Pandemic Darlings The pandemic economy, in original documents
Home Source documents U.S. unemployment insurance through the Covid-19 crisis

U.S. unemployment insurance through the Covid-19 crisis

Summary

A journal article by Francesco Spadafora of the Bank of Italy and International Monetary Fund, U.S. Unemployment insurance through the Covid-19 crisis, published in the Journal of Government and Economics, Vol. 9 (2023), article 100069. The abstract states that the paper describes the structural characteristics of the Unemployment Insurance system, compares its role in the 2008–09 Great Recession and the 2020 Covid-19 pandemic, and reviews evidence on demand-side and supply-side effects of emergency UI extensions. The introduction states that by August 2021 UI benefits had provided a lifeline for 53 million workers with an injection of more than $800 billion. Table 1 reports UI outlays by program from 2007 to 2022, including total benefits of 539.3 billion dollars in 2020. The article closes with a reference list.

Summary drafted by a model from the document's text below and checked by script against that text before publication. It is a navigation aid, not a reading of what the document proves. Where AI is used

Full text

Spadafora, Francesco



Article
U.S. unemployment insurance through the Covid-19 crisis



Journal of Government and Economics (JGE)


Provided in Cooperation with:
Elsevier




Suggested Citation: Spadafora, Francesco (2023) : U.S. unemployment insurance through the
Covid-19 crisis, Journal of Government and Economics (JGE), ISSN 2667-3193, Elsevier, Amsterdam,
Vol. 9, pp. 1-13,
https://doi.org/10.1016/j.jge.2023.100069




This Version is available at:
https://hdl.handle.net/10419/340709

Standard-Nutzungsbedingungen:                                               Terms of use:

Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen             Documents in EconStor may be saved and copied for your personal
Zwecken und zum Privatgebrauch gespeichert und kopiert werden.              and scholarly purposes.

Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle            You are not to copy documents for public or commercial purposes, to
Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich        exhibit the documents publicly, to make them publicly available on the
machen, vertreiben oder anderweitig nutzen.                                 internet, or to distribute or otherwise use the documents in public.

Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen              If the documents have been made available under an Open Content
(insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten,            Licence (especially Creative Commons Licences), you may exercise
gelten abweichend von diesen Nutzungsbedingungen die in der dort            further usage rights as specified in the indicated licence.
genannten Lizenz gewährten Nutzungsrechte.




                       https://creativecommons.org/licenses/by-nc-nd/4.0/
                                                                         Journal of Government and Economics 9 (2023) 100069


                                                                                  Contents lists available at ScienceDirect


                                                            Journal of Government and Economics
                                          journal homepage: www.journals.elsevier.com/journal-of-government-and-economics




U.S. Unemployment insurance through the Covid-19 crisis
Francesco Spadafora
Bank of Italy and International Monetary Fund, 700 19th St. NW, Washington, DC 20431, USA




A R T I C L E I N F O                                                  A B S T R A C T

JEL Classification:                                                    The Unemployment Insurance (UI) system in the United States has played a decisive lifeline role in effectively
E24                                                                    mitigating the economic and social impact of the Covid-19 pandemic, which prompted the largest expansion of
H7                                                                     UI programs in history, one that is unprecedented in scope, scale and cost. However, the crisis has once again
J64
                                                                       exposed some well-known challenges of the program, perhaps best epitomized by the steady decline since the
J65
                                                                       1980s of both the recipiency rate and the wage replacement rate, which is partly related to the UI system’s
Keywords:
                                                                       funding structure. As a result, on the eve of the pandemic less than one in three unemployed workers used to
Unemployment
Unemployment insurance
                                                                       collect UI benefits – of lower amounts and often for shorter periods of time than before – despite that the average
Job acceptance                                                         duration of unemployment had almost doubled in the aftermath of the Great Recession. The pandemic has also
Covid-19                                                               laid bare additional shortcomings – most notably in the effectiveness of the UI delivery infrastructure to provide
CARES Act                                                              timely and accurate payments – which have prompted further calls for modernizing the UI system. The objective
Countercyclical stabilization                                          of this paper is threefold: first, after a brief description of the main structural characteristics of the UI system, it
                                                                       compares the role played by UI in mitigating the impact of the 2008–09 Great Recession and the 2020 Covid-19
                                                                       pandemic; second, it reviews the empirical evidence from the pandemic on potential demand-side (countercy­
                                                                       clical stabilization) and supply-side (job-search disincentives) effects of emergency extensions of UI programs.
                                                                       Finally, it discusses the main lessons that the expansion of UI programs to respond to the pandemic can offer to
                                                                       inform – and enrich – the debate on whether and how to reform the UI system. The experience with the UI system
                                                                       provides fundamental lessons that can usefully inform the debate on whether and how to introduce – for example
                                                                       in Europe – a common unemployment insurance scheme for macroeconomic stabilization.




1. Introduction                                                                                                     federal law but administered by states under their own laws. The federal
                                                                                                                    government sets a national institutional (legislative and administrative)
    The adequacy of a country’s Unemployment Insurance (UI) system                                                  framework – under the oversight of the U.S. Department of Labor; states
has taken on distinctive importance during the Covid-19 pandemic and                                                (defined as the 50 states plus the District of Columbia, Puerto Rico and
the attendant unprecedented uncertainty surrounding the economic                                                    the U.S. Virgin Islands) establish most of the program’s key parameters,
outlook at both the national and global level. In the United States, the UI                                         namely eligibility criteria, benefit amounts and durations, tax rates and
system has once again played a decisive role in effectively mitigating the                                          taxable wage bases.
massive economic and social impact of the pandemic; the latter                                                          The UI system has three main statutory objectives: 1) providing
prompted the largest expansion of UI programs in history, one that is                                               adequate but temporary income support to workers who become un­
unprecedented in scope, scale and cost. By August 2021, UI benefits had                                             employed “through no fault of their own”; 2) automatic stabilizer for the
provided a financial lifeline for 53 million workers while helping sta­                                             macroeconomy, i.e., a countercyclical stimulus to mitigate recessions; 3)
bilize the economy with an injection of more than $800 billion (U.S.                                                stabilization of employment through the UI “experience rating” system,
Department of Labor-DOL, 2021c). As a result, UI was the most impor­                                                which aims at discouraging layoffs by employers.
tant single element of the fiscal response to the pandemic (Boesch et al.,                                              Against this background, even before the pandemic there had long
2021)                                                                                                               been ample consensus among many scholars and policymakers alike that
    The UI system was first introduced as part of the Social Security Act,                                          the UI system had not been adapted and modernized – notably at the
signed into law by President Roosevelt on August 14, 1935. UI is a form                                             federal level – to address the changes over the past few decades in the
of social insurance established as a federal-state partnership, based upon                                          structure of the U.S. economy and the characteristics of the labor



     E-mail address: fspadafora@imf.org.

https://doi.org/10.1016/j.jge.2023.100069
Received 21 November 2022; Received in revised form 11 April 2023; Accepted 25 April 2023
Available online 26 April 2023
2667-3193/© 2023 The Author(s). Published by Elsevier B.V. on behalf of The Academic Center for Chinese Economic Practice and Thinking, Tsinghua University and the Society for the Analysis of Government and Economics. This is
an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
F. Spadafora                                                                                                        Journal of Government and Economics 9 (2023) 100069


market. Ultimately, these changes have resulted in more and more un­                 tax rates), which provides most of the funding to the UI programs. As a
employed workers – especially self-employed and part-timers – who                    result, UI programs vary widely across states in terms of access,
often fail to qualify for UI benefits; a key reason for this unwarranted             coverage, level of benefits and duration.
outcome is that the states’ UI programs frequently set strict eligibility                The federal government pays only for the administrative costs of UI;
criteria based on employment history – e.g., minimum requirements in                 however, during recessions it has historically fully funded the emer­
terms of work time and wage earnings.                                                gency programs implemented to mitigate the impact on workers who
    The current UI system features a distinctive duality.                            lose their jobs
    On the one side, the state-based regular UI system – with its                        A key feature of the UI system is that it includes three main levels of
“patchwork” of 53 different state programs and rules – is often faulted              benefits,1 of both permanent and temporary (emergency) nature: 1)
for falling short of providing in normal times adequate protection and               regular state UI benefits; 2) federal-state Extended Benefits (EB); 3)
income support to unemployed workers. As a result, well before the                   emergency federal benefits.
Covid-19 pandemic, many observers believed that the ability of the UI
system to achieve its statutory objectives had been reduced: this is                 2.1. The state’s regular UI program
perhaps best epitomized by the fact that in 2019 – on the eve of the
pandemic – less than one in three unemployed workers were receiving                      Almost all wage and salary workers are covered by the state’s regular
UI benefits. Coverage and adequacy of benefits – in terms of level and               UI program, which provides partial and temporary financial assistance to
duration – as well as sufficiency of funding have long been identified as            unemployed workers – replacing about half of their previous wages, up
the overarching (and intertwined) issues of the regular UI system.                   to a maximum benefit amount – while they actively look for another job.
    On the other side, as demonstrated in unprecedented ways during the              On the eve of the Covid-19 pandemic in February 2020, average weekly
Covid-19 pandemic, the federal-based emergency UI programs (financed                 benefits were about $385 nationwide, but ranged from a low of
by the general revenues), which in recessionary times complement the                 $215–220 in Louisiana and Mississippi to $550 in Massachusetts2.
state-based UI programs, are usually remarkably effective. However,                      These differences reflect one of the most distinctive features of the UI
because most of these emergency programs need to be activated by                     system, namely the high degree of discretion given to states to set key
discretionary legislative action, they are exposed to lengthy political              requirements for UI benefits such as their level, duration and eligibility
negotiations and may thus suffer from implementation lags, which                     criteria. Despite the substantial variation of these parameters across
might also diminish their capabilities as (automatic) macroeconomic                  states, commonalities do exist. Workers usually qualify if they: i) are
stabilization mechanisms. Besides, there are concerns that expanding UI              unemployed “through no fault of their own”, as determined under each
benefits may generate job-search disincentives and moral hazard effects              state law; ii) meet monetary requirements, i.e., state-specific minimum
that unduly prolong unemployment.                                                    earnings and employment time during the so-called “base period” before
    The distinctive challenges brough about by the Covid crisis – and the            becoming unemployed or partially unemployed; iii) meet any non­
unprecedented and innovative temporary expansion of the UI programs                  mometary requirements, notably the so-called “work search re­
– have not only rekindled but also enriched the debate on how to reform              quirements”: ability and availability to work, actively seeking work.
the UI system to address its long-standing weaknesses. In this context,                  While from the late 1960s to 2011, the duration of UI benefits in
issues related to the equity of the UI system – primarily in terms of in­            regular UI programs was at least 26 weeks in all states, in the aftermath
come, race, gender and age – are now playing a more prominent role                   of the Great Recession, nine states reduced the maximum number of
that appropriately informs the design of the reform proposals.                       payable weeks (or capped them based upon the state’s unemployment
    With all due differences, the experience with the UI system in the               rate), given exceptionally high amount of UI benefits paid during the
United States provides fundamental lessons that can usefully apprise the
debate on whether and how to introduce in Europe a common unem­
                                                                                     Table 1
ployment insurance scheme for macroeconomic stabilization. Since                     UI Outlays by program ($bn).
European countries have already in place national unemployment in­
                                                                                              Total    State     Extended           EUC08     PEUC     PUA     FPUC
surance programs, the debate mostly revolves around the case for setting
                                                                                                       UI        Benefits
up a European unemployment re-insurance scheme that would top up
                                                                                      2007    32.4     32.4
those national programs while including safeguards against risks of                                              –                  –         –        –       –
                                                                                      2008    51.0     43.1      –                  7.9       –        –       –
redistribution and moral hazard (Spadafora, 2019; O’Leary et al., 2020).              2009    130.3    79.6      6.5                44.2      –        –       –
    The objective of this paper is threefold: first, after a brief description        2010    138.0    58.5      9.3                70.2      –        –       –
of the main structural characteristics of the UI system in the United                 2011    106.0    47.2      10.2               48.6      –        –       –
States, it compares the role played by UI in mitigating the impact of the             2012    82.9     43.1      3.0                36.9      –        –       –
                                                                                      2013    61.9     39.4                         22.6
2008–09 Great Recession and the 2020 Covid-19 pandemic; second, it
                                                                                                                 –                            –        –       –
                                                                                      2014    34.9     34.9      –                  –         –        –       –
reviews the empirical evidence from the pandemic on potential demand-                 2015    32.0     32.0      –                  –         –        –       –
side (countercyclical stabilization) and supply-side (job-search disin­               2016    31.7     31.7      –                  –         –        –       –
centives) effects of emergency extensions of UI programs. Finally, it                 2017    29.9     29.9      –                  –         –        –       –
                                                                                      2018    27.5     27.5
discusses the main lessons that the expansion of UI programs to respond                                          –                  –         –        –       –
                                                                                      2019    27.3     27.3      –                  –         –        –       –
to the pandemic can offer to inform – and enrich – the debate on whether              2020    539.3    143.6     4.2                –         28.6     79.7    283.2
and how to reform the UI system.                                                      2021    317.6    44.3      8.5                –         56.0     50.9    157.9
                                                                                      2022    24.5     24.3      0.3                –         –        –       –
2. Structural characteristics of the UI system                                       Source: author’s calculations on data from U.S. Department of Labor.3

