Poverty Reduction Is Not the Whole Story (2024)
Summary
A journal article, Poverty Reduction is Not the Whole Story: The COVID-19 Pandemic Response in Relation to Material Hardship, by Patrick Meehan and Trina Shanks of the School of Social Work, University of Michigan, published in the Journal of Family and Economic Issues (2024) 45:458–469. The authors use the Ypsilanti COVID-19 Study, a cross-sectional survey of 609 residents taken during the summer of 2020, and logistic regression models of bill-paying and food hardship. The abstract reports that disruptions to household finances, particularly job loss, significantly increased the likelihood of both hardships, that hardship type did not predict applying for SNAP or UI, and that UI was less accessible to low-income individuals experiencing hardship. The article reviews research on poverty measures and material hardship, sets out two research questions, and closes with its 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
Journal of Family and Economic Issues (2024) 45:458–469
https://doi.org/10.1007/s10834-023-09907-x
ORIGINAL PAPER
Poverty Reduction is Not the Whole Story: The COVID‑19 Pandemic
Response in Relation to Material Hardship
Patrick Meehan1 · Trina Shanks1
Accepted: 1 May 2023 / Published online: 16 May 2023
© The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2023
Abstract
As an absolute measure of deprivation poverty fails to capture the impact pandemic-related disruptions had on households.
In this study, we use data from the Ypsilanti COVID-19 Study, a cross-sectional survey of 609 residents taken during the
summer of 2020, to control for pandemic-related disruptions on bill-paying and food hardship. Using logistic regression
models in which specific forms of bill-paying (i.e. late paying rent, late paying utilities) and food hardships (i.e. eating
less over 7 days, worried food will run out) served as dependent variables, we find that disruptions to household finances,
particularly job loss, significantly increased the likelihood of experiencing bill-paying and food hardship, respectively. Our
study also controls for the type of hardship experienced to see which strategies households employed during the pandemic
to exit material hardship. Through logistic regression models on methods of exiting material hardship, we find the type of
hardship experienced was not predictive of applying for either SNAP or UI. Moreover, we find UI was less accessible to
low-income individuals experiencing hardship. The findings from our study elaborate the relationship between pandemic-
related disruptions and material hardship, and indicate to policymakers that preventing hardship in the first place is much
more meaningful to households than attempting to use policy to bring households out of hardship once they experience it.
Keywords COVID-19 · Material hardship · Poverty · Food insecurity · Rent · Utilities
Introduction O’Reilly (2020) observes that COVID-19 was not “an equal
opportunity pandemic” (p. 12). Rather, those impacted the
Almost three years on, there has been considerable research most were “the most marginalized” (p. 12). Gupta et al.
that the COVID-19 pandemic response worked well to keep (2020) found in their nationally representative survey of
families from falling into poverty. The combination of Eco- adults in September 2020 that four in ten families of color
nomic Impact Payments and the Pandemic Unemployment reported food insecurity in the prior 30 days, triple the rate
Compensation program not only kept families from falling of White families, and that more than three in ten families
into poverty (Han et al., 2020; Karpman & Zuckerman, of color worried about having enough to eat.
2021; Parolin et al., 2020), but actually increased house- Consequently, many observations on the policy response
hold incomes above replacement wages, particularly for to COVID-19 have focused not on poverty, which is an abso-
low-income families (Ganong et al., 2020). Material well- lute measure of well-being, but rather on material hardship.
being mirrored the policy response such that the expiration Heflin et al. (2009) conceive of material hardship as the abil-
of pandemic-related changes to Unemployment Insurance ity of households to meet basic necessities such as food,
(UI) in September 2020 led to increased food insecurity and shelter, and medical care. The traditional poverty measure
housing hardships (Cooney & Shaefer, 2021). ignores these forms of consumption. With respect to food,
Substantial variation existed in the pandemic experiences Schanzenbach (2020) defines food insecurity as the inability
of people of color and households with children at the time. to have “consistent, dependable access to enough food to live
an active, healthy lifestyle” (p. 2). Accordingly, Schanzen-
* Patrick Meehan bach and Pitts (2020) observe that in the early months of the
pjmeeh@umich.edu pandemic, food insecurity tripled among households with
children. Keith-Jennings et al. (2021) similarly observe food
1
School of Social Work, University of Michigan, Ann Arbor, hardship in Black and Latino households (see also Karpman
MI, USA
13
Vol:.(1234567890)
Journal of Family and Economic Issues (2024) 45:458–469 459
et al., 2020). In March 2021, these households were more of household consumption (Farrell et al., 2020; Han et al.,
than twice as likely as White adults to report that their 2020) that respond much more abruptly to changes in cir-
household did not get enough to eat in the last seven days. cumstance than can be captured from an annualized measure
These accounts suggest that far from holding hardship of poverty (Heflin, 2016; Heflin & Butler, 2012).
at bay, the policy response was inadequate to keep families In the early months of the pandemic there was a focus on
with children from suffering during the pandemic. Elliott changes in household circumstances was appropriate. For
et al. (2021) similarly find in their sample of mothers and example, Despard et al. (2020) observed in their nationally
grandmothers that COVID increased household expenses, representative sample from April to May 2020 that house-
particularly for those “with more people at home all day” holds experiencing COVID-related job loss were two to
(p. 4). three times more likely to experience hardships such as diffi-
In this study, we revisited the summer of 2020 to bet- culty paying for housing and putting off medical care. Using
ter understand how pandemic-related disruptions impacted data from the Well-Being and Basic Needs Survey, Karpman
material hardship for vulnerable households. We used data and Zuckerman (2021) identified that respondents whose
from the Ypsilanti COVID-19 Study to measure the impact households lost jobs during the pandemic were nearly three
of COVID-19-related disruptions on two specific types of times as likely to report problems paying utilities, and nearly
material hardship—bill-paying and food security—among four times as likely to report problems paying their rent or
an over sample of people of color, low-income households, mortgage than respondents who did not lose jobs. Schanzen-
and families with children. We further considered strate- bach (2020) separately found that among respondents to the
gies households experiencing hardship took to access the Household Pulse Survey from the U.S. Census Bureau, 21%
safety net or earn extra money. We found people of color, of those who lost jobs during COVID-19 reported not having
low-income households, and families with children were enough to eat.
indeed more likely to experience hardship, and that dis- Recognizing that the poverty measure, as designed, was
ruptions to household finances were more associated with unresponsive to the onset of the pandemic, several research-
food and bill-paying hardship than disruptions to health or ers attempted to perform updates in real time to demon-
household composition. And despite its fanfare as a bulwark strate pandemic-related deprivation and the responsiveness
against increased poverty during the pandemic (Ganong of public policy. Han et al. (2020) use a question on the
et al., 2020), we found that UI was less accessible to low- total cash income for the householder’s family for the previ-
income households experiencing hardship. The results of ous 12 months on the Basic Monthly Current Population
this study provide new insight on the impact of pandemic- Survey and compare that to official poverty thresholds to
related disruptions on material hardship that can inform the calculate the monthly poverty rate from January to Decem-
policy response in the future. Far from staving off hardship, ber 2020. Parolin et al. (2020) use the same data to cre-
the pandemic response in the U.S. left people of color and ate their own monthly poverty measure for October 2019
families with children hungry and unable to pay their rent to September 2020, based on projections of a family unit’s
and utilities. monthly resources. Parolin and colleagues use their derived
measure to show policies from the CARES Act—the Eco-
nomic Impact Payments and the Pandemic Unemployment
Literature Program—blunted a rise in poverty until the expiration of
the $600 weekly UI supplement at the end of July 2020.
