Effects of the Expanded Child Tax Credit on Employment Outcomes — NBER Working Paper 29823 (March 2022)

Summary

NBER Working Paper No. 29823, dated March 2022, by Elizabeth Ananat, Benjamin Glasner, Christal Hamilton and Zachary Parolin, on the employment effects of the expanded Child Tax Credit (CTC). The paper notes that from July through December 2021 most U.S. families with children received monthly payments of $300 per child under age six and $250 per child between the ages of 6 and 17. Using difference-in-differences analyses of the monthly Current Population Survey and the Census Pulse from April through December 2021, the authors report very small, inconsistently signed and statistically insignificant impacts on employment and labor force participation among adults in households with children. They conclude that the expanded CTC did not have negative short-term employment effects that offset its documented reductions in poverty and hardship.

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

Issuer: Source documents
Document type: Effects Of The Expanded Child Tax Credit On Employment Outcomes Evidence From Real World D
Case: Effects Of The Expanded Child Tax Credit On Employment Outcomes Evidence From Real World Data From April To December 202

About this paper. Across samples and model specifications, the authors find "very small, statistically insignificant, and inconsistently signed impacts" of the expanded Child Tax Credit on employment. The data run from April to December 2021, so the result covers the short run. NBER working papers, this one included, have not been peer-reviewed.

Cite this document

“Effects of the Expanded Child Tax Credit on Employment Outcomes — NBER Working Paper 29823 (March 2022).” Original at www.nber.org. Archived at Pandemic Darlings, https://pandemicdarlings.com/source-documents/effects-of-the-expanded-child-tax-credit-on-employment-outcomes-evidence-from-real-world-data-from-april-to-december-202/, SHA-256 978eed4f674dfae374f7f75de0f0d0db3e6120c0ecc9bda73ea5d7fe6b69cf26.

Full text

NBER WORKING PAPER SERIES

EFFECTS OF THE EXPANDED CHILD TAX CREDIT ON EMPLOYMENT OUTCOMES:
EVIDENCE FROM REAL-WORLD DATA FROM APRIL TO DECEMBER 2021

Elizabeth Ananat
Benjamin Glasner
Christal Hamilton
Zachary Parolin

Working Paper 29823
http://www.nber.org/papers/w29823

NATIONAL BUREAU OF ECONOMIC RESEARCH
1050 Massachusetts Avenue
Cambridge, MA 02138
March 2022

The authors appreciate helpful feedback from Sandra Black, Sophie Collyer, Megan Curran,
Irwin Garfinkel, Dan Hamermesh, Rob Hartley, Jane Waldfogel, and Christopher Wimer. We
acknowledge funding from the Anne E. Casey Foundation and The JPB Foundation. The views
expressed herein are those of the authors and do not necessarily reflect the views of the National
Bureau of Economic Research.

At least one co-author has disclosed additional relationships of potential relevance for this
research. Further information is available online at http://www.nber.org/papers/w29823.ack

NBER working papers are circulated for discussion and comment purposes. They have not been
peer-reviewed or been subject to the review by the NBER Board of Directors that accompanies
official NBER publications.

© 2022 by Elizabeth Ananat, Benjamin Glasner, Christal Hamilton, and Zachary Parolin. All
rights reserved. Short sections of text, not to exceed two paragraphs, may be quoted without
explicit permission provided that full credit, including © notice, is given to the source.
Effects of the Expanded Child Tax Credit on Employment Outcomes: Evidence from Real-World
Data from April to December 2021
Elizabeth Ananat, Benjamin Glasner, Christal Hamilton, and Zachary Parolin
NBER Working Paper No. 29823
March 2022
JEL No. H2,J18,J22

ABSTRACT

Studies have established that the expanded Child Tax Credit (CTC), which provided monthly
cash payments to most U.S. families with children from July to December 2021, substantially
reduced poverty and food hardship. Other studies posit, however, that the CTC payments may
generate negative employment effects that could offset its potential poverty-reduction effects.
Scholars have simulated employment scenarios assuming various labor supply elasticities, but
less work has empirically assessed how the monthly payments affected employment outcomes
using real-world data. To evaluate employment effects, we apply a series of difference-in-
differences analyses using data from the monthly Current Population Survey and the Census
Pulse, both from April through December 2021. Across both samples and several model
specifications, we find very small, inconsistently signed, and statistically insignificant impacts of
the CTC both on employment in the prior week and on active participation in the labor force
among adults living in households with children. Further, labor supply responses to the policy
change do not differ for households for whom the CTC’s expansion eliminated a previous work
incentive. Thus, our analyses of real-world data suggest that the expanded CTC did not have
negative short-term employment effects that offset its documented reductions in poverty and
hardship.