    As a federal-state partnership, the UI system subjects states to some
minimal federal requirements to ensure both that the program provides
a basic level of protection for eligible workers and serves as a macro­
economic stabilizer in economic downturns (Stone and Chen, 2014).                     1
                                                                                        The system also includes a few special state and federal programs, for
    Within these basic safeguards, states pay for the actual benefits                example unemployment compensation for federal employees (UCFE) and for
provided to workers and – critically – enjoy extensive flexibility in                newly discharged veterans (UCX), disaster unemployment assistance (DUA),
setting such key parameters as eligibility criteria, benefit levels and              and short-time compensation (STC).
duration as well as the tax structure (e.g., state taxable wage base and              2
                                                                                        Department of Labor, Monthly Programme and Financial Data.

                                                                                 2
F. Spadafora                                                                                                  Journal of Government and Economics 9 (2023) 100069


long downturn (GAO, 2015; Table 1; Isaacs, 2019). As a result, on the            2.3. Temporary UI programs
eve of the Covid-19 pandemic, in eight states the duration was fewer
than 26 weeks – to as low as 12 weeks in Florida and North Carolina in               The recognition that, because of design flaws, the EB program has
times of low unemployment – with 20 weeks being the new maximum in               been unable to respond rapidly and effectively to recessions, has led
six of such states. However, the trend was reversed in response to the           federal lawmakers to enact temporary federally funded supplementary
Covid-19 pandemic, as in 2020 some states increased the maximum                  programs on an ad hoc basis in every recession since 1958. Before the
number of weeks payable in regular UI programs. Currently, twelve                pandemic-driven recession in 2020, Congress had acted eight times – in
states provide less than 26 weeks and two states (Massachusetts and              1958, 1961, 1971, 1974, 1982, 1991, 2002 and 2008 – to establish
Montana) provide more (30 and 28 weeks) (CBPP, 2023).                            temporary UI programs to provide additional weeks of benefits to in­
    The state regular UI program is almost totally funded by taxes on            dividuals who had exhausted state UI benefits (Isaacs and Whittaker,
employers, something that is hailed as one of the system’s key unique            2014b).
features; only three states (Alaska, New Jersey, and Pennsylvania)                   It is important to note that unemployed workers usually qualify for
collect taxes also from workers (DOL, 2020b, p. 2–4). More specifically,         EB and temporary emergency programs only if they are initially found
the regular UI program is financed primarily by state payroll taxes under        eligible for state UI benefits and are not disqualified during the time they
the State Unemployment Tax Act (SUTA) and, to a lesser extent, by                receive regular benefits. These requirements may undermine the
federal payroll taxes under the Federal Unemployment Tax Act (FUTA).             coverage of UI programs as a source of income support insofar as states
                                                                                 that restrict access to regular UI benefits might make ineligible workers
                                                                                 unable to receive benefits of any type (i.e., not only regular, but also
2.2. The extended benefits (EB) program                                          extended or emergency ones).

    The permanent Extended Benefits (EB) program was enacted by                  2.3.1. The 2008 emergency unemployment compensation program
Congress with the Federal-State Extended Unemployment Compensa­                      The Emergency Unemployment Compensation (EUC08) program
tion Act of 1970 (EUCA) to provide – during recessionary periods and in          was created on June 30, 2008 to provide benefits to individuals who had
high-unemployment states – additional weeks of benefits to workers               exhausted regular state UI benefits. Although the EUC08 program
who have exhausted their regular state UI benefits before they find              shared many features of previous emergency programs, an important
another job.                                                                     innovation was that the EUC08 was initiated much earlier in the
    The EB program is normally co-financed by the federal government             recession than were previous emergency programs, thereby providing it
and the states on an equal 50–50 basis. However, beginning with the              with a potentially greater stabilization role (Nicholson and Needels,
Great Recession, and more recently during the Covid-19 pandemic, the             2011).
federal government stepped in to fully finance the EB program.                       The EUC08 program was fully funded by the federal government and
    A key feature of the EB program is its automatic activation, which           was modified 11 times (Isaacs and Whittaker, 2014a). It stood up in
depends on a few statutory and optional “triggers” (Isaac and Whittaker,         comparison to other emergency UI programs in that it lasted more than 5
2014b; Bauer et al., 2020) that allow states to provide up to 13 or 20           years, until December 28, 2013 – as last extended by the American
weeks of EB when the state insured unemployment rate (IUR) or the total          Taxpayer Relief Act of 2012 – mirroring the unprecedented length of the
unemployment rate (TUR) reach some thresholds highs.                             Great Recession.
    EB triggers have well-known drawbacks, with implications for in­                 At its peak, between November 2009 and September 2012, the four
come support and countercyclical capabilities of UI. A long-standing             “tiers” of EUC08 provided up to 34 weeks of emergency federal benefits
flaw is the so-called “triggering off” problem, i.e., the fact that the          in all states and up to 53 weeks in states with a Total Unemployment
triggers may result in an automatic premature deactivation of the EB             Rate of 8.5–9.0 percent or higher (up to 63 weeks between February
program in times when the unemployment rate remains high and sus­                2012 and May 2012 in states where EB was not active). As a result, in
tained but not increasing, (Isaacs and Whittaker, 2014b).                        states with high unemployment rates and where all four tiers of EUC08
    This problem was magnified in the aftermath of the Great Recession,          benefits were available, eligible unemployed workers could receive a
which featured a sluggish recovery and large numbers of states simul­            total maximum of 99 consecutive weeks of UI: 26 weeks of regular, 53
taneously experiencing a protracted period of very high unemployment,            weeks of EUC08, and 20 weeks of EB (Isaac and Whittaker, 2014a;
notably its long-term component. In fact, long-term unemployment in              Table 1; Needels et al., 2016).
the aftermath of the Great Recession didn’t begin declining until 2012,              The extension by up to 73 weeks of federal benefits payable under
well after the end of the recession. Against this background, Congress           EUC08 and EB combined represented the longest potential duration of
had to step in twice, in 2010 and again in 2013, to keep EB in place, by         benefits in the history of the UI system (Needels et al., 2016), far
giving states the option of temporarily extending from two to three years        exceeding prior extensions of federal benefits: in fact, before the Great
the “lookback” period for the IUR and TUR triggers (Bauer et al., 2020).         Recession no temporary federal program had provided more than 33
    The triggering off problem resurfaced – with somewhat new nuances            weeks of benefits (Weidinger, 2020b, p. 7). The extension reflected the
– in the recent recession caused by the Covid-19 pandemic. Because of            fact that – as noted before – the Great Recession and its aftermath were
the swift and severe nature of the economic shock from the pandemic,             characterized by the longest average unemployment duration since
the TUR and IUR triggers have automatically activated the EB program             World War II, which exceeded 40 weeks for 7 months in 2011 and 2012,
in nearly every state, in some cases less than one month after the passage       almost twice as high as the average unemployment duration at any other
of the CARES Act (Bauer et al., 2020). By June 2020, 52 states had EB on         point since that War (Needels et al., 2016).
(the only exception being South Dakota).
    Despite this unusually responsive triggering on of EB, the require­
ment that the IUR be rising remained problematic throughout the
pandemic crisis, although it was mitigated by the fact that unemploy­
ment declined more rapidly than during the Great Recession. The un­               3
                                                                                     State UI: Monthly Programme and Financial Data https://oui.doleta.gov/
employment rate peaked at a record level of close to 15 percent in April         unemploy/claimssum/5159report.asp; EB: UI Data Summary https://oui.
2020, but by March 2022 had virtually returned to the pre-pandemic               doleta.gov/unemploy/data_summary/DataSum.asp; EUC08: ET Financial Data
level (3.5 percent in February 2020). As a result, by December 27,               Handbook 394 https://oui.doleta.gov/unemploy/hb394/hndbkrpt.asp; PEUC-
2020, EB was active in 24 states (from 52 in June 2020) and by April             PUA-FPUC: Unemployment Insurance Data, CARES Act funding to States
2022, no state had EB on.                                                        https://oui.doleta.gov/unemploy/docs/cares_act_funding_state.html

                                                                             3
F. Spadafora                                                                                                  Journal of Government and Economics 9 (2023) 100069