To understand how the pandemic response prevented pov- Even these adjustments to poverty calculations, though,
erty but nevertheless contributed to hunger and hardship it obfuscate the disruption the pandemic caused families.
is necessary to explain the difference between poverty and Elliott et al. (2021) interview 54 mothers and grandmoth-
material hardship. The federal poverty measure calculates an ers in North Carolina to better understand their experience
“Economy Meal Plan” for a given family size multiplied by navigating the social safety net at the height of COVID-19.
three (Heflin et al., 2009, p. 747). Everyone under the thresh- They find nearly all of their subjects’ household bills had
old is poor while everyone over is not, regardless of the cost increased with more people at home, making the usual
of living, or the fact that groceries account for far less of a strategies for stretching food out for the month less effec-
household’s budget than was the case when the measure was tive. Mothers were the most unsung of all the heroes of
created in 1963. In this way, household income is treated the pandemic, according to O’Reilly (2020), who writes,
as an absolute barometer of deprivation. Material hardship, “no one is recognizing let along supporting mothers as
on the other hand, recognizes income as instrumental to frontline workers or acknowledging and appreciating what
deprivation, but that deprivation itself is experienced when mothers are managing and accomplishing in their homes
basic needs such as food, shelter, and medical care go unmet under unimaginable circumstances” (p. 12). Monte (2020)
(Heflin et al., 2009). These basic needs represent patterns uses data from the Household Pulse Survey to note that in
13
460 Journal of Family and Economic Issues (2024) 45:458–469
the early months of the pandemic adults living with chil- All of this is to suggest that each pandemic-related dis-
dren were more likely to report not having enough to eat ruption may not have contributed equally to each house-
than adults not living with children. In this context, SNAP hold hardship. Using data from the Survey of Income and
was much more helpful to families during the pandemic Program Participation, Heflin (2016) finds that job loss,
than the $600 UI supplement (Elliott et al., 2021; Hembre, controlling for other forms of disruption, predicts both
2020; Moffitt & Ziliak, 2020). bill-paying and food hardships. Given the eviction mora-
Parolin (2021) also observes that people of color can toria in place during the pandemic, though, job loss may
question the poverty-reducing accomplishments of the have had no relationship to late or missed rent payments
pandemic response. For example, “72% of families report (see Despard et al., 2020). Similarly, it may be an oversim-
that they also experienced food insufficiency prior to the plification to associate pandemic-related job loss with food
onset of the pandemic” (p. 1). Indeed, using Household insecurity (see Karpman & Zuckerman, 2021; Schanzen-
Pulse Survey data Keith-Jennings et al. (2021) find that bach, 2020) without controlling for additional factors such
one-fifth of children in Black and Latino households did as pandemic-related changes in health or household com-
not have enough to eat in the last seven days, three times position. No study has previously attempted to disentangle
the rate of White children. Pandemic-related jobs losses the impact COVID-related disruptions had on bill-paying
were highest among Latino households, while Black and food hardships, respectively.
households were significantly more likely than White In a similar fashion, each policy response to the pan-
households to have difficulty making housing and other demic may have distinct relationships to multiple hard-
bill payments (Despard et al., 2020). ships. Heflin and Butler (2012) find that the distinctiveness
Deprivations such as missed or delayed payments and of bill-paying and food hardships, respectively, influences
food insecurity in the spring and summer of 2020 are the strategies households use to exit them (see also Heflin
understood to be related to COVID-19. But the sheer vast- et al., 2011). Food hardships can be massaged, they find,
ness of the disruption the pandemic wrought—job losses, through the stretching of groceries and the foregoing of
household composition changes, health complications, meals in ways that bill-paying hardships cannot. In this
etc.—obscures precisely which disruptions were most way, the existence of eviction moratoria in the first year
meaningful to household hardships. Identifying the rela- of the pandemic may have allowed families to triage their
tionship between specific forms of disruption to specific bill-paying hardships in favor of food in ways not possible
forms of hardship is not a picayune exercise; it is vital to before the pandemic. This policy response may have been
our understanding of the effectiveness of the pandemic particularly helpful for households not otherwise eligible
response. for the expanded UI benefits.
Heflin and Butler (2012) speculate that bill-paying Given that the inaccessibility of expanded UI ben-
hardships and food hardships represent distinct constructs efits to low-income families was recognized at the time
that cannot be reduced to a unidimensional understanding (Han et al., 2020; Moffitt & Ziliak, 2020), caution is war-
of hardship. Food hardships are distinct from bill-paying ranted in ascribing too much credit for their ability to pull
hardships for a variety of reasons according to Heflin et al. households out of hardship. In other words, the success
(2009). First, more so than bill-paying, food consumption of expanded UI benefits may have depended on the spe-
is sensitive to income fluctuations. Second, food hardships cific hardships households faced among those eligible to
vary in their duration, from a few days to a few months receive them (see also Ganong et al., 2020).
in most cases. Third, food hardships can be concentrated
within different members of the household. Caregivers, for
example, can go without food in favor of their children in
a way that cannot be done with bill-paying. For their part, Research Questions
Heflin et al. suggest bill-paying hardships may be distinct
from food hardships because the processes resulting from Using Heflin’s (2016) multidimensional understanding of
an inability to pay basic bills are themselves unique. The material hardship, our first research question is as follows:
first missed payment, for example, either for rent or utilities, which pandemic-related disruptions were associated with
rarely results in catastrophic consequences, and households food and bill-paying hardships, respectively?
can negotiate their payments before the consequences begin Conceiving of these hardships as distinct experiences,
to escalate. In the early months of the COVID-19 pandemic our second research question is as follows: did those expe-
many states and the Centers for Disease Control had insti- riencing food or bill-paying hardship use distinct strategies
tuted eviction moratoria. This may have given households to exit hardship?
the flexibility to relax their bill-paying priorities in favor of
food.
