.
COVID changed things
As a reminder, unemployment insurance is a mix between federal policy
that changes automatically with the economy, state decisions about their
own las and eligibility, and how programs are administered: if the
website is easy to navigate on your phone, if navigators can help you
with your app, if there's enough staff to answer the phones. All these
moving parts make it easier or harder to get UI.
During the pandemic, the US made a huge change to how unemployment
insurance works. Whether this was a valid or effective decision to help
people in need is up for debate. It does show us at least one thing,
that recipiency isn't fixed. Due to changes in who was eligible and how
to show your eligibility, recipiency skyrocketed - reaching above 100%
of the unemployed population. Some of this is fraud, but that's less
relevant for the point at hand. This is evidence that you can get more
people into the unemployment system.
Although this seems obvious, it's important to recognize: recipiency
isn't fixed around 30 percent of unemployed workers. It's a choice that
states and the federal government has made about how people should take
up benefits.
This is changing. In the past few years, the federal government has
provided guidance to states that they should make it easier to get
unemployment benefits. These memos and letters have focused on what's
called equitable access to the benefits. US DOL has been focused on
requiring that state workforce agencies allow “the consistent and
systematic fair, just, and impartial treatment of all beneficiaries of,
applicants to, and participants in the UI program” (U.S. Department of
Labor, 2023). These mandates aren’t just an extra requirement. A lot of
money was made available after the COVID-19 pandemic for state workforce
agencies to respond to these concerns:
$260 million for projects that improve equitable access
to benefits and
$200 million for changes
following a consultative assessment of state workforce agencies'
capabilities. But, many of these programs have not been rigorously
evaluated to see if they work.
If recipiency can change, what's happening in states with higher
recipiency?
There's a lot of theories about why recipiency varies so much. In recent
years, academic research has looked at how to improve access to
benefits. There's a lot of approaches that states have tried in
administering the program and also a lot of differences in policy. This
patchwork of policy and implementation has led to some outcomes that
look like a lot of American history and politics.
Northern states have tended to make it easier to get unemployment
insurance benefits, and as such have better recipiency rates. These
differences are likely caused by lower take-up and more constrictive
policy. Black potential claimants, in particular,
take up benefits less
than other potential claimants, especially in the South.
But states also vary over time. If we look at the bigger states from
earlier, we can see how often they have a certain recipiency rate for
each month they report it. If a rate is more common, that rate will be a
larger bump in the graph below. If a state's recipiency rate doesn't
change much in the past two decades, it would be a really tall bump. If
the state's changed a lot, it would spread out.
What's happening in New Jersey? Over the past two decades, New Jersey
has had a lot of differences in its recipiency rate. The answer is a
combination of policy and implementation. Alongside some changes to
unemployment insurance law, the New Jersey Office of Innovation worked
with NJ DOL to modernize their system with a focus on making it easier
to use for residents. Since 2022,
NJ DOL has had
a new app, SMS reminders, mobile-accessibility, and plain language
descriptions of the law. The impacts are striking. UI is easier to
access, and you can see it in the recipiency rate.
That gets us back to the original question: How many people miss out on
unemployment insurance? We can't necessarily get that answer and
certainly can't use the recipiency rate to get. One solution would be to
get better information on who isn't getting benefits. A better
recipiency rate would help states make decisions to help their
residents. Turning the lights on in a dark room lets you navigate to
where you want to go.
Even if the recipiency rate is a flawed measure at best, states can
still figure out how to make sure that the people who have paid into
unemployment insurance actually get their benefits. New Jersey has done
it. And New Jersey is an indication that when they do, we can see it in
the recipiency rate.