2.3.2. The pandemic UI programs in the 2020 cares act                            2010, nonfarm payroll employment had fallen by about 8.7 million jobs
    The Coronavirus Aid, Relief and Economic Security Act (CARES Act),           from its pre-recession peak of January 2008, which was regained only in
passed by Congress on March 27, 2020, took some of the boldest actions           the spring of 2014. Because of this sizable and long-lasting adverse
to mitigate the human hardship and economic damage caused by the                 impact on the labor market – which raised concerns of a “jobless” re­
Covid-19 pandemic; it resulted in an unprecedented expansion of UI               covery – the Great Recession prompted the first major expansion of UI
programs, eligibility and level of benefits.                                     programs (Nicholson et al., 2014, p. 188).
    The legislation has three major UI components, which were initially              Total weekly UI recipients – including both regular state and emer­
set to expire on December 31, 2020 (DOL, 2020c):                                 gency federal programs – started rising in late 2007, from around 2.2
                                                                                 million on average in September-October, and accelerated substantially
                                                                                 in the summer of 2008, hitting 7 million by year-end (Fig. 1). UI re­
    1 Additional weeks of federally funded benefits (i.e., an extension of       cipients kept increasing in 2009, remaining above 10 million per week
      the potential benefit duration-PBD) provided by the Pandemic               between November 2009 and May 2010 and reaching a record-high of
      Emergency Unemployment Compensation (PEUC), which was the                  12 million in January 2010.
      current version of the extra weeks of emergency federal benefits               Regular state UI programs borne the brunt of the assistance through
      (like EUC08) that policymakers had enacted in past recessions.             the summer of 2008, covering 3.1 million unemployed per week in the
      Importantly, PEUC was not linked to a state’s unemployment rate.           first three quarters of that year (referred to as “insured unemployment”
      It originally provided 13 weeks of federally funded UI to workers          in the statistics of the Department of Labor and measured by the so-
      who had exhausted their regular state benefits, to be paid ahead of        called “continued UI claims”). The EUC08 program began operating
      any EB (in line with previous emergency programs). The                     since June 2008 and was the key vehicle for the federal government to
      December 2020 Continued Assistance Act (CAA) extended the                  provide supplementary unemployment assistance, covering up to 50
      maximum number of PEUC weeks a worker could receive from 13                percent of total UI recipients in 2010 – 5.4 million weekly at its peak
      to 24. The March 11, 2021 American Rescue Plan Act (ARPA)                  between January and May – and close to 30 percent (1.3 million) when it
      further increased the maximum number of weeks to 53 and                    expired in December 2013.
      extended PEUC’s duration through September 6, 2021.                            It is noticeable that by the time the EUC08 program expired at end-
    2 A federal supplement to UI benefits: the Federal Pandemic Unem­            2013, total weekly UI recipients were still twice as much higher – 4.4
      ployment Compensation (FPUC) provided $600 a week to supple­               million on average in December – than on the eve of the Great Recession;
      ment the weekly payments determined under state UI laws for all            they would go down to below pre-recession levels only by the third
      programs (regular UI, PEUC, EB, and PUA). FPUC initially expired           quarter of 2015.
      on July 31, 2020. Given the lack of political agreement on                     The EB program covered weekly about 1 million unemployed
      extending the FPUC, the “Lost Wages Assistance” program,                   workers at its peak in September-October 2009. Overall, between July
      authorized by then President Trump’s memorandum of August 84,              2008 and September 2013, nearly 24 million unemployed workers
      introduced a temporary $300 weekly supplement for 5–6 weeks                benefitted from the EUC08 and EB programs (Council of Economic
      after July 31. The FPUC applicability gap lasted until December            Advisers, 2013).
      26, when it was re-established by the December 2020 CAA and                    The impact of the Covid-19 pandemic led to an expansion of UI
      extended through March 14, 2021, with a reduced supplement of              programs that is unprecedented in scale, scope and cost, despite that the
      $300. The March 2021 ARPA further extended this $300 supple­               recession lasted only two months and was thus the shortest in history.
      ment through September 6, 2021.                                            Between February and April 2020, nonfarm payroll employment fell by
    3 An expansion of coverage to extend unemployment assistance to              almost 22 million jobs; however, by June 2022, employment had
      the many workers who normally are not eligible for regular state           broadly regained the pre-pandemic peak of February 2020.
      UI benefits: the Pandemic Unemployment Assistance (PUA) applies                On the eve of the pandemic, regular state UI programs – the first line
      primarily to the self-employed, independent contractors (“gig”             of defense – had been covering 2 million unemployed workers per week
      workers), workers seeking part-time work, those who do not have            (Fig. 2). The number of workers receiving state UI benefits (“continued UI
      a long enough work history, and those who must leave work for              claims”) increased from 2 million in the first week of March 2020 to 8.3
      compelling family reasons. The December 2020 CAA increased the             million in the last week of the same month, reaching a record-high of 23
      maximum number of PUA weeks a person could receive from 39 to              million in the second week of May. Because of the distinctively rapid and
      50 and extended the program’s duration through March 14, 2021.             unprecedented nature of the Covid-19 shock, at the very beginning
      The March 2021 ARPA further increased the maximum number of                regular state UI programs had to bear the brunt of the sudden spike in
      weeks to 79 and extended PUA’s duration through September 6,               unemployment that followed the widespread introduction of lockdowns
      2021.                                                                      in spring 2020.
                                                                                     Even more striking is the increase of unemployed workers receiving
   While the PEUC closely resembles the expansion of federal emer­               both state and federal UI support. Following the activation of federal
gency unemployment benefits enacted in previous recessions, the                  emergency programs (EB, PUA, PEUC) by early April 2020, total UI
expansion of UI eligibility and generosity resulting from the PUA and the        claims peaked at over 33 million weekly in June 2020 (from 2 million in
FPUC is noteworthy in scope, scale and cost.                                     the first week of March) and remained above 15 million until May 2021,
                                                                                 more than one year into the pandemic.
3. UI performance during the great recession and the Covid-19                        The federal “pandemic” programs brough about an unprecedented
pandemic                                                                         expansion of UI coverage: since November 2020, PUA and PEUC covered
                                                                                 around 70–75 percent of total unemployed workers receiving UI bene­
   The Great Recession – officially lasting from December 2007 to June           fits, with a peak of almost 17 million in August 2020. When they expired
2009 – was the longest and most-severe one since World War II. Between           on September 6, 2021, over 8.5 million workers were still receiving
December 2007 and October 2009, the unemployment rate doubled to                 benefits from PUA and PEUC, as opposed to only 2.4 million from the
10 percent and did not fall below 5 percent until late 2015; by February         regular state UI (Fig. 2). These pandemic programs are the driving
                                                                                 reason why the UI system has been able to deliver such wide assistance
                                                                                 during the Covid-19 downturn, overcoming the narrow eligibility of
 4
    Memorandum Authorizing the Other Needs Assistance Program for Major          traditional UI benefits. In the same vein, the FPUC added a substantial
Disaster Declarations Related to Coronavirus Disease 2019.                       increase in the level of benefits. Reflecting the expiration of the

                                                                             4
F. Spadafora                                                                                                      Journal of Government and Economics 9 (2023) 100069




          Fig. 1. UI recipients by programme 2007–15. Source: author’s elaborations on DOL data: https://oui.doleta.gov/unemploy/DataDashboard.asp




          Fig. 2. UI recipients by programme 2020–21. Source: author’s elaborations on DOL data: https://oui.doleta.gov/unemploy/DataDashboard.asp


pandemic emergency programs in September 2021, the number of                        a lower extent, the Pandemic Unemployment Assistance (PUA): these
workers receiving UI benefits dropped abruptly to a low of about 2                  facilities brough about an unprecedented expansion of both UI eligibility
million weekly by December 2021, 90 percent of which from the regular               and the level of benefits, at an overall cost of $440 billion and $130
state UI.                                                                           billion, respectively, over 2020–21.
                                                                                        To appreciate the unprecedented generosity of the FPUC – initially
                                                                                    set at $600 weekly and then reduced to $300 since December 2020 – it
3.1. Costs of the pandemic UI programs                                              suffices to recall that in the 2009 ARRA stimulus package, weekly UI
                                                                                    benefits were increased by just $25 (Burtless, 2020). Another compari­
   The unprecedented expansion of coverage, eligibility and level of UI             son points to the fact that, in January 2020, the nationwide average UI
benefits in response to the Covid-19 pandemic is mirrored by the                    weekly benefit was about $385: adding $600 to this amount yields an
massive increase of outlays. In 2020–21, total UI programs cost a cu­               average weekly benefit of $985, an increase of about 155 percent
mulative of $857 billion, of which about $656 billion from the emer­                (Burtless, 2020).
gency pandemic programs (PUA, PEUC and FPUC), which expired on                          The overall cost of the three new federal pandemic facilities – $656
September 6, 2021. Regular state UI benefits cost $188 billion, while EB            billion– has greatly exceeded early estimates by the CBO (2020, p. 9),
outlays were limited to less than $13 billion only (Table 1).                       which in April 2020 had estimated that outlays for UI would increase “by
   The scale of the exceptional pandemic support is best epitomized if              a total of $263 billion in 2020 and 2021 as a result of these provisions”, of
one considers that the total cost of all UI programs in response to the             which $176 in 2020 for the FPUC alone. Given its extension from
Great Recession amounted to $570 billion over six years (2008–2013); a              December 2020 through September 6, 2021, the cost of FPUC has grown
key difference is that the largest share of the total cost ($311 billion) was       because of the rising share of individuals exhausting the regular state UI
accounted for by the regular UI program; like during the pandemic,                  benefits.
payments under the EB program were rather limited ($29 billion) and a                   It is also remarkable to note that the CBO had estimated that the PUA
substantial share of the total cost derived from the emergency federal              program would be claimed by 5 million people at a total cost of $35
program – the EUC08 – whose price tag hit $230 billion.                             billion in 2020 and 2021: in the event, about 15 million workers were
   As previously noted, at its peak in February-March 2010 EUC08 was                receiving the PUA at its peak in August 2020, while the total cost of the
covering 5.9 million unemployed workers; at their peaks in August                   program amounted to $130 billion.
2020, PUA and PEUC were covering 17 million such workers.
   While the PEUC closely resembles the expansion of federal emer­
gency unemployment benefits enacted in previous recessions – and its                3.2. Financial impact of the great recession and the Covid-19 pandemic
cost has been relatively contained ($85 billion) – a key feature of the             on state trust funds and UI taxation
federal pandemic programs is that the largest sources of additional costs
are the Federal Pandemic Unemployment Compensation (FPUC) and, to                      State trust funds are used to pay UI benefits to eligible recipients of

                                                                                5
F. Spadafora                                                                                                      Journal of Government and Economics 9 (2023) 100069