13
Journal of Family and Economic Issues (2024) 45:458–469 461
Data and Methods rate of 13.1%. The response rate for the rest of the sam-
ple is harder to calculate because the research team was
Our data came from the Ypsilanti COVID-19 Study, a unaware of the number of potential respondents on the
cross-sectional needs assessment of residents taken in various non-profit listservs that emailed invitations. The
the summer of 2020. A local human service organization average age of respondents in the entire sample was 37.7
spearheaded the project, which eventually came to include (SD 11.1). The sample was 74.3% female, 51.8% White,
the Washtenaw County Racial Equity Office, and faculty 33.4% Black, 11.6% Latinx, and 56.7% of respondents had
of a research university in the area. The human service household incomes below $40,000 in 2019. The character-
organization provided direct service to residents of the istics of the sample over fit for race and income compared
Ypsilanti Housing Commission, the city’s public housing to the Ypsilanti population in general, which in 2020 was
authority. The organization had a strong desire to know the 26.5% Black and 6.0% Latinx, respectively, with a median
prevalence of COVID-19 among residents and additional household income of $40,028 (U.S. Census Bureau, 2020).
services they may need at a time when the city lacked a
COVID-19 testing center. Members of the partnership col- Pandemic‑Related Disruptions
laborated on the survey instrument and considered many
ways to measure not only the presence of COVID-19, but Heflin (2016) identifies disruptions in employment and
its impact on individuals and families. The language for income, household composition and residence, and health
many of the survey items was taken from the Detroit Metro as potential contributors to a multidimensional understand-
Areas Communities Study, which was concurrently field- ing of material hardship. The COVID-19 pandemic touched
ing a rapid-response survey to Detroit area residents on each of these forms of disruption, and each was a focus of
similar topics. The involvement of the Washtenaw County the survey instrument.
Racial Equity Office was particularly valuable in crafting
items that spoke to racial disparities in the ability to work Housing Disruptions
from home, knowledge of individuals with COVID-19,
attitudes toward the (eventual) vaccines, as well as the One item was used to indicate disruptions to housing cir-
responsiveness of elected officials to the pandemic. In cumstances and household composition. It read, “Since the
the immediate aftermath of George Floyd’s murder, the COVID-19 pandemic began in the U.S. (around March 1,
partnership also included survey items on attitudes toward 2020), have any of the following changes taken place in your
law enforcement. Because no identifying information of living arrangements (select all that apply)?” Answer choices
respondents was collected, the survey instrument and data included (1) I moved to a new place to live, (2) someone
collection procedures were deemed exempt by the institu- moved into my household, (3) someone moved out of my
tional review board of the participating university. household, (4) no changes. Each of these answer choices was
The survey instrument was administered through Qual- dichotomized as unique indicators of housing disruptions
trics. To make it as accessible to as many residents as pos- related to COVID-19.
sible, the research partnership created a Spanish-language
version of the instrument that respondents could elect to Health Disruptions
answer.
Respondents to the survey were a non-probability sam- Three items were used to indicate disruptions to health. The
ple of Ypsilanti residents. The research partnership met first asked respondents if they had been tested for COVID-
with representatives of several non-profit organizations 19. Testing at the time of data collection was not easily
serving the Ypsilanti community, who agreed to email an accessible in Ypsilanti, so individuals to be tested were most
invitation to participate in the study to their respective likely experiencing COVID-related symptoms. The second
listservs. Residents of the Ypsilanti Housing Commission asked respondents if a friend or family member had COVID-
also received an email invitation to participate. In adher- 19. The third asked respondents if they knew someone who
ence to equitable research practice, respondents received died from COVID-19. These two questions indicated some
a $20 incentive for completing the survey. degree of exposure to COVID that may have impacted the
Data collection occurred between June 12 and August individual enough to change their behavior.
21, 2020. A total of 676 surveys were started. Of these,
there were 609 usable responses. There were approxi- Financial Disruptions
mately 1,000 Ypsilanti Housing Commission residents
during the period of data collection. We received 131 Three items were used to indicate disruptions to household
responses from individuals in this group, for a response finances. The first asked respondents to indicate their cur-
rent employment status. Their answer choices were (1) I am
13
462 Journal of Family and Economic Issues (2024) 45:458–469
still working in the same job I had before the COVID-19 the pandemic’s potential impact on food insecurity in the
pandemic, (2) I lost my job and am looking for work, (3) short term.
I have been temporarily laid off from the job I had before
the COVID-19 pandemic, (4) I am on sick leave or other Exit Strategies
leave from the job I had before the COVID-19 pandemic,
(5) I am now working at a different job than I had before the Despite being a cross-sectional survey, data collection for
COVID-19 pandemic, (6) other. Our analysis dichotomized the Ypsilanti COVID-19 Study occurred at such a time
those who had lost their job and were looking for work, and (June–August 2020) that respondents could reasonably be
all other respondents. The second item asked respondents, expected to (1) accurately recall their circumstances before
“Has the COVID-19 pandemic affected how you are spend- the pandemic (around March 1, 2020), as well as to (2) be
ing money?” Respondents could answer (1) Yes, I have been aware of and utilize pandemic-related programs passed
spending a lot more money, (2) Yes, I have been spending a under the CARES Act such as the Economic Impact Pay-
little more money, (3) Yes, I have been spending a little less ments and the Pandemic Unemployment Compensation.
money, (4) Yes, I have been spending a lot less money, or
(5) No, I have not changed the amount I spend. Our analysis Social Program Participation
dichotomized those who indicated their spending increased
a little or a lot, and all other respondents. The third item was Heflin et al. (2011) conceive of social programs such as
identical to the second, except that it asked about borrowing. those passed under the CARES Act as potential exit strate-
Similarly, our analysis dichotomized those who indicated gies from hardship. One item was used to indicate social
their borrowing had increased a little or a lot, and all other program participation as an exit strategy from material hard-
respondents. ship. It reads, “Since March 1, 2020, have you received any
of the following forms of public assistance or emergency
Bill‑Paying Hardship safety net services (select all that apply)?” Respondents
had 11 answers choices, among which were SNAP or Food
One item was used to identify bill-paying hardship modeled Stamps, and Unemployment Insurance (UI) benefits. Each of
after Heflin et al. (2009). It reads, “Since March 1, 2020, these was dichotomized to indicate their participation since
have you been late paying for any of the following (select the start of the pandemic.
all that apply).” Answer choices included (1) mortgage or
rent, (2) loans (e.g. student loans, car loan), (3) credit card, Individual Strategies
(4) utility or water bill, (5) phone, internet, cable, (6) I have
not been late with any payments. We recorded respond- Heflin et al. (2011) also recognized idiosyncratic strategies
ents who chose mortgage or rent or utility or water bill as to exit material hardship that need to be accounted for in
experiencing utility hardship so that our analysis would be relation to social program participation. One item was used
consistent with previous research on material hardship, and to indicate individual strategies to exit material hardship.