Elizabeth Ananat Christal Hamilton
Barnard College Columbia University
Columbia University 1255 Amsterdam Ave.
3009 Broadway New York, NY 10027
Office 1019 Milstein Building cgh2139@columbia.edu
New York, NY 10027
and NBER Zachary Parolin
eananat@barnard.edu Columbia University
and Bocconi University
Benjamin Glasner zachary.parolin@columbia.edu
Columbia University
1255 Amsterdam Ave.
New York, NY 10027
bng2111@columbia.edu
INTRODUCTION

From July through December 2021, most families with children in the United States (U.S.)

received monthly cash payments of $300 per child under age six and $250 per child between the

ages of 6 and 17. The first monthly payment was distributed to families of 59.3 million children in

July, while subsequent payments reached more than 61 million children (U.S. Department of

Treasury, 2021). Early assessments of the expanded CTC payments suggest they had increased

low-income families’ spending on food items (Perez-Lopez, 2021), had strongly reduced child

poverty rates (Parolin, Collyer, Curran, & Wimer, 2021), and had strongly reduced food

insufficiency (Parolin, Ananat, Collyer, Curran, & Wimer, 2021). Recent empirical research has

had less to say, however, on the expanded CTC’s effects on employment outcomes (one exception

being Roll, Hamilton, & Chun, 2021).

Economic theory predicts competing effects of the expanded CTC on parental labor force

participation and employment. On one hand, the elimination of the phase-in portion of the previous

CTC raises incomes while lowering the return to work, which could reduce parental labor force

participation and employment. As we discuss below, most simulations of the CTC’s effects thus

assume some negative employment effects, though to varying magnitudes (Bastian, 2022; Corinth,

Stadnicki, Meyer, & Wu, 2021; Goldin, Maag, & Michelmore, 2021; National Academies of

Sciences, Engineering, and Medicine, 2019). On the other hand, eliminating the phase-in, and

therefore providing cash to families that are not working, may provide credit-constrained parents

with resources to address barriers to employment, such as by allowing them to secure childcare or

get auto repairs, which could increase labor force participation and employment. Rachidi (2021),

for example, finds that five percent of monthly CTC recipients report that the extra income helped

them work more, compared to an equivalent share (5 percent) who say that the CTC helped them

2
work less. Similarly, data from the Census Household Pulse Survey suggest that more than 10

percent of CTC recipients have used the funds for childcare expenses (Roll, Hamilton, & Chun,

2021). Moreover, increased aggregate demand as a result of higher incomes, particularly among

populations such as low-income families with children that have a high marginal propensity to

consume, may have positive macroeconomic effects that contribute to higher aggregate

employment (Fisher, Johnson, Smeeding, & Thompson, 2020).

In recent simulation work, scholars have modeled the potential employment effects of the

fully-refundable CTC and, in doing so, have reached varying conclusions. The National

Academies of Sciences, Engineering, and Medicine (2019) report on cutting child poverty in half

suggests that a child allowance similar to the reformed CTC might reduce employment by 150,000

workers. Focusing on the parameters of the version of the monthly CTC passed into law, Goldin,

Maag, and Michelmore (2021) likewise estimate small employment responses. Brill et al. (2021)

estimate a decline of 296,000 full-time equivalent jobs. Bastian (2022) estimates that 413,000

adults would stop working. Corinth, Stadnicki, Meyer, and Wu (2021), in contrast, estimate that

the expanded CTC would lead 1.5 million working parents (2.6 percent of all working parents) to

exit the labor force. The differences in the simulations are driven by different assessments

regarding the proper size and the types of employment elasticities applied in the models.

This study diverges from the simulation approach and the related debate over the assumed

size of employment elasticities; we instead build on earlier work by empirically assessing the

employment effects of the expanded CTC using available data pre- and post-expansion.

Specifically, we apply a series of difference-in-differences analyses using data from the monthly

Current Population Survey files from April 2021 through December 2021, and the Census

Household Pulse Survey microdata collected from April 14, 2021 through December 13, 2021. In

3
both datasets, we leverage variation in policy eligibility between households with children and

those without, as well as variation in CTC payment amount received by households based on the

ages of children, the number of children, and pre-reform income levels.