the regular state UI programs. Each state deposits its SUTA revenues into           quarter of 2022; the net balance turned in a small surplus by the first
its individual state UI trust fund account within the federal Unemploy­             quarter of 2022, which increased to $27 billion by the third quarter of
ment Trust Fund (UTF) held at the U.S. Treasury. Although there are no              that year. Title XII advances peaked at close to $55 billion in the second
federal requirements for the level of funds that should be kept in a state          quarter of 2021 and were down to $27 billion by the third quarter of
trust fund, statea operate on a so-called “forward funding” basis, i.e.,            2022.
they build up reserves in expansionary years in anticipation of paying                  Similar to the impact of the Great Recession, the deterioration of
higher amounts of benefits during recessionary times (DOL, 2020a).                  state trust funds’ balances and the attendant borrowing from the federal
    This notwithstanding, during deep or prolonged recessions SUTA                  unemployment trust fund raise the risk that states will respond to the
taxes and UTF account balances may be insufficient to cover the states’             binding financial constraints by reducing UI benefits and/or increasing
legal obligation to pay UI benefits: federal law thus provides a loan               taxes, also considering that in some states additional contributions are
mechanism – through the Title XII program – whereby states may                      required by law when trust fund balances drop to specified points.
borrow funds from the Federal Unemployment Account (FUA), which is                      In fact, it is important to emphasize that most – if not all – states have
the federal loan account within the UTF.                                            trust fund balance targets written into their state law, with triggers built
    As of June 2008 – on the eve of the Great Recession – state trust funds         in to adjust the tax rates annually according to the state’s trust fund
had an aggregate balance of $40 billion and a negligible amount (less               balance (GAO, 2010, p. 7; DOL, 2020b, p. 2–18). In the past, periods of
than $180 million) of Title XII loans (Fig. 3). The impact of the Great             elevated UI benefit outlays have historically been followed by higher
Recession is dramatically visible in the fact that these loans peaked at            taxes to restore the solvency of state trust funds (Weidinger, 2020a;
$47 billion in the first quarter of 2011, when the aggregate balance of             Towson, 2020). For example, average state UI tax rates increased from a
state trust funds hit a low of $7.6 billion (the net trust fund balance was         low of 2.3 percent in 2008 to a post-Great Recession peak of 3.5 percent
thus close to -$40 billion). It took more than two years for the net                in 2012, before declining over the next eight years – to a low of 1.72
aggregate trust fund balance to go back into a small surplus ($1.7                  percent by end-2020. The impact of Covid is visible already in 2021,
billion) in the second quarter of 2013. Moreover, in the aftermath of the           when the average rate increased to 1.89 percent. Preliminary estimates
Great Recession, states’ borrowing needs for their UI programs far                  by the Department of Labor suggest that, for the United States as a
exceeded the available federal UI trust fund reserves, forcing the FUA              whole, the average UI tax rate would increase to 2.3 percent for 2022 as
itself to borrow funds from the U.S. Treasury between 2008 and 2015 to              a whole.6 However, the rapid improvement in the state UI trust fund
finance loans to the state accounts (Whittaker, 2018).                              balances mitigates the risk of UI tax increase; moreover, states are
    The steady replenishment of the state trust funds since 2013 is visible         authorized to use federal Fiscal Recovery Funds to replenish UI trust
in the fact that by the end of 2019 – on the eve of the Covid-19 crisis –           funds to pre-pandemic levels.
these funds had a sizable aggregate balance of about $76 billion (DOL,                  Finally, it is worth noting that the links between UI tax rates and
2020a). The amount of outstanding Title XII loans stood at only $63                 financial conditions of state trust funds may result in both automatic
million (owned by the U.S. Virgin Islands), down from a peak of $47                 stabilization of trust fund balances and procyclical effects on the busi­
billion in March 2011 (DOL 2020a). It is important to note that the                 ness cycle, which may erode at least some of the stabilizing macroeco­
solvency of state trust funds is a critical factor in determining the overall       nomic effects of paying UI benefits (GAO, 2010, p. 24).
performance of the UI system in terms of coverage, duration and level of
UI benefits.                                                                        4. Demand- and supply-side effects of pandemic UI programs
    The outbreak of the Covid-19 pandemic implied a massive spike in
regular UI benefit payments by states, which in 2020 alone reached an                   UI benefits face policymakers with a rather standard tradeoff that
all-time record of almost $144 billion (Table 1), an amount close to half           characterizes any form of insurance, in this specific case: how to effec­
the entire cost of regular UI benefits during the six years of the Great            tively provide income support in the face of involuntary unemployment
Recession and its aftermath (2008–2013) and almost twice as much as                 while avoiding disincentives to work and broader moral hazard effects.
the $76 billion aggregate trust funds balance at the start of 2020. Similar             The emergency UI programs adopted in response to the pandemic
to what happened in the aftermath of the Great Recession, the massive               have provided an opportunity to reassess – under admittedly unique
increase in UI benefits triggered by the Covid-19 pandemic forced states            circumstances – two long-standing issues highlighted in the literature
to begin borrowing again from the Federal Unemployment Account: in                  about the impact of UI benefit extensions (i.e., duration and/or gener­
2020, 22 states took Title XII advances (DOL, 2021a).                               osity). On the demand side, the focus has been on assessing the coun­
    As a result, by December 2020 the aggregate balance of state trust              tercyclical capabilities of emergency UI programs, their possible
funds (net of Title XII advances) had turned into a deficit of $21 billion          Keynesian effects and the size of the fiscal multipliers; on the supply
(DOL, 2021a); outstanding Title XII federal advances hit $45.5 billion.             side, empirical research has paid most attention to evaluate the potential
The financial situation further deteriorated in the first quarter of 2021,          impact of UI benefits on job-search incentives (i.e., willingness to accept
when the net deficit peaked at $27 billion.                                         a job) and ultimately employment.
    At end-2020, five states (California, Illinois, Massachusetts, New                  The fact that, in the early summer of 2021, 24 states opted out the
York and Texas,) had a combined deficit of $37.7 billion in their trust             pandemic UI programs before their legislated expiration date of
funds, a major turnaround from an aggregate surplus of $11.5 billion at             September 6 has provided a unique setting for testing empirically the
end-20195.                                                                          demand- and supply- side effects of UI benefit extensions.
    However, contrary to what happened during the Great Recession,
financial conditions of state UI trust funds have improved much faster in           4.1. UI’s demand-side effects and countercyclical capabilities
the aftermath of the Covid-19 pandemic. Reflecting the economic re­
covery and the attendant improvement of the employment situation, UI                    Macroeconomic stabilization is one of the key functions of the UI
outlays declined to $44 billion in 2021, compared to almost $144 billion            system, which is generally believed to be highly cyclical and responsive
in 2020, and further down to only $24 billion in 2022 (Table 1). As a               to distress in the U.S. economy. Furthermore, temporary extensions of
result, the balance of state trust funds began to improve as early as the           the duration of UI benefits (as well as increases of their level) are usually
second quarter of 2021 and by end-2021 had increased to $39 billion                 seen as a most effective instrument for countercyclical macroeconomic
(from $25 billion at end-2020) and further up to $54 billion by the third           stabilization in downturns (Yang et al., 2010), when the insurance value


 5                                                                                   6
     Author’s elaborations on data from DOL (2020a and 2021b).                           https://oui.doleta.gov/unemploy/avg_employ.asp.

                                                                                6
F. Spadafora                                                                                                      Journal of Government and Economics 9 (2023) 100069




Fig. 3. Trust Fund and loan balances 2007–2021 (quarterly). Source: author’s elaboration on U.S. Department of Labor data https://oui.doleta.gov/unemploy/data_
summary/DataSum.asp


of unemployment benefits is greater for a given level of moral hazard                evidence on both the marginal propensity to consume and the total
risk, which is anyway lower in downturns given weak conditions in the                output multiplier of UI – which factors in possible disincentive effects on
labor market (Dube, 2021).                                                           labor supply – and settle on a multiplier for federally financed UI pro­
    As emphasized by Hellwig (2021), the distinctive stimulative power               grams of between 0.5 and 1 when monetary policy does not respond,
of UI programs goes beyond the very injection of cash into the economy               while underscoring that the evidence cannot rule out a multiplier as
that supports consumption – i.e., the traditional Keynesian effect – and             small as 0 or as large as 2. The authors also specify that the multiplier
rather arises from their crucial role in reducing idiosyncratic household            may be larger for UI extensions than increases in benefit levels and
income risk, which in turn gives rise to several powerful channels                   smaller when monetary policy is active. More recently, Hellwig (2021)
through which benefits may support aggregate demand: for example,                    provides estimates of fiscal multipliers of fairly similar size to previous
preventing mortgage default, preserving housing wealth, reducing pre­                estimates found in the literature, confirming the ability of temporary UI
cautionary savings.                                                                  benefit extensions to stimulate employment and aggregate demand.
    Even before the pandemic, empirical research had documented the                      The Covid-19 pandemic has provided a major test case for the UI
strong demand effects from the extension of UI benefits under emer­                  system to demonstrate its full potential as a macroeconomic stabilizer,
gency federal programs. In a study using county-level data and a                     all the more so given the unprecedented expansion of UI benefits in
difference-in-difference approach, Hellwig (2021) estimates the relative             scope and size, even when compared to the Great Recession. Because of
magnitude of supply and demand effects of benefits extension and finds               the combined effect of a sudden rise in the unemployment rate and the
that the boost to aggregate demand from additional weeks of UI benefits              expanded pandemic programs, between May and July 2020 unem­
is more than enough to offset any labor market disincentives. This                   ployment benefits amounted to close to 7.0 percent of total monthly
positive effect is driven by such non-tradable sectors as retail and con­            personal income, a record far exceeding the 1.3 percent peak during the
struction. In this latter sector, one reason is that UI benefits play a strong       Great Recession.
role in stabilizing house prices and thus generate a wealth effect for                   It has been estimated that UI benefits, as a share of wage and salary
homeowners.                                                                          income, provided an economic boost roughly four times as great during
    One specific reason highlighted in the literature why UI benefits are            the pandemic as during any previous recessions (Bivens and Banerjee,
an optimal option for providing stimulus during downturns is because                 2021). In fact, data from the Bureau of Economic Analyst show that UI
unemployed workers have a higher marginal propensity to consume                      benefits as a ratio of employees’ wages and salaries reached a peak of 2.6
(MPC), of around 0.8 according to recent estimates (Ganong and Noel,                 percent in January 2010, largely because of the support from the EUC08
2019): consumer spending is thus generally highly sensitive to UI ben­               program; during the Covid-19 pandemic, the above-mentioned ratio
efits, which means that increasing their generosity can be an effective              peaked at over 15 percent in June 2020 and remained close to 5 percent
macroeconomic stabilization tool.                                                    through May 2021, largely because of the new FPUC benefit. With the
    Kekre (2022) emphasizes the role of expanded UI benefits as a                    early termination of pandemic programs in many states beginning in
discretionary tool in the stabilization of short-run fluctuations and                June 2021, the ratio decreased to 3.4 percent by August. When these
studies the output and employment effects of UI in a general equilibrium             programs all expired by law in early September, the ratio further
framework with incomplete markets, search frictions, and nominal ri­                 dropped to 0.43 percent in October 2021, still higher than the
gidities. Calibrating the model to the U.S. economy during the Great                 pre-pandemic 0.26 percent of February 2020.
Recession, the author finds an important stabilization role of UI, with an               Using estimates of the marginal propensity to consume, Autor et al.
output multiplier ranging between 0.6 and 1.8 and an unemployment                    (2022) offer a back-of-the-envelope comparison of the degree of stim­
rate that would have been as much as 0.4 percentage points higher in the             ulus provided by three main programs adopted in response to the
absence of the extensions to the EUC08 program. These results depend to              pandemic: the Paycheck Protection Program, the Economic Impact
a significant extent on the heterogeneity in the marginal propensity to              Payments (“stimulus payments”) and pandemic UI programs such as
consume MPCs – i.e., higher for the unemployed than the employed –                   PUA. The authors’ calculations show that the overall MPC out of UI
the decline in agents’ precautionary saving (because of lower income                 benefits is higher than the other two programs. This is notable if one
risks provided by UI) and an accommodative monetary policy (i.e., no                 considers that 40 percent of PUA recipients are estimated to be high
increase in the nominal interest rate).                                              income self-employed business owners (Boesch et al., 2021) that have
    Chodorow-Reich and Coglianese (2019) summarize the empirical                     lower MPCs than the one assumed for standard UI recipients.