because rent and utilities were qualitatively more important It reads, “Since the start of the pandemic (around March 1,
to respondents than the other answer choices. 2020), have you been paid for any of the following activities
(select all that apply)?” There were 7 answer choices avail-
able to respondents. Our analysis considers three: (1) child
Food Hardship
or elder care services, (2) housecleaning, and (3) driving or
ride-sharing such as Uber or Lyft. Each of these was dichot-
Two items were used to identify hardships related to food
omized to indicate their use since the start of the pandemic.
insecurity. The first asked respondents, “In the past seven
Table 1 shares the characteristics of respondents to the
days, were you worried you would run out of food because
Ypsilanti COVID-19 Study.
of a lack of money or other resources?” The second asked
The analysis strategy for Research Question 1 followed
respondents, “In the past seven days, did you eat less than
from Heflin and Butler (2012):
you thought you should because of a lack of money or other
resources?” For both questions, respondents could answer (
h
)
log = 𝛽0 + 𝛽1 HSDi + 𝛽2 HLTH i + 𝛽3 FIN i + 𝛽4 Xi + ei
yes, no, or don’t know. Respondents were identified as expe- 1−h i
riencing food hardship if they answered yes to either ques- ( )
tion. This approach differs from Heflin and Butler (2012) where log 1−h h
is the likelihood of experiencing material
i
who consider the volume of food consumed over the last hardship (late paying rent, late paying utilities, eating less,
12 months. Framing food hardship consistently with Heflin or worrying about food) for individual i ; HSDi is a vector of
and Butler, as well as Heflin et al. (2009), would obfuscate variables capturing housing disruptions (i.e. moved to new
13
Journal of Family and Economic Issues (2024) 45:458–469 463
Table 1 Sample characteristics (N = 609) Table 1 (continued)
Baseline characteristic % The average age of the sample was 37.7, with a standard deviation of
11.1
Ypsilanti Housing Commission resident 26.1
Disruptions to housing circumstances
residence since COVID, another person moved in since
Moved to new residence 8.0
COVID, another person moved out since COVID); HLTH i
Another person moved in 5.7
is a vector of variables capturing disruptions to health (i.e.
Another person moved out 4.6
tested for COVID, friend or family member had COVID, and
Disruptions to health
Tested for COVID-19 31.7
know someone who died from COVID); FIN i is a vector of
Friend or family member had COVID-19 39.9 variables capturing disruptions to financial circumstances
Know someone who died from COVID-19 18.5 (i.e. lost job since COVID, spending since COVID, and bor-
Disruptions to household finances rowing since COVID), and Xi is a vector of demographic
Lost job since COVID-19 25.9 variables (i.e. housing income, rent current residence, can
Spending increased since COVID-19 41.7 afford a $400 unexpected expense, age, race, any children in
Borrowing increased since COVID-19 23.3 the household, and overall health).
Demographics For Research Question 2 we modeled five different meth-
Female 74.3 ods for exiting material hardship, conditional on experienc-
White 51.8 ing hardship in the first set of models:
Black 33.4
( )
Latinx 11.6 y
log =𝛽0 + 𝛽1 Hi + 𝛽2 HSDi + 𝛽3 HLTH i
All other races 3.2 1−y i
Can afford $400 unexpected expense 56.3
+ 𝛽4 FIN i + 𝛽5 Xi + ei
Rent current residence 57.3
Any children in the household 69.3
( )
where log y
is the likelihood of attempting one of the
Household income in 2019 1−y i
$0–$9,999 13.7 five exit strategies considered— applying for SNAP or UI,
$10,000–$19,999 12.2 and three different ways of earning extra money—and Hi is
$20,000–$29,999 13.8 a vector of variables related to the type of hardship
$30,000–$39,999 17.0 experienced.
$40,000–$49,999 12.7
$50,000–$74,999 14.6
$75,000 or more 16.3
Results
Overall health
Excellent 15.4
Very good 45.2
Pandemic‑Related Disruptions and Material
Average 28.4
Hardship
Fair 8.4
Poor 2.6 Table 2 shows the relationship of pandemic-related disrup-
Type of hardship tions to bill-paying and food hardship, respectively. Of the
No hardship 49.4 various disruptions COVID-19 put households through,
Bill-paying and food 23.6 disruptions to household finances had the most consistent
Bill-paying only 12.8 relationship to both bill-paying and food hardships. The
Food only 14.1 loss of one’s job since the start of the COVID-19 pandemic,
Bill-paying hardship indicators for example, significantly increased the likelihood of being
Late paying rent since COVID-19 21.2 late with rent ( 𝛽 = 1.055, p < 0.000), being late with utili-
Late paying utilities since COVID-19 26.8 ties ( 𝛽 = 0.651, p < 0.007), eating less over the last 7 days
Food hardship indicators
(𝛽 = 0.753, p < 0.001), and worrying you might run out of
Did you eat less last 7 days to save food? 30.9
food ( 𝛽 = 0.618, p < 0.014), controlling for all other dis-
Worried food will run out last seven days? 30.5
ruptions and demographic characteristics. This finding was
Hardship exit strategies
consistent with research from the early months of the pan-
Applied for SNAP benefits 38.7
Applied for UI benefits 23.3
demic (Despard et al., 2020; Karpman & Zuckerman, 2021),
Earned money for childcare since COVID 7.7 but with the benefit of controlling for additional forms of
Earned money for housecleaning since COVID 8.7 disruption.