Across both samples and several model specifications, we find very small, statistically

insignificant, and inconsistently signed impacts of the CTC on employment in the prior week and

on being active in the labor force. Findings are similar when evaluating heterogeneous effects

across income levels. Of note, effects for families who earned less than $25,000 prior to the reform,

and who therefore face the largest decline in the return to work due to the elimination of the CTC

phase-in, are indistinguishable from the effects for higher-income families, in contrast to the

predictions of simulation work. The short-term consequences of the CTC for employment,

estimated using real-world outcomes, are not consistent with simulations that suggest that the

CTC’s long-run employment effects will meaningfully offset its observed reductions in poverty

and hardship (Corinth et al., 2021). They are, instead, consistent with other real-world results

showing declines in poverty and food insufficiency in families with children as a result of the CTC

expansion.

We acknowledge that long-run employment effects, should the program be made

permanent, could be different from those observed in 2021. However, theory suggests that the

temporary nature of the 2021 program, along with other exceptional features of the 2021 labor

landscape, arguably make the effects of the 2021 expansion a lower bound for any negative labor

supply effects of a permanent expansion.

4
POLICY CONTEXT

Prior to the expanded CTC, tax filers could receive a maximum CTC of $2,000 per child

per year, but many children did not receive the full refundable amount. 1 The CTC credit phased in

at a rate of 15 cents per dollar of earnings, providing an effective wage subsidy but denying benefits

to those with low or no earnings. As a result, one in three children did not receive the full benefit

value because their families did not earn enough income to qualify. Children with single parents,

those in rural areas, Black and Latino children, and those in larger families were disproportionately

ineligible for the full credit (Collyer, Harris, & Wimer, 2019; Curran & Collyer, 2020).

Following parameters similar to the American Family Act (a bill first introduced in both

the Senate and House of Representatives in 2017 and reintroduced in 2019), the American Rescue

Plan (ARP) temporarily transformed the CTC into a nearly-universal child allowance for 2021. 2

Specifically, the ARP included three fundamental changes to the CTC. First, it made the CTC

available to almost all children—including those previously excluded due to their families’ low

income—by removing the earnings requirement and making the credit fully refundable. Second,

it raised the maximum annual credit amount to $3,000 for children ages 6–17 and $3,600 for

children under age six. Third, beginning mid-July 2021, the ARP delivered the credit in monthly

installments of up to $250 per older child or up to $300 per child under six years of age, for a

period of six months.3

1
For additional information on the history of the Child Tax Credit, see Crandall-Hollick (2021), Crandall-Hollick
(2018), and Garfinkel et al. (2016).
2
The expansion to the CTC in the ARP mirrors the proposed reforms in the American Family Act (AFA) with one
exception: in the AFA, the credit would begin to phase out for heads of household with earnings above $120,000 or
and joint filers with Adjusted Gross Incomes (AGI) over $180,000. In the ARP, the credit began to phase out for
families with AGIs above $112,500 or $150,000 per year, depending on filing status, but it only phased out until
matching the credit values that a family would receive under prior law. This alteration was made because the Biden
administration committed to not raising taxes for those with incomes below $400,000 per year.
3
Because the payments began halfway through the year, families received half of the full amount of their credit in
2021 and the remainder after filing their 2021 taxes in 2022.

5
Early research suggests that the expanded CTC generated large reductions in child poverty

(Parolin, Collyer, et al., 2021), reduced food insufficiency among households with children

(Parolin, Ananat, et al., 2021), increased spending on childcare (Roll, Hamilton, & Chun, 2021),

and helped as many working parents “work more” as to “work less” (Rachidi, 2021). Though

economic theory largely predicts, and has often demonstrated, that an unconditional cash transfer

can lead to employment declines, recent studies evaluating similar child benefit programs in

Canada—the Canadian Universal Child Care Benefit (UCCB) and the Canada Child Benefit

(CCB)—found no evidence of a labor supply response (Baker, Messacar, & Stabile, 2021).

For that reason, recent research simulating the effects both of a permanent child allowance

(National Academies of Sciences, Engineering, and Medicine, 2019) and of a permanent version

of the 2021 CTC expansion (Corinth et al., 2021) has modeled the income elasticity of parent

earnings with respect to the transfer as zero. However, differences in assumptions about the price

elasticity of parent earnings—relevant because of the expansion’s erasure of the phase-in range of

the credit, which eliminated a wage subsidy for the lowest-earning parents—led to different

predictions about the effects of this expansion on parent employment, with Corinth et al. (2021)

suggesting employment declines that range from 150,000 adults to 1.5 million adults.