                                                                                 7
F. Spadafora                                                                                                    Journal of Government and Economics 9 (2023) 100069


    The $600 FPUC introduced by the CARES Act clearly stands out in                implementation lags – typically ramp up transfers late in a recession, if
terms of generosity; several studies have thus analyzed both its coun­             not after when it has ended – which may undermine their countercy­
tercyclical impact and the potential for moral hazard effects on labor             clical effectiveness; second, the EB program – the automatic part of
supply. The latter can be first appreciated by assessing the FPUC impact           federal benefit extensions – has historically been modest in size: in fact,
on the UI replacement rate (Burtless, 2020; Ganong et al., 2020). With a           during the pandemic regular state UI benefits amounted to $188 billion,
typical UI replacement rate of around 40 percent of past earnings as of            while EB outlays were limited to less than $13 billion (Table 1).
end-2019, the $600 FPUC has raised this rate to about 105 percent;                     Against this background, it has been argued that reforms to enhance
Ganong et al. (2020) estimate that 76 percent of unemployed workers                the automatic stabilizer properties of UI should first apply to regular UI
eligible for regular UI had statutory replacement rates above 100                  benefits, focusing on raising the recipiency rates and increasing the
percent: for the median worker, the $600 supplement nearly tripled the             weekly benefit amount (Chodorow-Reich and Coglianese, 2019). In this
typical benefit levels, replacing 145 percent of lost income. As a result,         regard, it is worth noting that, because of the temporary measures
many unemployed workers, especially those earning below-average                    included in the CARES Act – notably the relaxation of UI eligibility
wages, received weekly UI benefits that are greater than the weekly                criteria – the average recipiency rate of regular UI benefits exceptionally
earnings they lost, a “big bazooka in terms of countercyclical stimulus”           increased from 28 percent in 2019 to 70 percent in 2020 – with a peak of
(Burtless, 2020). It is also important to note that the extraordinarily high       around 90 percent in the second and third quarters – before declining to
replacement rates delivered by enhanced unemployment benefits may                  36 percent in 2021 and 25 percent in 20227.
have diminished the recipients’ marginal propensity to consume (Autor                  Many reform proposals focus on the EB program and its triggers
et al., 2022).                                                                     (West et al., 2016; O’Leary and Wandner, 2018; Chodorow-Reich and
    The unique impact of the FPUC – through exceptional replacement                Coglianese, 2019; Bivens et al., 2021). The latter, as noted in Section 2.1,
rates higher than 100 percent – is also confirmed by the evidence that,            have long been criticized as inadequate to timely respond to economic
when the $600 supplement was available, the spending of unemployed                 downturns, thus contributing to making the UI system underperform as
households actually rose after a job loss, both in absolute terms and              a macroeconomic stabilizer. For example, soon after the Great Recession
relative to the spending of employed households (Ganong et al., 2021).             the Advisory Council found that the temporary federal extensions of
This spending increase stands in sharp contrast to normal times when               unemployment benefits had been “extremely inefficient”, as they were
spending falls sharply after a job loss and is particularly striking because       neither well timed nor well targeted (Isaacs and Whittaker, 2014b, p. 9).
overall spending was substantially depressed during the pandemic.                      Proposed reforms of the EB program rest on the principle that, for
    All in all, the $600 boost from the FPUC is believed to have helped            automatic stabilizers to serve their role, they need to trigger on in a
turn what would have been a very sharp and protracted reduction in                 timely manner at the onset of a downturn, but also need to continue to
spending into a gain in spending compared to the baseline (Dube, 2021,             support the recovery – rather than “putting the brakes on” by an early de-
p. 4). According to some estimates (Ganong et al., 2021), the FPUC                 activation (Dube, 2021; Chodorow-Reich and Coglianese, 2019). To this
supplement increased total spending by 2.0–2.6 percent between April               end, proposals include: i) making EB fully federally funded in order to
and July 2020.                                                                     remove the fiscal disincentives for states to opt out and only use the
    Further evidence on the demand-side impact of pandemic UI pro­                 default IUR trigger; ii) reforming the triggers to enhance the automatic
grams is provided by Coombs et al. (2021), who study the effects of the            extensions of benefits during periods of extremely high unemployment,
decision by 22 states to end early – in June 2021 as opposed to the                including by reinstating the national trigger abolished in 1981 (Hellwig,
expiration date of September 6 – all the above-mentioned supplemental              2021); iii) extending the additional weeks of EB benefits to address
benefits. The authors find that UI recipiency dropped by 35 percentage             long-term unemployment; iv) removing look-back provisions from EB
points among workers who were unemployed and receiving UI at the                   triggers that make automatic extensions turn off during periods of pro­
end of April 2021. Through the first week of August, average UI benefits           longed unemployment.
for these workers fell by $278 per week and earnings rose by $14 per                   Most of the proposals focus on designing the triggers to best link the
week, thus offsetting only 5 percent of the loss in income. Spending fell          duration of UI benefits to the state of the economy, a choice that involves
by $145 per week, as the loss of benefits led to a large immediate decline         important tradeoffs (e.g., level or change in the target variable, typically
in consumption. Despite this evidence supporting the macroeconomic                 unemployment; national or state unemployment rates). Recent simula­
relevance of cutting UI benefits, the authors underscored that these ef­           tions by Chodorow-Reich et al., 2022 suggest that policies designed to
fects might be overstated by the fact that the sample is composed                  trigger immediately at the onset of a recession (or even before it starts)
entirely of low-income and credit-constrained workers who are likely to            result in benefits extensions that occur in less slack labor markets. Some
respond more strongly to a loss of benefits than higher-income workers             simulations point to the benefits of a national trigger. Rather surpris­
affected by the same policy.                                                       ingly, the simulations suggest that, despite their ad hoc nature and risks
    In the face of their success, the pandemic UI programs introduced by           of implementation lags, UI benefit extensions legislated by Congress
the CARES Act have rekindled the debate about whether these federal                during the past crises compare favorably ex post to common proposals
extensions to complement the regular state UI programs during re­                  for automatic triggers.
cessions improve their countercyclical capabilities. On the one hand, it is
recognized that the need for longer durations of UI benefits is magnified          4.2. UI’s supply-side effects and moral hazard risks
during and (frequently) even after recessions, because job openings in­
crease slowly and many UI recipients experience prolonged periods of                   The labor market effects of UI benefits have been extensively studied
unemployment well after the end of a recession (O’Leary and Wandner,               so there exists ample literature on the potential job-search disincentives
2020). The Great Recession is a most prominent example in this regard,             effects and moral hazard costs of unemployment insurance (Giupponi
as the rise of long-term unemployment has come to represent one of its             et al., 2022). In principle, these effects might be magnified for pandemic
most visible legacies.                                                             UI programs, given their unprecedented generosity, and have provided a
    On the other hand, it is emphasized that, even in a severe recession,          key argument to those who oppose a structural expansion of the UI
regular state UI programs have provided the bulk of the increase in                system.
transfers to unemployed workers, notably at the early stage (Chodor­                   The academic literature is overall divided about the extent to which
ow-Reich and Coglianese, 2019). For their part, federal emergency UI               UI benefits create supply-side distortions by affecting job search
benefits have historically played a surprisingly small role in providing
macroeconomic stimulus early in recessions, for two main reasons: first,
emergency programs lack automaticity and are often subject to                       7
                                                                                        https://oui.doleta.gov/unemploy/data_summary/DataSum.asp

                                                                               8
F. Spadafora                                                                                                   Journal of Government and Economics 9 (2023) 100069


incentives and labor market outcomes; the strength of these effects re­           impact is quite modest when compared to the overall effect of Covid-19
mains fiercely debated (Chodorow-Reich and Coglianese, 2019).                     itself. Importantly, Holzer et al. (2021) emphasize that some studies’
    Some recent microeconomic studies find no effect of benefit gener­            evidence of minimal distortions of pandemic UI benefits on labor supply
osity on unemployed workers’ willingness to accept a job, others provide          should be seen as an anomaly due to the unique circumstances of the
evidence that extending the UI benefit duration leads to reduced job              pandemic: from the early stages of the pandemic in 2020 through the
search effort, though the effect declines in recessions (Hellwig, 2021 and        spring of 2021 – when uncertainty and social distancing were domi­
the references therein).                                                          nating factors – the familiar relationship between unemployment
    This notwithstanding, a fairly shared conclusion of the empirical             duration and UI benefit generosity may have somewhat broken down.
research is that the duration of unemployment spells is strongly                  Subsequent studies would support the pre-pandemic consensus that the
responsive to the generosity of unemployment insurance (Giupponi                  duration of unemployment increases with the generosity of unemploy­
et al., 2022). Against this background, it is remarkable that the findings        ment benefits: in this regard, the summer of 2021 could be interpreted as
of the available studies on the effects of pandemic UI programs are               a period of “healing” and of returning normalcy for the labor market
somewhat less unambiguous than those highlighted by the                           (Holzer et al., 2021).
pre-pandemic literature, as several of them suggest that the distortions              Other studies have specifically focused on the impact of pandemic UI
to labor supply are minimal. Part of this divergence from the previous            programs on job-search incentives.
consensus can be accounted for by the unique circumstances of the                     Marinescu et al. (2021) study the impact of the FPUC on job appli­
pandemic (Holzer et al., 2021): for example, labor supply may be less             cations and vacancy creation and find that a 10 percent increase in
sensitive to the generosity of UI benefits when workers face health               unemployment benefits caused a 3.6 percent decline in applications, but
concerns; furthermore, the elasticity of unemployment duration with               did not decrease vacancy creation; hence, FPUC increased labor market
respect to UI benefit generosity may have been affected by the sub­               tightness (vacancies/applications). Since the findings suggest that
stantial increase of household savings arising from both direct cash              tightness was unusually depressed during the period when FPUC was
payments from various government programs and less opportunity to                 available, the authors conclude that, altogether, their results imply that
spend money due to social distancing measures.                                    the positive effect of FPUC on tightness was likely welfare improving:
    A first batch of empirical studies have focused on the impact of              FPUC decreased competition among applicants at a time when jobs were
specific pandemic UI programs – PUA and FPUC – on employment.                     unusually scarce.
    After surveying several of these studies, Dube (2021, p. 4) concludes             Furthermore, recognizing that the unprecedented increase in UI
that, overall, the evidence points to a surprisingly small impact of              generosity caused weekly benefit payments to exceed prior earnings for
benefit generosity on employment, at both the micro and macro levels.             most recipients, Petrosky-Nadeau and Valletta (2021) analyze the job
For their part, Ganong et al. (2021) find that the FPCU supplement                acceptance decision for a wide range of U.S. workers by estimating a
generated minimal disincentive effects on employment, which                       “reservation UI benefit”. They conclude that only a small fraction of
decreased by only 0.2–0.4 percent as a result of the reduction in job             workers would turn down an offer to return to work at their previous
search. Further evidence comes from studies that focused on the impact            wage and would rather prefer to keep receiving expanded UI payments
of the decision by 24 states to opt out of the pandemic PUA and FPUC              provided by the CARES Act; besides, like in Ganong et al. (2021) the
programs in the early summer of 2021, that is, before their legislated            disincentive effects of the $600 supplemental payments on job-finding
expiration date of September 6                                                    rates were found to be modest: the additional income provided to the
    The relevance of this decision can be best appreciated by noting that         unemployed through the CARES Act (and subsequent legislation) likely
in early June 2021, about 75 percent (11.2 million) of unemployed                 acted as an effective targeted fiscal transfer supporting aggregate de­
workers were receiving PUA and PEUC benefits. When they expired on                mand, while having little impact on the unemployment rate via labor
September 6, these benefits were still supporting 8.5 million unem­               supply effects.
ployed workers. Because of the early withdrawal decision, pandemic                    The PUA benefit – at times referred to as the “PUA experiment”
supplemental UI benefits were eliminated entirely for over 2 million              (Greig et al., 2022) – has received particular attention because of its
workers and reduced by $300 per week for over 1 million workers.                  features mark a departure from traditional UI programs, in that it
    Using anonymous bank transaction data and a difference-in-                    enabled self-employed and other workers not covered by UI (e.g.,
differences research design, Coombs et al. (2021) measure the effect of           part-timers and those lacking a sufficient long work history) to become
withdrawing pandemic UI benefits on the financial and employment                  eligible for many Covid-19 related reasons beyond involuntary job loss.
trajectories of unemployed workers in states that withdrew benefits               However, it is accepted that between 40 and 50 percent of PUA re­
(“Withdrawal” states), compared to workers with the same unemploy­                cipients were self-employed. PUA dramatically expanded UI eligibility
ment duration in states that retained these benefits (“Retain” states).           as it represented around 40 percent of total UI claims.
They find that ending pandemic UI benefits increased employment by                    Greig et al. (2022) do not see clear evidence that PUA recipients
4.4 percentage points in Withdrawal states relative to Retain states.             exhibited greater work disincentives than traditional UI recipients. PUA
    In the same vein, Albert et al. (2022) assess the impact of the FPUC          did not provide a flat benefit but rather the weekly benefit amount was
withdrawal on the pace of hiring, given that this policy change was               linked to the income declared by the applicant. The authors examine the
intended in part to ease hiring challenges employers faced from tight             flows of recipients entering and exiting the PUA program versus the
labor markets. The authors compare the labor market outcomes between              traditional UI program and find that the exit rate – a proxy for job search
the 27 states (including the District of Columbia) that maintained the            – is only slightly lower (by 5 percent) for PUA than the corresponding
FPUC in place (“keep UI” states) and the 24 states that eliminated it (“cut       rate for traditional UI in 2020 and broadly similar in 2021. This would
UI” states). Statistical analysis shows that, over the full time the policy       signal that PUA did not generate especially strong work disincentives.
difference was maintained (July to September 2021), the states that cut               Finally, the pandemic UI programs stand out also in terms of their
UI benefits experienced a relative increase in hiring rates. However, the         distributional impact, which had been the object of several empirical
size of the effect—about 0.2 percentage point—is quite small relative to          studies.
monthly hiring rates of around 4 to 5 percentage points.                              Ganong et al. (2022) argue that the expansion of UI programs was
    For their part, Holtzer et al. (2021) run a counterfactual exercise and       highly progressive in that they offset income losses and delivered the
estimate that eliminating FPUC and PUA would have lowerd by 0.3                   most benefit to lower-income workers, not least because job losses were
percentage points the national unemployment rate in both July and                 concentrated in low-wage service sectors. For example, the FPUC, unlike
August 2021 while employment would have been 0.2 and 0.1 points                   typical stabilization tools such as broad-based stimulus checks, has been
higher, respectively. As noted before, the authors acknowledge that this          targeted to a subset of the population hard hit by the recession and may