Earned money for ridesharing since COVID 7.1
13
464 Journal of Family and Economic Issues (2024) 45:458–469
Table 2 Logistic regression results for pandemic-related disruptions on food and bill-paying hardships
Variable Late paying rent Late paying utilities Eat less last 7 days Worried about food
𝛽 SE p 𝛽 SE p 𝛽 SE p 𝛽 SE p
Disruptions to housing circumstances
Moved to new residence since COVID 0.082 0.412 0.842 0.069 0.389 0.859 0.362 0.383 0.345 − .557 0.419 0.184
Another person moved in since COVID − .408 0.533 0.444 − 1.31 0.612 0.032 − .688 0.547 0.208 − .298 − .531 0.574
Another person moved out since COVID − 1.58 0.784 0.044 0.176 0.497 0.723 − .063 0.488 0.897 − .381 0.528 0.470
Disruptions to health
Tested for COVID 0.546 0.275 0.047 − .085 0.264 0.748 0.161 0.258 0.533 0.934 0.261 0.000
Friend or family member had COVID 0.277 0.272 0.307 0.184 0.259 0.476 − .196 0.249 0.432 0.191 0.262 0.468
Know someone who died from COVID 0.125 0.312 0.689 0.105 0.299 0.726 − .225 0.308 0.467 − .305 0.310 0.325
Disruptions to household finances
Lost job since COVID 1.055 0.254 0.000 0.651 0.242 0.007 0.753 0.236 0.001 0.618 0.251 0.014
Spending since COVID 0.211 0.092 0.021 0.282 0.084 0.001 0.143 0.082 0.082 0.285 0.086 0.001
Borrowing since COVID 0.277 0.121 0.022 0.032 0.112 0.776 0.182 0.113 0.107 0.351 0.119 0.003
Demographics
Household income 0.076 0.086 0.374 − .186 0.079 0.018 − .250 0.078 0.001 − .181 0.081 0.025
Rent current residence 0.670 0.324 0.039 0.101 0.293 0.731 0.224 0.289 0.438 0.532 0.306 0.083
Can afford a $400 unexpected expense 0.261 0.282 0.355 − .215 0.255 0.399 0.109 0.255 0.668 − .048 0.266 0.857
Age 0.011 0.012 0.362 0.012 0.012 0.304 − .002 0.011 0.845 0.010 0.012 0.411
Black 0.802 0.307 0.009 0.643 0.274 0.019 − .304 0.279 0.276 0.269 0.286 0.347
Latinx 1.236 0.375 0.001 0.965 0.356 0.007 1.513 0.348 0.000 1.759 0.368 0.000
All other races 0.452 0.712 0.526 − 1.62 1.066 0.129 − .564 0.697 0.418 0.051 0.657 0.938
Any children in the household 0.235 0.318 0.459 1.314 0.321 0.000 0.704 0.281 0.012 1.011 0.312 0.001
Overall health − .106 0.146 0.468 0.025 0.134 0.851 0.261 0.129 0.043 − .054 0.138 0.695
N 521 521 515 514
Bold values indicate p < .05
Pandemic-related increases in spending was also related In contrast to disruptions in household finance, disrup-
to food and bill-paying hardship. It significantly increased tions in housing circumstances decreased the likelihood
the likelihood of being late with rent (𝛽 = 0.211, p < 0.021), of experiencing bill-paying hardship. For example, having
being late with utilities (𝛽 = 0.282, p < 0.001), and worrying another individual in the household move out significantly
you might run out of food (𝛽 = 0.285, p < 0.001), controlling decreased (𝛽 = − 1.58, p < 0.044) the likelihood of being late
for other factors, including the loss of one’s job. This finding with rent, controlling for other factors. Having an individual
was consistent with Elliott and colleague’s (2021) sample of move into the household significantly decreased (𝛽 = − 1.31,
mothers and grandmothers in North Carolina, and suggests p < 0.032) the likelihood of being late with utilities. Heflin
the pandemic interfered with established spending patterns (2016) found that the fluidity of household composition had
in ways that adversely affected both bills and food. different effects on hardship, and “may be a strategy that
Our last measure of disruptions in household finances low-income households use in varying ways to meet compet-
concerned increased borrowing. It similarly cut across food ing demands” (p. 369).
and bill-paying hardship. Pandemic-related increases in bor- Finally, we considered disruptions to health related to
rowing significantly increased the likelihood of being late COVID-19. Of our three indicators, only testing for COVID-
with rent (𝛽 = 0.277, p < 0.022) and worrying you might run 19 showed a relationship to material hardship. It was associ-
out of food (𝛽 = 0.351, p < 0.003), controlling for other fac- ated with an increased likelihood of being late with rent (𝛽 =
tors. That increased borrowing should touch both bill-pay- 0.546, p < 0.047), and worrying food will run out (𝛽 = 0.934,
ing and food hardship may suggest these individuals were p < 0.000), controlling for other factors. Our other measures
in more dire circumstances than other respondents. With of health disruptions were admittedly indirect, so it was
increased borrowing comes increased debt, which may have understandable these did not show a relationship to mate-
been a measure of last resort after having already been late rial hardship. Testing’s relationship to material hardship may
with rent and worrying you might run out of food. suggest these individuals were genuinely sick, or that their
13
Journal of Family and Economic Issues (2024) 45:458–469 465
circumstances required them to go out more because they the CARES Act did not address the issues that triggered the
had less ability to stay at home and wait out the pandemic. hardship in the first place (Heflin & Butler, 2012).
Importantly, certain demographic characteristics that Disruptions in housing circumstances were associated
were constant throughout the pandemic showed strong asso- with multiple strategies to exit material hardship, although
ciations with food and bill-paying hardship, controlling for in conflicting ways. For example, moving to a new resi-
changes in circumstances. For example, higher household dence ( 𝛽 = − 1.74, p < 0.001) and having an individual
incomes in 2019 significantly decreased the likelihood of move out of the household significantly decreased ( 𝛽 =
being late with utilities ( 𝛽 = − 0.186, p < 0.018), eating less − 1.80, p < 0.038) the likelihood of receiving SNAP. On the
over the last 7 days (𝛽 = − 0.250, p < 0.001), and worrying other hand, those same disruptions were associated with an
food will run out ( 𝛽 = − 0.181, p < 0.025), controlling for increased likelihood of earning money from childcare ( 𝛽
other factors. Race was also predictive of material hardship = 1.88, p < 0.001 for moving to a new residence, 𝛽 = 2.14,
entry. Black respondents were associated with an increased p < 0.033 for an individual moving out of the household).
likelihood of being late with rent (𝛽 = 0.802, p < 0.009) and Also, moving to a new residence was also associated with an
being late with utilities ( 𝛽 = 0.643, p < 0.019), controlling increased likelihood of earning money from housekeeping (𝛽
for other factors. Latinx respondents were associated with = 1.29, p < 0.044). These results may be an indication that
an increased likelihood of being late with rent (𝛽 = 1.236, for the respondents in our sample who experienced disrup-
p < 0.001), being late with utilities ( 𝛽 = 0.965, < 0.007), tions in housing circumstances, these disruptions may have
eating less over the last 7 days ( 𝛽 = 1.513, p < 0.000), and been conditional on their performing these additional tasks
worrying food will run out ( 𝛽 = 1.011, p < 0.001). These for pay, perhaps at a relative or friend’s residence.
findings were consistent with Karpman et al. (2020), who Disruptions in household finances were not predictive of
found Latino households to be the most disadvantaged early receiving SNAP benefits. It may be that a substantial num-
in the pandemic. Lastly, it should be noted that having chil- ber of our Ypsilanti Housing Commission respondents were
dren was also a source of bill-paying and food hardship. already receiving SNAP prior to the onset of the pandemic
Having children was associated with an increased likelihood and elected not to answer that item affirmatively on the
of being late with utilities (𝛽 = 1.314, p < 0.000), eating less survey instrument. This would be consistent with Parolin
over the last 7 days ( 𝛽 = 0.704, p < 0.012), and worrying (2021). Increased spending was associated with an increased
food will run out (𝛽 = 1.011, p < 0.001), controlling for other likelihood ( 𝛽 = 0.437, p < 0.027) of earning money from
factors. These results were consistent with prior research childcare. Table 3 also showed that losing one’s job did sig-
from the early months of the pandemic (Elliott et al., 2021; nificantly increase the likelihood of receiving UI (𝛽 = 2.13,
Gupta et al., 2020; Keith-Jennings et al., 2021; Monte, 2020; p < 0.000) for the respondents in our sample.