Rather than projecting future employment losses, this study investigates real-world data to

assess whether the temporary CTC expansion had a plausibly causal effect on parents’ labor

market engagement during its installments of six monthly payments in 2021. While these estimates

reflect the effects of a temporary, rather than permanent, policy change, the effects of a temporary

cut in the return to work on labor supply are likely to provide a lower bound for any negative

effects of a permanent change. Holding the income elasticity of parent labor supply at zero, as is

now conventional in the literature, means that only the compensated price elasticity is at play.

6
Compensated intertemporal labor supply elasticities will be larger in response to temporary wage

changes than in response to permanent wage changes due to incentives for re-timing (see, e.g.,

Biddle, 1988). That is, the policy provides a time-limited opportunity for a non-work spell at a

lower price than usual, which will cause a much higher number of parents to re-time a desired non-

work spell toward the low-price period than would take time out of the labor force in any given

period if the price were permanently lowered. In addition, there are several reasons parents would

have been more likely to substitute away from employment during 2021 than they might be in

more typical times. First, the threat of COVID temporarily increased the cost to many Americans

of working, particularly in the public-facing roles most often available to the low-wage workers

most affected by the elimination of the phase-in. Second, child care and school continued to be

unpredictable in the second half of 2021, again making work temporarily less feasible and

attractive for parents. Third, continuing worker shortages over the period implied parents would

be able to find jobs in 2022 if they took time off in 2021, temporarily lowering the opportunity

cost of delayed job matching. All of these factors increased parents’ incentives to re-time non-

work spells toward late 2021 in response to the temporarily lower cost of such spells. Even if only

some parents perceived the change in CTC as temporary, while others perceived it as permanent,

the aggregate decline in employment should still provide a lower bound for the aggregate decline

under an announced permanent CTC change. Thus, empirical estimates of the effects of the

temporarily expanded 2021 CTC on parental employment are likely to provide a lower bound for

any parental employment effects of a permanently expanded CTC.

DATA AND METHODS

Data

7
This study uses data from the Current Population Survey (CPS) and the Census Household

Pulse Survey (Pulse). The CPS is a nationally representative monthly household survey conducted

jointly by the U.S. Census Bureau and the Bureau of Labor Statistics. As the primary source of

employment statistics for the US population, the CPS gathers data on labor force participation and

demographics in its basic monthly survey, as well as data on other dimensions of social and

economic well-being through various topic-oriented supplements. We use data from April 2021

through December 2021 basic monthly surveys. We exclude the month of July, given that we

cannot distinguish pre-treatment versus post-treatment responses within July.

We supplement the CPS data with analyses from the Census Household Pulse survey. The

Pulse was introduced in April 2020 by the U.S. Census Bureau to obtain up-to-date and nationally

representative information on the social and economic wellbeing of households across the U.S.

Participants are randomly selected to participate in the survey, which has been used to track trends

in material hardship, subjective well-being, and other social and economic indicators throughout

the COVID-19 pandemic (Bauer et al., 2018; Bitler, Hoynes, & Schanzenbach, 2020; Cai,

Woolhandler, Himmelstein, & Gaffney, 2021; Morales, Morales, & Beltran, 2020; Schanzenbach

& Pitts, 2020; Twenge & Joiner, 2020; Ziliak, 2021). For this study, we use data collected from

April 14, 2021 through December 13, 2021 (Waves 28–40). The first payment of the expanded

CTC was delivered to recipients on July 15, 2021, which falls prior to the beginning of Wave 34

of the Pulse (Wave 34 spans July 21 to August 2, 2021).

In Wave 34 of the Pulse, the U.S. Census Bureau altered the position of its employment

question within the survey, leading to different levels of item non-response for employment status

relative to prior waves. Given that the change in item non-response affects both our treatment and

control groups, it should not bias our findings. In particular, even if the change were to affect

8
families with children differently from childless families, one of our analytical strategies,

discussed below, identifies differences in the magnitude of the change in CTC benefit due to the

expansion among families with children. These differences are idiosyncratic functions of the age

and number of children and family income, and do not correspond neatly to otherwise salient

differences between families that might drive differences in effects of the survey change. Thus,

our robust approach of using multiple difference-in-differences specifications allows us to produce

credible estimates of the CTC’s effects on employment within the Pulse, along with parallel

estimates using the CPS, which was unaffected by any such change.

Sample

We limit our primary samples to adults between the ages of 18–65 years old. Our pooled

CPS monthly sample includes 504,364 adults. For our pooled Pulse data, we exclude adults

residing in households that had imputed values for the number of children in the household, as

errors in the imputed values could bias our estimates. Our final Pulse primary analytical sample

includes 478,026 adults. Descriptive statistics for both samples are shown in Table 1.