                                                                              9
F. Spadafora                                                                                                    Journal of Government and Economics 9 (2023) 100069


have reversed income patterns which would have otherwise arisen                    single program fully financed and administered at the federal level. Most
across income levels, occupations, and industries (Ganong et al., 2020).           proposals are instead centered on introducing – or expanding – mini­
    As far as PUA is concerned, Autor et al. (2022) argue that the inci­           mum federal standards for the key UI parameters: eligibility re­
dence of unemployment insurance during the pandemic was weighted                   quirements as well as level, duration and funding of benefits.
towards both the upper and lower tails of the U.S. household income                    The peculiar experience during the pandemic has not only rekindled
distribution. The authors estimate that over 50 percent of pandemic                but also enriched with new issues the debate on reforming the UI system,
unemployment insurance payments went to the bottom fifth and                       with issues related to the equity of the UI system – primarily in terms of
second-to-bottom fifth of households but also – and rather surprisingly –          income, race, gender and age – now playing a more prominent role that
that the top fifth of households received a bit more than one-quarter of           appropriately informs the design of the reform proposals (Biven et al.,
unemployment insurance benefits: the latter finding can largely be                 2021; Stettner and Pancotti, 2021; Stone, 2021).
accounted for by the PUA, which extended UI benefits to the                            At the same time, one needs to recognize that the unique nature of
self-employed, often business owners with high incomes.                            the COVID-19 downturn – triggered by a public health emergency –
    In the same vein, Greig et al. (2022) argue that PUA disproportion­            makes it difficult to disentangle enduring policy lessons from those that
ately benefited lower-income families and those more marginally                    are exclusive to the pandemic (Ganong et al., 2022).
attached to the labor market. This conclusion relies inter alia on the fact            Despite this caveat, three important UI reform issues exposed by the
that PUA recipients received their labor income through direct deposit             pandemic stand out: whether the expanded coverage of UI temporarily
in a dramatically lower percentage.                                                achieved through the pandemic programs should be made – at least in
    It should also be noted that the FPUC has had an impact also in terms          part – a permanent feature of the post-pandemic system; the need for
of the debate between targeted versus universal stimulus payments                  stronger IT and administrative frameworks to maximize the advantages
(Ganong et al., 2021). For the last 20 years, the federal government has           of an expanded coverage; the role of Short-Term Compensation schemes
regularly used universal or near-universal tax rebate payments at the              to discourage layoffs and complement UI benefits.
onset of recessions; targeting payments to certain particularly vulner­                On the first issue, several authors support to keep a permanent
able households – such as the unemployed workers – can be an alter­                version of pandemic programs in place. The starting point is the
native approach to fiscal stimulus. The spending impacts from targeted             recognition of a rather striking feature of the UI system before the
transfers are indeed substantial, even in a pandemic, and larger than              pandemic, namely that in ordinary times (i.e., except for recessions)
estimated spending responses to universal transfers in the past. In this           most unemployed workers don’t receive state regular UI benefits: on the
context, however, one should also consider the fiscal externality of               eve of the pandemic, less than one in three unemployed workers was
increasing the generosity of unemployment insurance, which is seen by              collecting them. The coverage of the UI system is traditionally measured
many as relatively large: according to Giupponi et al. (2022), the cost to         by the recipiency rate – the national percentage of unemployed workers
the government of an additional $1 of unemployment insurance ranges                receiving regular state UI benefits – which has trended downward over
from $1.50 to $2.50.                                                               the past 40 years: it was 28 percent in 2019, down from a peak of 44
                                                                                   percent in 1980 and 36 percent on the eve of the Great Recession (DOL
5. Key challenges and reform proposals                                             UI chartbook, https://oui.doleta.gov/unemploy/chartbook.asp).
                                                                                       Low recipiency rates reflect a number of reasons, notably a combi­
    The debate on how to reform the UI system and overcome the critical            nation of restrictive monetary and nonmonetary eligibility criteria
weaknesses resulting from its “patchwork” structure of 53 different state          (Chodorow-Reich and Coglianese, 2019). In particular, the high
programs and rules is a long-standing one and by far predates the                  thresholds of minimum earnings and/or minimum amount of work time
pandemic. The starting point is the recognition that the last major re­            requested by most states to be eligible for UI exclude many workers who
form of the system dates back to 1976 and was not really comprehensive             become unemployed. Furthermore, the fact that the take-up rate of UI
(O’Leary and Wandner, 2020). In the meanwhile, since the 1980s many                benefits among eligible individuals is less than one may reflect several
benefits of the UI program have eroded (O’Leary et al., 2020, p. 3) while          administrative and possibly psychological hurdles (Chodorow-Reich
the U.S. economy has undergone significant structural changes, notably             and Coglianese, 2019). Informational asymmetries may also play a role:
in terms of industry and occupational mix of employment as well as the             Boesch et al. (2021) note that the most common reason why the un­
demographics of the labor force, with increased female labor force                 employed do not apply for UI benefits is because they believe they are
participation and voluntary part-time employment.                                  not eligible.
    In this context, the pandemic has re-exposed key challenges that had               Against this background, most proposals to improve UI eligibility call
long been the objects of various reform proposals, more recently in the            for three main changes: 1) harmonizing monetary eligibility criteria
aftermath of the Great Recession. The latter prompted numerous                     across states, notably adoption of alternative base periods for calculating
comprehensive reports that set out detailed recommendations for re­                earnings; 2) extending eligibility to part-time workers and individuals
form, but they were not followed up by legislative action at the federal           seeking part-time employment; 3) granting UI eligibility to people who
level.                                                                             leave their jobs for “good cause” reasons, most notably compelling
    Although an in-depth analysis of the pre-pandemic UI reform pro­               family reasons, such as caring for family members or inadequate
posals goes beyond the scope of this paper, in general they aim at                 childcare. The importance of these reasons – and their potential to affect
addressing weaknesses in three main areas: coverage and adequacy of                UI recipiency rates – have been made even more prominent by the
benefits; sufficiency of funding; automatic countercyclical capabilities8.         pandemic.
    Most reform proposals foresee in one way or the other an expanded                  In fact, the latter has dramatically exposed a critical feature of the
role for the federal government to increase the adequacy, equity and               regular state UI programs, namely that they are not designed to cover all
efficiency of state UI benefits across the country. The more far-reaching          unemployed workers, i.e., they do not cover people who leave their jobs
proposals (Wandner, 2020; Dube, 2021) call for converting UI into a                voluntarily, those looking for their first jobs, and those reentering the
                                                                                   job market after leaving voluntarily; furthermore, self-employed
                                                                                   workers, “gig” workers, undocumented workers and students are tradi­
 8                                                                                 tionally not eligible to apply for UI benefits (Alcalá Kovalski and
    See West et al (2016), Wandner (2018) and O’Leary and Wandner (2018)
for some of the main comprehensive reform proposals put forth before the           Sheiner, 2020; Dube, 2021). The stringency of the normal eligibility
COVID pandemic and Bivens et al (2021), Dube (2021), Furman et al (2020),          standards has been exposed by the elevated number of unemployed
Simon (2021), Spadafora, (2022), Wandner (2020) and Weidinger (2020b) for          workers who during the pandemic have relied on the Pandemic Unem­
those advanced in the aftermath of the pandemic.                                   ployment Assistance program, which was explicitly introduced to cover

                                                                              10
F. Spadafora                                                                                                      Journal of Government and Economics 9 (2023) 100069