Schanzenbach & Pitts, 2020) and spoke to the challenges Our results showed that the likelihood of receiving UI
families with children faced at this time. since the start of the pandemic significantly increased (𝛽 =
0.298, p < 0.028) as household income increased, controlling
Hardship and Exit Strategies for other factors. This was consistent with research from
early in the pandemic that the expanded UI benefits were less
Table 3 shows the results of our models of exit strategies accessible to low-income individuals who had lost jobs (Han
from material hardship. There was very little to suggest from et al., 2020; Moffitt & Ziliak, 2020). The opposite relation-
our sample of Ypsilanti residents that the type of hardship ship was found for SNAP. There, the likelihood of receiving
faced in the early months of the pandemic was predictive of SNAP significantly decreased ( 𝛽 = − 0.237, p < 0.047) as
an individual’s exit strategy. Indeed, bill-paying hardship household income increased. Black respondents were also
significantly decreased (𝛽 = − 1.68, p < 0.045) the likelihood significantly more likely ( 𝛽 = 0.879, p < 0.021) to receive
of earning money from Uber or Lyft. Food hardship was not SNAP since March 1, 2020, as were Latinx respondents (𝛽
predictive of any exit strategy. Heflin and Butler (2012) simi- = 2.24, p < 0.000), and respondents with children (𝛽 = 2.01,
larly found difficulty predicting exit strategies based on the p < 0.000), controlling for other factors. These findings were
type of hardship faced. They speculate that households expe- all consistent with research that found SNAP to be the most
riencing food hardship alone were unlikely to recover due to responsive safety net program during the pandemic (Hem-
poor household functioning. Heflin (2016) further found that bre, 2020; Moffitt & Ziliak, 2020).
once a disruption occurs, households often “lack resilience
and have difficulty regaining equilibrium” (p. 370). In the
early months of the pandemic, the lack of a clear pathway
out of hardship once it was experienced suggested the Eco-
nomic Impact Payments and the expanded UI benefits under
13
466
13
Table 3 Logistic regression results on strategies to exit material hardship
Variable Applied for SNAP Applied for UI Childcare Housecleaning Uber/Lyft
𝛽 SE p 𝛽 SE p 𝛽 SE p 𝛽 SE p 𝛽 SE p
Type of hardship
Bill-paying only − .641 .391 .101 .542 .405 .181 .443 .601 .461 − .622 .609 .307 − 1.68 .841 .045
Food only − .550 .389 .158 − .341 .470 .468 .975 .595 .102 − .139 .544 .799 − .157 .520 .762
Disruptions to housing circumstances
Moved to new residence since COVID − 1.74 .514 .001 .047 .564 .934 1.88 .575 .001 1.29 .642 .044 .368 .703 .601
Another person moved in since COVID − 1.33 .730 .068 − .502 .777 .518 1.23 .908 .175 .738 .841 .380 .311 .877 .722
Another person moved out since COVID − 1.80 .870 .038 − 2.10 1.19 .077 2.14 1.00 .033 1.07 1.05 .309 1.43 .966 .139
Disruptions to health
Tested for COVID − .752 .364 .039 − .532 .392 .175 .952 .522 .068 .405 .496 .414 .034 .558 .951
Friend or family member had COVID .211 .346 .540 .417 .375 .266 − .889 .533 .095 − .299 .514 .560 − 1.78 .633 .005
Know someone who died from COVID .688 .436 .114 − .200 .436 .646 .956 .522 .068 − .894 .649 .169 .592 .636 .352
Disruptions to household finances
Lost job since COVID − .527 .335 .116 2.13 .381 .000 − .012 .494 .979 .882 .464 .057 .335 .459 .465
Spending since COVID .016 .113 .884 − .030 .122 .803 .437 .198 .027 .264 .173 .127 .278 .186 .136
Borrowing since COVID .108 .138 .436 .034 .149 .816 .103 .195 .597 − .295 .193 .126 − .028 .202 .890
Demographics
Household income − .237 .119 .047 .298 .135 .028 .258 .188 .169 .424 .191 .027 .311 .192 .106
Rent current residence .361 .407 .375 − .574 .461 .213 .725 .662 .274 .304 .583 .602 .415 .526 .430
Can afford a $400 unexpected expense − .349 .343 .309 − .456 .398 .252 − .898 .527 .088 1.44 .577 .013 1.16 .544 .032
Age − .023 .019 .214 − .036 .021 .088 .024 .029 .411 .024 .034 .469 .052 .031 .096
Black .879 .380 .021 .967 .447 .031 − .519 .607 .392 .366 .583 .530 .295 .644 .647
Latinx 2.24 .510 .000 .254 .562 .651 − .565 .749 .450 − .962 .690 .164 1.75 .657 .008
All other races .320 1.02 .754 .709 1.29 .585 − − − − − − 1.56 1.44 .279
Any children in the household 2.01 .496 .000 − .766 .454 .092 1.21 .906 .181 1.28 .915 .161 1.05 .924 .255
Overall health .129 .189 .495 − .067 .222 .760 − .326 .308 .289 − .468 .319 .143 .522 .296 .078
N 262 262 257 257 262
Bold values indicate p < .05
Journal of Family and Economic Issues (2024) 45:458–469
Journal of Family and Economic Issues (2024) 45:458–469 467
Discussion disruptions to household finances, perhaps through an
American version of the U.K.’s Coronavirus Job Reten-
The notion of pandemic amnesty has been floated to absolve tion Scheme, in which the central government subsidized
policymakers of all the things unknown at the onset of the cost to furlough workers at home rather than lay them
COVID-19 in order to render a constructive conversation off entirely (Brewer & Tasseva, 2021). In the U.S., laying
on how to plan for future pandemics (Oster, 2022). Among workers off and then making UI more generous was not
the unknowns in the early months of the pandemic was the an effective method for pulling the most vulnerable out
necessity of masking. A similar unknown was what to do of pandemic-related material hardship (see also Moffitt
for all the workers required to stay home in the spring of & Ziliak, 2020).
2020 so that we could “flatten the curve”. Among the poli- Finally, the disproportionate impact of pandemic-
cies to emerge out of this period were the Economic Impact related disruptions to households of color and families
Payments and the Pandemic Unemployment Compensation with children was unknown to policymakers in the spring
program. These have been heralded as having prevented of 2020, although Parolin (2021) persuasively argues this
a massive increase in poverty (Cooney & Shaefer, 2021; should not have been the case. We found that control-
Ganong et al., 2020; Karpman & Zuckerman, 2021). The ling for pandemic-related disruptions, households of
spirit of pandemic amnesty should extend not only to these color were more likely to experience bill-paying and
programs in particular, but also to the consensus regarding food hardship in the early months of the pandemic than
their effectiveness at preventing poverty. White households in our sample. In addition, we found
In absolute terms the pandemic response may have pre- that households with children were more likely to expe-
vented poverty, but in this study, we have considered how rience bill-paying and food hardship. COVID-19 was
poverty as an absolute measure was inadequate for under- not, as others have noted, an equal opportunity disruptor
standing the impact pandemic-related disruptions had on (O’Reilly, 2020). In comparison to UI, the importance
vulnerable households. We should rather view pandemic- of SNAP may have been unknown to policymakers in
related disruptions in early 2020 through the prism of mate- the spring of 2020. As others have noted (Elliott et al.,
rial hardship, which refers to a household’s ability to meet 2021), SNAP was the clear strategy out of hardship for
basic expenses such as rent, utilities, food, and medical care households with children in our sample, more so than the
(Heflin et al., 2009). Viewed from this perspective, there is expanded UI benefits.
consensus that the pandemic itself was quite harmful to low-
income households, households of color, and families with
children (see Gupta et al., 2020; Karpman & Zuckerman, Limitations
2021; Schanzenbach & Pitts, 2020). Our study went further
than previous considerations of pandemic-related material The Ypsilanti COVID-19 Study was a non-probability
hardship by showing which disruptions were associated with sample of residents of one municipality, at one point in
bill-paying and/or food hardship. We then conceptualized time in the summer of 2020. Although the sample was over
variation in how these hardships were experienced to explore representative of people of color and low-income indi-
their relationship to strategies to exit hardship. viduals in Ypsilanti, it was not generalizable to the city
We found many insights that were unknown to policy- as a whole, nor to the United States. Nevertheless, many
makers at the onset of the pandemic. For example, dis- of the descriptive statistics in the Ypsilanti COVID-19
ruptions to household finances, particularly through job Study mirrored those of nationally representative studies
loss, cut across bill-paying and food hardships, control- fielded around the same time, particularly concerning dis-
ling for other forms of pandemic-related disruption. And ruptions to households of color and families with children
yet once a household experienced either bill-paying or (Despard et al., 2020; Karpman et al., 2020). These simi-
food hardship, there was no clear path out of the hardship larities added concurrent validity to the Ypsilanti COVID-
through SNAP or UI. Along the same lines, we found 19 Study.
that among those experiencing hardship—bill-paying, The indicators of health disruptions used in this study
food, or both—income was predictive of UI. In other were indirect and showed little relationship to food or
words, those most likely to benefit from the extraordinary bill-paying hardship. It may be that more direct indica-
generosity of the Pandemic Unemployment Compensa- tors of pandemic-related health complications will show
tion program were less likely to access it, such as low- significant relationships to multiple dimensions of material
income households behind on their rent or worried their hardship.
food might run out (see also Ganong et al., 2020). In the Finally, this study was making inferences on the rela-
future, pandemic response should prioritize preventing tionships between pandemic-related disruptions and multi-
ple dimensions of material hardship while simultaneously
13
468 Journal of Family and Economic Issues (2024) 45:458–469
measuring the relationships between material hardship Declarations
and strategies to exit hardship. Conceptually, this ana-
lytic strategy may be more appropriately applied to a Compliance with Ethical Standards This study was provided IRB ex-
emption from the University of Michigan. Although it did involve re-
panel study than a cross-sectional one like the Ypsilanti search on human subjects, no identifying information was collected,
COVID-19 Study. However, the rapidly evolving nature and the incentive participants received for completing the survey in-
of the pandemic response argued in favor of our data. The strument–$20—was not coercive. Participants were informed prior to
Pandemic Unemployment Compensation program expired consenting to participate that information provided would be used for
research purposes, as well as information to contact the study team
at the end of July 2020. Fielding a panel study to capture in with questions.
one instance the impact of COVID disruptions, then field-
ing another panel to capture the impact of the pandemic Conflict of interest The study team has no conflicts of interest to dis-
close.
response was impractical. We were fortunate that our
period of data collection allowed for the accurate recall of
respondents when programs passed under the CARES Act
were still in operation. References
Brewer, M., & Tasseva, I. V. (2021). Did the UK policy response
to Covid-19 protect household incomes? The Journal of Eco-
Conclusion and Implications nomic Inequality, 19, 433–458. https:// d oi. o rg/ 1 0. 1 007/
s10888-021-09491-w
Although three years removed from spring 2020, it is Cooney, P., & Shaefer, H. L. (2021). Material hardship and mental
health following the COVID-19 relief bill and American rescue
important to remember that the COVID-19 pandemic plan act. Poverty Solutions. University of Michigan. Retrieved
created profound disruptions to already vulnerable indi- December 28, 2022, from http://poverty.umich.edu/publications/
viduals and families with children. These disruptions mater ial-hardship-and-mental-health-following-the-covid-19-
relief-bill-and-american-rescue-plan-act/
impacted their ability to pay their bills and provide food
Despard, M., Grinstein-Weiss, M., Chun, Y., & Roll, S. (2020).
for the households. These are properly understood as dis- COVID-19 job and income loss leading to more hunger and
tinct hardships that can be glossed over when the pan- financial hardship. Brookings Institution. Retrieved December
demic response is viewed from the perspective of poverty. 28, 2022, from http://brookings.edu/blog/up-front/2020/07/13/
covid-19-job-and-income-loss-leading-to-more-hunger-and-finan
In absolute terms the pandemic response helped families
cial-hardship/
avoid poverty, but pandemic-related disruptions such as Elliott, S., Satterfield, S. J., Solorzano, G., Bowen, S., Hardison-
jobs loss did throw them into hardship. And once faced Moody, A., & Williams, L. (2021). Disenfranchised: How lower
with hardship, it is not clear the pandemic response pro- income mothers navigated the social safety net during the COVID-
19 pandemic. Socius, 7. https://doi.org/10.1177/2378023121
vided pathways for families out of hardship.
1031690
In the future, a viral pandemic is not the only disrup- Farrell, D., Ganong, P., Greig, F., Liebeskind, M., Noel, P., & Vavra,
tion imaginable to vulnerable families with children. In J. (2020). Consumption effects of unemployment insurance dur-
reality many things can cause a disruption such as a plant ing the COVID-19 pandemic. JP Morgan Chase & Co. Institute.