many otherwise ineligible workers.                                                   system as a fundamental source of income support for workers and –
    As a result, perhaps the single most prominent reform proposal based             equally important – a powerful countercyclical instrument for the
on the experience of the pandemic is to keep in place a permanent                    economy as a whole, with minimal evidence of job-search disincentives.
version of PUA (Dube, 2021; Ganong et al., 2022; Greig et al., 2022) to              The unprecedented expansion of UI coverage and generosity to provide
structurally expand UI coverage and provide income support to those                  financial relief to millions of suddenly unemployed workers and their
categories of workers more marginally attached to the labor force. Be­               families is rightly considered to be a success story (Dube, 2021). The
sides, a standing PUA-like program would provide time to establish                   budgetary cost of deploying the new “pandemic” UI programs has
protocols and enhance systems with a view to ultimately avoiding the                 inevitably been high, but pales in comparison to the economic and social
inefficiencies arising from the need to establish an entirely new program            support provided by these programs.
when facing a crisis, which implies processing peaks of claims volume                    The lifeline role played by these programs is vividly demonstrated by
that would likely hamper the timely delivery of benefits.                            the fact that in early September 2021 – when they expired – benefits
    The second most supported reform proposal calls for strengthening                provided by the Pandemic Unemployment Assistance (PUA) and
the so-called UI delivery infrastructure – most notably, Information-                Pandemic Emergency Unemployment Compensation (PEUC) programs
Technology-IT and administrative frameworks – with a view to not                     were supporting about 75 percent (8.5 million) of total unemployed
only maximizing the efficiency of UI – in terms of timely and accurate               workers receiving UI payments.
payments to eligible workers that magnify income support and coun­                       Despite this success, the implementation of the pandemic programs
tercyclical stimulus while minimizing the room for fraud – but also                  has once again highlighted several limitations of the state-based regular
allowing an effective design of UI benefits.                                         UI system, while acknowledging the significant heterogeneity across
    In fact, several authors (Boesch et al., 2021; Autor et al., 2022;               states in terms of performances and key UI parameters. In the event, like
Ganong et al., 2022) have emphasized that, during the pandemic, the                  during the Great Recession, the system has required significant emer­
often-outdated IT systems and limited administrative capacity of                     gency legislative interventions to ensure that the microeconomic and
state-based UI programs have substantially constrained optimal policy                macroeconomic support was adequate in scope and size to address the
design, with the result that the U.S. was forced to provide emergency aid            unique challenges posed by the pandemic. The latter has also exposed
“using a fire hose rather than a fire extinguisher” (Autor et al., 2022). The        new challenges, first and foremost the constraints posed by an inade­
FPUC is a case in point, in that it had to be introduced as a flat nation­           quate UI delivery infrastructure, not only in terms of ability to make
wide weekly supplement of $600 because IT and administrative weak­                   timely payments to UI recipients but also for the optimal design of
nesses made it infeasible to adopt a flexible supplement that would                  expanded UI programs.
target a higher wage replacement rate tailored to pre-job loss earnings,                 The pandemic has inevitably rekindled the debate on whether and
which likely would create fewer inefficiencies in terms of work disin­               how to fundamentally reform the UI system, as the last significant re­
centives. In other words, although the flat $600 FPUC was highly pro­                form dates back to 1976. Reform proposals by academics and policy­
gressive, it was not optimal UI policy, as it was largely based on a                 makers alike have long focused on addressing well-known weaknesses of
national average pre-job loss earnings level of workers who were un­                 the system, namely low recipiency and replacement rates and duration
employed before the pandemic began.                                                  of UI benefits in some states shorter than the long-established 26 weeks.
    A third prominent reform proposal is to expand the role of the so-               President Biden’s 2022 Budget itself included a set of principles that are
called Short-Time Compensation (or work) schemes (STCs), which                       intended to serve as the basis for any major reform of the UI system,
provide employers with the option of temporarily reducing workers’                   including more equitable and progressive financing mechanisms (DOL,
hours – as an alternative to layoffs – and replacing via STCs some of the            2021c).
workers’ lost income. In other terms, STCs aim at discouraging layoffs in                A distinctive reason why the Covid-19 crisis has revived the calls for
the first place but are best placed to address brief downturns rather than           reforming the UI system is because the pandemic has dramatically
a firm’s permanent fall in labor demand or production (see Giupponi                  exposed that regular UI programs tend to disproportionately leave un­
et al., 2022, for an in-depth discussion).                                           covered exactly the types of workers most affected by the pandemic,
    In Europe, STCs have represented the primary tool to protect workers             namely the self-employed, new entrants into the labor force, and part-
from the impact of the pandemic. On the contrary, STCs are available in              time workers with short employment and earnings histories. These
only 26 U.S. states and make up only a small fraction – less than 1                  workers are frequently ineligible for UI benefits because they may fail to
percent – of overall UI benefits (Boesch et al., 2021; Giupponi et al.,              meet the states’ often strict eligibility requirements; they were covered
2022). This major difference between Europe and the U.S. in the use of               by the new PUA program, whose success is testified by the fact that it
STCs reflects a diametrically opposed approach to respond to acute – and             supported more than 15 million workers at its peak usage in August
unprecedented during the pandemic – strains in the labor markers: the                2020 and around 5 million workers when it expired in early September
U.S. policy was centered on insuring the income of workers – by                      2021. This massive but temporary expanded coverage is pointed out as a
aggressively increasing the generosity of UI benefits– while in Europe               demonstration that piecemeal fixes to the UI system should give way to
the emphasis was on preserving the relationship between workers and                  reforms that establish a structurally wider safety net.
firms – by providing generous subsidies for hours reduction and tem­                     As a result, in the aftermath of the pandemic most attention is being
porary layoffs through STCs (Giupponi et al., 2022).                                 paid to the proposals that aims at permanently extending UI coverage to
    Despite that STCs in a recession are funded by the federal govern­               previously ineligible workers, with a view to closing the mismatch be­
ment, they show an overall low take up rate, which is usually accounted              tween often outdated UI requirements and the changed nature of the U.
for by such reasons as limited awareness among employers and several                 S. workforce during the past two decades. At the same time, it is
administrative constraints (Boesch et al., 2021). This notwithstanding,              recognized that the choices made often hurriedly under the pressure
these is some consensus that STCs should be seen as a complementary                  from the pandemic may not necessarily be the best ones for a structural
tool to other UI programs. As a result, reform proposals to add STCs to a            expansion of UI coverage (Boesch et al., 2021).
state’s toolkit intend to bolster the ability of the UI system to precent                Many proposals call for making permanent the pandemic-induced
layoffs.                                                                             expansion of UI eligibility to those workers who are forced to volun­
                                                                                     tarily quit their jobs because of “good causes”, most notably compelling
6. Concluding remarks                                                                family reasons. Equally important, it is emphasized that the weaknesses
                                                                                     of the UI system penalize disproportionately women and workers of
   The Covid-19 pandemic has once again confirmed the critical role                  color (Goger et al., 2020; Bivens et al., 2021; Stettner and Pancotti,
played during recessions by the U.S. Unemployment Insurance (UI)                     2021; Stone, 2021). In this context, a key welcome novelty of the

                                                                                11
F. Spadafora                                                                                                             Journal of Government and Economics 9 (2023) 100069


ongoing debate is the prominence taken on by issues related to the eq­                  massive financial pressure that the pandemic exerted on the state trust
uity of the UI system – primarily in terms of income, race, gender and                  funds, more states might display the same tendencies that occurred in
age – in appropriately informing the design of the reform proposals.                    the aftermath of the Great Recession, namely cutting benefit levels or
     The current context of a tight U.S. labor market and a faster-than-                shortening their duration to help replenish these funds’ reserves. This
expected decline in the unemployment rate – by now back to the pre-                     risk would exacerbate what is by many highlighted as the fundamental
pandemic level – likely run counter to the urgency to act, while the                    funding problem faced by the UI system, namely that neither unem­
difficulties of finding political consensus on reforming UI do not appear               ployment benefits nor UI tax rates and taxable wage bases are uniformly
to be lower than in the past. This is perhaps best epitomized by the                    indexed – i.e., by a federal standard – to consistently provide adequate
decision, in mid-2021, by about half of states to discontinue the                       revenues and benefits over time.
pandemic programs earlier than their expiration in early September.                         On the other hand, other observers discount this risk by pointing to
Unsurprisingly, this decision has received different interpretations: on                the rapid improvement as early as mid-2021 in the state trust fund
the one hand, it has been justified on the ground that job vacancies                    balances – even net of Title XII loans – because of the sizable decline in
appeared to be amply available and the pandemic UI programs were                        state UI payments and the robust economic recovery; besides the
seen as providing disincentives to apply for them; on the other hand, the               financial aid provided by federal legislation approved during the
decision has been criticized because in June-July 2021 the pandemic                     pandemic presents states with an opportunity – admittedly quite
programs were still supporting around 10 million unemployed workers.                    disputed – to replenish unemployment insurance funds without
     At a more general level, the choice of discontinuing these programs                increasing the payroll tax burden on local employers.
has been pointed out as a consequence of much control on the UI system                      Like before the pandemic, many reform proposals also aim at
ceded by the federal government to states, something that is seen by                    improving the macroeconomic countercyclical role of the UI system,
some as a failure to equitably protect working people (Bivens et al.,                   which has been critical in mitigating the recessionary impact of the
2021, p. 1). Many reform proposals thus call for a stronger role for the                pandemic; the focus is on enhancing the automaticity of UI’s stabiliza­
federal government, first and foremost to introduce nationwide federal                  tion capabilities, most notably by improving the design of the “triggers”
minimum standards for benefits eligibility, duration, and levels, with a                of the Extended Benefits program to ensure that it remains active for as
view to overcoming the substantial variations across states.                            long as unemployment is elevated.
     A common goal to be drawn out of the lessons from pandemic is to                       Reforming the UI system has always proved to be a difficult if not
make the UI system “recession-ready” (Ganong et al., 2022) and minimize                 contentious and divisive task since its inception. As noted by Blaustein
– if not eliminate – the traditional trade-off between speed and accuracy               (1993, p. 149), “unemployment insurance began in controversy; it was never
that characterizes the UI delivery infrastructure. To this end, it has been             to be free of controversy”. Whether the political consensus to act boldly
suggested that states should approach the UI delivery infrastructure as it              during the pandemic – forged by the dramatic challenges it posed – will
were economic disaster preparedness, by building adequate adminis­                      extend to undertake comprehensive reforms of the UI system inevitably
trative capacity to aptly target, calibrate, and deploy timely its responses            remains an open question at this juncture.
in an emergency. It is thus recognized that, ultimately, this goal would                    What is true is that the experience with the UI system in the United
imply a more active role by the federal government in responding to                     States provides fundamental lessons that can usefully inform the debate
acute labor market strains, most notably in providing technology and                    on whether and how to introduce in Europe a common unemployment
data infrastructure that could enable an effective design of UI programs                insurance scheme for macroeconomic stabilization.
– e.g., flexible benefit levels set at a target income replacement rate – but
also stronger and smoother eligibility verification and fraud prevention                Statements and declarations
(Greig et al., 2022).
     In fact, in the aftermath of the pandemic unprecedented attention is                - The author did not receive support from any organization for the
being devoted to upgrade the technological infrastructure of the UI                        submitted work
system, whose fragmentation along state lines has often been a cause of                  - The author has no relevant financial or non-financial interests to
delays in providing timely and equitable income support to unemploy­                       disclose
ment workers during the pandemic.9 To this end, in August 2021 the U.                    - The views expressed are those of the author and do not necessarily
S. Department of Labor announced a series of measures – once again                         reflect those of the Bank of Italy or the International Monetary Fund.
labelled “UI Modernization” (DOL, 2021c) – to address such most im­
mediate challenges as promoting equitable access, ensuring timely
payment of UI benefits, and developing IT solutions to modernize                        Declaration of Competing Interest
outdated state technology. At the same time, this UI modernization
effort recognizes that longstanding problems further exposed by the                         The authors declare that they have no known competing financial
pandemic can be addressed only through a comprehensive UI reform. In                    interests or personal relationships that could have appeared to influence
fact, the emergency temporary measures adopted during the pandemic –                    the work reported in this paper.
while successful – were not designed to address the fundamental
objective of restoring the UI system’s ability to provide adequate support              Acknowledgements
for workers in normal times.
     Accordingly, the need for structural reforms remains a matter of                       For their useful comments, I would like to thank Alfonso Rosolia,
debate. On the one hand, many observers believe that reforms are all the                Luigi Federico Signorini and an anonymous reviewer. The usual dis­
more necessary to fix the system’s underlying weaknesses and avoid a                    claimers apply.
return to the pre-pandemic status quo of low recipiency rates, gaps in
coverage and inadequate funding mechanisms. Besides, these reforms                      References
are meant to be more stringent because of the risk that, in the face of the
                                                                                        Albert, S., Lofton, O., Petrosky-Nadeau, N., Valletta, R.G., 2022. Unemployment
                                                                                            Insurance Withdrawal. Federal Reserve Bank of San Francisco Economic Letter,
                                                                                            2022-09.
 9
   Lack of modernization of states’ IT systems has been identified as one of the        Alcalá Kovalski, M., Sheiner, L., 2020. How Does Unemployment Insurance work? And
                                                                                            how is It Changing During the Coronavirus pandemic? Brookings Institution blog,
primary reasons underlying the difficulties in implementing the three key                   July 20.
pandemic UI programs introduced by the CARES Act (DOL, Office of Inspector              Autor, D., Cho, D., Crane, L.D., Goldar, M., Lutz, B., Montes, J.K., Peterman, W.B.,
General, 2021).                                                                             Ratner, D.D., Villar Vallenas, D., Yildirmaz, A., 2022. The $800 Billion Paycheck