Retrieved December 28, 2022, from https://papers.ssrn.com/sol3/
closure, or a water crisis as with Flint, Michigan, or Jack-
papers.cfm?abstract_id=3654274
son, Mississippi, or an unexpected medical emergency. Ford School. (n.d.) Detroit metro area communities study. University
As Parolin (2021) observes, many of our most vulner- of Michigan. Retrieved December 28, 2022, from http://fords
able families live on a knife’s edge avoiding hardship chool.umich.edu/research-centers-depar tments/detroit-metro-
area-communities-study
when a disruption can hit of unknown duration. Poverty
Ganong, P., Noel, P., & Vavra, J. (2020). U.S. unemployment replace-
as an absolute measure will fail to capture the nuance of ment rates during the pandemic. Journal of Public Economics,
a household’s experience with disruptions and provide 191, 1–12. https://doi.org/10.1016/j.jpubeco.2020.104273
policymakers accurate information as to the disruption’s Gupta, P., Gonzalez, D., & Waxman, E. (2020). Forty percent of Black
and Hispanic parents of school-age children are food insecure.
impact. Our findings suggest when disruptions hit, poli-
Urban Institute. Retrieved December 28, 2022, from http://urban.
cymakers should focus on preventing job loss. This prior- org/research/publication/forty-percent-black-and-hispanic-paren
ity is likely to limit bill-paying and food hardships that ts-school-age-children-are-food-insecure
follow in the immediate aftermath of the disruption. The Han, J., Meyer, B., & Sullivan, J. (2020). Income and poverty in the
COVID-19 pandemic. NBER Working Paper Series. Working
alternative, allowing job loss to occur and using the UI
Paper 27729. Retrieved December 28, 2022, from http://www.
system to make families whole, is not an effective way of nber.org/papers/w27729
pulling households out of hardship once it hits. Heflin, C. (2016). Family instability and material hardship: Results
from the 2008 Survey of Income and Program Participation. Jour-
nal of Family and Economic Issues, 37, 359–372. https://doi.org/
10.1007/s10834-016-9503-6
13
Journal of Family and Economic Issues (2024) 45:458–469 469
Heflin, C. H., & Butler, J. S. (2012). Why do women enter and exit O’Reilly, A. (2020). “Trying to Function in the Unfunctionable”:
from material hardship? Journal of Family Issues, 34(5), 631–660. Mothers and COVID-19. Journal of the Motherhood Initiative
https://doi.org/10.1177/0192513X12442822 for Research and Community Involvement, 11(1). Retrieved from
Heflin, C., London, A. S., & Scott, E. K. (2011). Mitigating material https://jarm.journals.yorku.ca/index.php/jarm/article/view/40588
hardship: The strategies low-income families employ to reduce the Oster, E. (2022). Let’s declare a pandemic amnesty. The Atlantic.
consequences of poverty. Sociological Inquiry, 81(2), 223–246. Retrieved December 28, 2022, from http://theatlantic.com/ideas/
https://doi.org/10.1111/j.1475-682X.2011.00369.x archive/2022/10/covid-response-foregiveness/671879/
Heflin, C., Sandberg, J., & Rafail, P. (2009). The structure of material Parolin, Z. (2021). What the COVID-19 pandemic reveals about racial
hardship in U.S. households: An examination of the coherence differences in child welfare and child well-being: An introduction
behind common measures of well-being. Social Problems, 56(4), to the special issue. Race and Social Problems, 13(1), 1–5. https://
746–764. https://doi.org/10.1525/sp.2009.56.4.746 doi.org/10.1007/s12552-021-09319-2
Hembre, E. (2020). Examining the social safety net response to the Parolin, Z., Curran, M., Matsudaira, J., Waldfogel, J., & Wimer, C.
COVID-19 pandemic. Department of Economics, University of (2020). Monthly poverty rates in the United States during the
Illinois-Chicago. Retrieved December 28, 2022, from https://ssc. COVID-19 pandemic. Poverty and Social Policy Discussion Paper.
wisc.edu/~hembre/wp-content/uploads/2020/09/TANF_COVID. Center on Poverty and Social Policy.
pdf Schanzenbach, D.W. (2020). Not enough to eat: COVID-19 deep-
Karpman, M., & Zuckerman, S. (2021). Average decline in material ens America’s hunger crisis. Food Research and Action Center
hardship during the pandemic conceals unequal circumstances. Report. Retrieved December 28, 2022, from http://frac.org/wp-
Urban Institute. Retrieved from https://www.urban.org/sites/defau content/uploads/Not-Enough-to-Eat_Hunger-and-COVID.pdf
lt/files/publication/104060/average-decline-in-material-hardship- Schanzenbach, D.W., & Pitts, A. (2020). How much has food insecu-
during-thepandemic-conceals-unequal-circumstances_0.pdf rity risen? Evidence from the Census Household Pulse Survey.
Karpman, M., Zuckerman, S., Gonzalez, D., & Kenney, G.M. (2020). Institute for Policy Research Rapid Research Report. Retrieved
The COVID-19 pandemic is straining families’ abilities to afford December 28, 2022, from https://www.ipr.northwester n.edu/
basic needs: Low-income and Hispanic families the hardest hit. documents/repor ts/ipr-rapid-research-repor ts-pulse-hh-data-10-
Urban Institute. Retrieved December 28, 2022, from http://urban. june-2020.pdf
org/research/publication/covid-19-pandemic-straining-families- U.S. Census Bureau. (2020). Social explorer tables: ACS 2020 (5-year
abilities-afford-basic-needs estimates). socialexplorer.com. Retrieved December 28, 2022,
Keith-Jennings, B., Nchako, C., & Llobrera, J. (2021). Number of fami- from https://www.socialexplorer.com/tables/ACS2020_5yr/R1326
lies struggling to afford food rose steeply in pandemic and remains 8806.
high, especially among children and households of color. Center
for Budget and Policy Priorities. Retrieved December 28, 2022, Publisher's Note Springer Nature remains neutral with regard to
from https://cbpp.org/research/food-assistance/number-of-famil jurisdictional claims in published maps and institutional affiliations.
ies-struggling-to-afford-food-rose-steeply-in-pandemic-and
Moffitt, R. A., & Ziliak, J. P. (2020). COVID-19 and the U.S. safety Springer Nature or its licensor (e.g. a society or other partner) holds
net. Fiscal Studies, 41(3), 515–548. https://doi.org/10.1111/1475- exclusive rights to this article under a publishing agreement with the
5890.12243 author(s) or other rightsholder(s); author self-archiving of the accepted
Monte, L.M. (2020). New census Household Pulse Survey shows manuscript version of this article is solely governed by the terms of
households with children lost income, experienced food short- such publishing agreement and applicable law.
ages during pandemic. U.S. Census Bureau. Retrieved December
28, 2022, from http://census.gov/library/stories/2020/05/adults-
in-households-with-children-more-likely-to-report-loss-in-emplo
yment-income-during-covid-19.html
13
File and source
- File
- poverty-reduction-is-not-the-whole-story-the-covid-19-pandemic-response-in-relation-to-material-hardship.pdf
- Size
- 687,246 bytes
- SHA-256
- 31d2f3a5b6354c3dd73945c408034d7cb945bb754cd0904da476284a8722486d
- Original
- doi.org