                                                                                   12
F. Spadafora                                                                                                                          Journal of Government and Economics 9 (2023) 100069

    Protection Program: Where Did the Money Go and Why Did It Go There? NBER.                     Kekre, R., 2022. Unemployment Insurance in Macroeconomic Stabilization. Becker
    Working Paper 29669.                                                                               Friedman Institute for Economics at University of Chicago. Working Paper No. 2021-
Bauer, L., Edelberg, W., Parson, J., 2020. Unemployment Insurance Extended Benefits                    28.
    Will Lapse Too Soon Without Policy Changes. Brookings Institution blog, August 6.             Marinescu, I., Skandalis, D., Zhao, D., 2021. The impact of the federal pandemic
Bivens, J., Boteach, M., Deutsch, R., Diez, F., Dixon, R., Galle, B., Gould-Werth, A.,                 unemployment compensation on job search and vacancy creation. J. Public Econ.
    Marquez, N., Roberts, L., Shierholz, H., Spriggs, W., 2021. Reforming Unemployment                 200.
    Insurance: Stabilizing a System in Crisis and Laying the Foundation for Equity. A             Needels, K., Nicholson, W., Lee, J., Hock, H., 2016. Exhaustees of Extended
    joint report of the Center for American Progress, Center for Popular Democracy,                    Unemployment Benefits Programs: Coping with the Aftermath of the Great
    Economic Policy Institute, Groundwork Collaborative, National Employment Law                       Recession. Mathematica Policy Report.
    Project, National Women’s Law Center, and Washington Center for Equitable                     Nicholson, W., Needels, K., 2011. The EUC08 Program in Theoretical and Historical
    Growth. June.                                                                                      Perspective. U.S Department of Labor. February.
Bivens, J., Banerjee, A., 2021. How to Boost Unemployment Insurance As a Macroeconomic            Nicholson, W., Needels, K., Hock, H., 2014. Unemployment Compensation During the
    Stabilizer. Lessons From the 2020 Pandemic Programs. Economic Policy Institute.                    Great Recession: Theory and Evidence. National Tax Journal. March.
    October.                                                                                      O’Leary, C.J., Barnow, B.S., Lenaerts, K., 2020. Lessons from the American Federal-State
Blaustein, S.J., 1993. Unemployment Insurance in the United States: The First Half-                    Unemployment Insurance System for a European Unemployment Benefits System.
    Century. W.E. Upjohn Institute for Employment Research.                                            W.E. Upjohn Institute for Employment Research. Working Paper No. 16-264.
Boesch, T., Lim, K., Nunn, R., 2021. What Did and Didn’t Work in Unemployment                     O’Leary, C.J., Wandner, S.A., 2018. Unemployment Insurance Reform: evidence-Based
    Insurance During the Pandemic. Federal Reserve Bank of Minneapolis. August.                        Recommendations. editor. In: Wandner, S.A. (Ed.), Unemployment Insurance
Burtless, G., 2020. Unemployment Insurance as Social Protection and Stimulus during                    Reform: Fixing A Broken System. W.E. Upjohn Institute for Employment Research.
    the Coronavirus Crisis. The Brookings Institution Economic Studies. April.                         Chapter 5.
Center for Budget and Policy Priorities, 2023. How Many Weeks of Unemployment                     O’Leary, C.J., Wandner, S.A., 2020. An Illustrated Case For Unemployment Insurance
    Compensation Are Available? Policy Basics. April.                                                  Reform. W.E. Upjohn Institute for Employment Research. Working Paper No. 19-
Chodorow-Reich, G., Coglianese, J., 2019. Unemployment Insurance and                                   317.
    Macroeconomic Stabilization. The Hamilton Project. May.                                       Petrosky-Nadeau, N., Valletta, R.G., 2021. UI Generosity and Job Acceptance: Effects of
Chodorow-Reich, G., Ganong, P., Gruber, J., 2022. Should we have automatic triggers for                the 2020 CARES Act. Federal Reserve bank of San Francisco. Working Paper
    unemployment benefit duration and how costly would they be? Am. Econ. Assoc.                       2021–13.
    Papers Proc. 112, 112–116.                                                                    Simon, A., 2021. Unemployment Insurance At a crossroads: Tracing program Design
Coombs, K., Dube, A., Jahnke, C., Kluender, R., Naidu, S., Stepner, M., 2021. Early                    During and Beyond COVID-19. American Enterprise Institute. October.
    Withdrawal of Pandemic Unemployment Insurance: Effects on Earnings,                           Spadafora, F., 2022. Don’t Let Me Down: Unemployment Insurance in the United States.
    Employment and Consumption. Harvard Business School. Working Paper No. 22-                         Bank of Italy Occasional Paper No. 673.
    046.                                                                                          Spadafora, F., 2019. European Integration in the Time of Mistrust. Bank of Italy
Congressional Budget Office (2020), Preliminary Estimate of the Budgetary Effects of H.R.              Occasional. Paper No. 512.
    748, CARES Act, Public Law 116-136, April. https://www.cbo.gov/publication/                   Stettner, A., Pancotti, E., 2021. 1 in 4 Workers Relied On Unemployment Aid During the
    56334.                                                                                             Pandemic. The Century Foundation, Unemployment Commentary, 17 March.
Council of Economic Advisers, 2013. The Economic Benefits of Extending                            Stone, C., Chen, W., 2014. Introduction to Unemployment Insurance. Center on Budget
    Unemployment Insurance. December.                                                                  and Policy Priorities.
Dube, A., 2021. A Plan to Reform the Unemployment Insurance System in the United                  Stone, C., 2021. Congress Should Heed to President Biden’s Call for Fundamental UI
    States. The Hamilton Project. Policy proposal 2021-03April.                                        Reform. Center on Budget and Policy Priorities. May.
Furman, J., Geithner, T., Hubbard, G., Kearney, M.S., 2020. Promoting Economic                    Towson, T., 2020. How Severely will COVID-19 Impact SUI Tax Rates? Equifax Insights
    Recovery After COVID-19. The Aspen Institute, Economic Strategy Group. June 16.                    Blog, 18 March.
Ganong, P., Noel, P., 2019. Consumer spending during unemployment: positive and                   U.S. Department of Labor (2020a), State Unemployment Insurance - Trust Fund Solvency
    normative implications. Am. Econ. Rev. 109 (7).                                                    Report 2020.
Ganong, P., Greig, F., Liebeskind, M., Noel, P., Sullivan, D.M., Vavra, J., 2021. Spending        U.S. Department of Labor (2020b), Comparison of State UI Laws 2020.
    and Job Search Impacts of Expanded Unemployment Benefits: Evidence from                       U.S. Department of Labor (2020c), Unemployment Insurance Program Letter No. 14-20.
    Administrative Micro Data. Baker Friedman Institute for Economics at University of                 April 2.
    Chicago. Working Paper No. 2021-19.                                                           U.S. Department of Labor (2021a), State Unemployment Insurance - Trust Fund Solvency
Ganong, P., Greig, F., Noel, P., Sullivan, D.M., Vavra1, J., 2022. Lessons learned from                Report 2021.
    expanded unemployment insurance during COVID-19. In: Edelberg, W., Sheiner, L.,               U.S. Department of Labor (2021b), State Unemployment Insurance - Tax Measures
    Wessel, D. (Eds.), Recession Remedies: Lessons Learned from the U.S. Economic                      Report 2020.
    Policy Response to COVID-19. The Hamilton Project, Hutchins Center on Fiscal and              U.S. Department of Labor (2021c), Unemployment Insurance Modernization, Fact Sheet,
    Monetary Policy at Brookings.                                                                      August.
Ganong, P., Noel, P., Vavra, J., 2020. US Unemployment Insurance Replacement Rates                U.S. Department of Labor, Office of Inspector General (2021), Covid-19: States Struggled to
    During the Pandemic. Baker Friedman Institute for Economics at University of                       Implement CARES Act unemployment Insurance Programs, May.
    Chicago. Working Paper No. 2020-62.                                                           U.S. Government Accountability Office (2010), Unemployment Insurance Trust Funds:
Giupponi, G., Landais, C., Lapeyre, A., 2022. Should we insure workers or jobs during                  Long-standing State Financing Policies Have Increased Risk of Insolvency. April.
    recessions? J. Econ. Perspect. 36 (2), 29–54. N.                                              U.S. Government Accountability Office (2015), Unemployment Insurance: States’
Goger, A., Hadden Loh, T., George, C., 2020. Unemployment Insurance is Failing                         Reductions in Maximum Benefit Durations Have Implications for Federal Costs. April.
    Workers During COVID-19. Here’s how to Strengthen it. Brookings Institution                   Wandner, S.A., 2018. Why the unemployment insurance program needs to be reformed.
    Report. April.                                                                                     editor. In: Wandner, S.A. (Ed.), Unemployment Insurance Reform: Fixing A Broken
Greig, F., Sullivan, D.M., Anderson, S., Ganong, P., Noel, P., Vavra, J., 2022. Lessons                System. W.E. Upjohn Institute for Employment Research. Chapter 1.
    Learned from the Pandemic Unemployment Assistance Program During COVID-19.                    Wandner, S.A., 2020. Options For Unemployment Insurance Structural and
    JPMorgan Chase Institute, Washington, D.C.                                                         Administrative Reform: Proposals and Analysis. W.E. Upjohn Institute for
Hellwig, K.P., 2021. Supply and Demand Effects of Unemployment Insurance Benefit                       Employment Research. Policy Paper No. 2020-020.
    Extensions: Evidence from U.S. Counties. IMF. Working Paper 21/70.                            Weidinger, M., 2020a. Is the Unemployment Insurance System Ready For the Next
Holzer, H.J., Hubbard, R.G., Strain, M.R., 2021. Did Pandemic Unemployment Benefits                    recession? American Enterprise Institute blog postMarch.
    Reduce Employment? Evidence from Early State-Level Expirations in June 2021.                  Weidinger, M., 2020b. Extended: A Review of the Current and Proposed Duration of
    NBER. Working Paper 29575.                                                                         “Pandemic” Unemployment Benefits. American Enterprise Institute. June.
Isaacs, K.P., 2019. Unemployment Insurance: Consequences of Changes in State                      West, R., Dutta-Gupta, I., Grant, K., Boteach, M., McKenna, C., Conti, J., 2016.
    Unemployment Compensation Laws. Congressional Research Service report. October                     Strengthening Unemployment Protections in America. Center for Economic Progress,
    23.                                                                                                National Employment Law Project, Georgetown Center on Poverty and Inequality.
Isaacs, K.P., Whittaker, J.M., 2014a. Emergency Unemployment Compensation (EUC08):                     June.
    Status of Benefits Prior to Expiration. Congressional Research Service report. August         Whittaker, J.M., 2018. The Unemployment Trust Fund (UTF): State Insolvency and
    11.                                                                                                Federal Loans to States. Congressional Research Service. February.
Isaacs, K.P., Whittaker, J.M., 2014b. Extending Unemployment Compensation Benefits                Yang, S., Lasky, M., Page, B., 2010. Policies For Increasing Economic Growth and
    During Recessions. Congressional Research Service report. October 14.                              Employment in 2010 and 2011. Congressional Budget Office. January.




                                                                                             13


File and source

File
u-s-unemployment-insurance-through-the-covid-19-crisis.pdf
Size
1,122,844 bytes
SHA-256
ba81179047ae74d867a1ba8efe184e35878be1280b36e6eb692c1e32ad742c0c
Our copy
u-s-unemployment-insurance-through-the-covid-19-crisis.pdf
Original
doi.org
Back to top