Measures

Outcomes

In the CPS, we investigate the impact of the CTC expansion on the extensive margin of the

supply of labor using both employment status, as measured by an indicator variable for

employment at the time of the interview, and whether adults were members of the labor force in

the week prior to their CPS interview. The Pulse does not collect data on labor force participation,

limiting our analyses of the Pulse sample to only effects on employment status.

9
Receipt of the CTC

The Pulse includes its own question about whether anyone in the household received a

CTC payment under the American Rescue Plan (U.S. Department of Treasury, 2021), which we

use to estimate treatment-on-the-treated (TOT) effects in addition to intent-to-treat (ITT) effects

based on the presence of children. By contrast, the monthly CPS does not gather information on

CTC receipt. We therefore code a household as eligible for receiving CTC payments in the CPS if

they had at least one child present in the household, and we estimate only ITT effects in the CPS.

Income Subgroups

We also conduct subgroup analyses by pre-expansion household income. In the CPS, we

define subgroups using a categorical income variable that includes “money from jobs, net income

from business, farm or rent, pensions, dividends, interest, social security payments, and any other

money income received” among family members in the prior year. In the Pulse, we use a similar

categorical income variable that includes “pre-tax income” from 2019 (waves 28 to 33) or 2020

(waves 34 onward).

Analytical Strategy

Our analytical strategy follows that of Parolin, Ananat, Collyer, Curran, and Wimer (2021).

To assess the effect of the CTC on the labor market outcomes we employ difference-in-differences

models using the following equation:

𝑦𝑦𝑖𝑖𝑖𝑖 = 𝛽𝛽1 𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑡𝑡 + 𝛽𝛽2 𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑖𝑖 + 𝛽𝛽3 (𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃 ∗ 𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇)𝑖𝑖𝑖𝑖 + 𝛽𝛽4 𝑋𝑋 + +𝜀𝜀𝑖𝑖𝑖𝑖 (1)
𝑖𝑖

where yit is one of our two outcomes of interest (employment status; labor force participation).

PostCTC is a binary indicator of whether the time of the survey was after July 15th, 2021, the date

when the expanded CTC was first administered. X is a vector of household head characteristics—

age, sex, and education status—and both state and month fixed effects.

10
We specify our treatment variable, Treatment, in three separate ways. In our first

specification, we operationalize a binary treatment indicator that is equal to one if a household had

children and set to zero if no child was present in the household. In our second specification, we

estimate models using a continuous measure of treatment intensity to capture the fact that the

effects of the CTC are likely to vary by the age of children (as families with children under age six

receive larger monthly benefit values), the number of children in the home, and the relative value

of the new CTC benefits compared to what the household likely received from the CTC prior to

the reform. Because we do not have information on pre-reform CTC receipt, we use data from the

2019 U.S. Current Population Survey Annual Social and Economic Supplement (CPS ASEC) to

estimate the mean pre- and post-reform benefit values for bins defined by the number of adults in

the household (ranging from 1 to 10), the number of children in the household (ranging from 0 to

10), and eight categorical pre-tax income bins (from under $25,000 annually scaling up to more

than $200,000 per year). We compute the mean pre-reform refundable CTC benefits as observed

for each family unit in the CPS ASEC. We then simulate the additional post-reform benefits that

each family is eligible for using detailed policy rules from the CTC reform as specified in the 2021

American Rescue Plan. We subtract the pre-reform benefit value from the post-reform benefit

value to create a “net benefit” indicator for each family unit. 4 Finally, we calculate the weighted

mean of the net benefit value for each of the bins defined above. We then import this value into

our 2021 Pulse and monthly CPS data, matching on the number of adults, number of children, and

total family income of respondents. Figure 1 displays the mean net CTC benefits by income bin

and family size. In our third specification, we calculate the change in income from work plus

4
In a sensitivity test, we also evaluate the results when adjusting the benefits for family size using the modified OECD
equivalent scale. The OECD scale begins with a value of 1 for a single adult, then adds 0.5 for each child in the home
and 0.3 for each additional adult in the home. The results are not meaningfully different from our primary findings.

11
expanded CTC, relative to 2019 income from work plus prior CTC benefit, in order to reflect the

change in the return to work faced by a household.

In each model, β3 is our primary coefficient of interest, as it identifies, when using the

binary treatment indicator, whether adults in households with children were more or less likely to

be engaged in the labor market relative to adults in childless households after the introduction of

the CTC changes. When using the continuous dollar treatment, it represents the effect of eligibility

for an additional $100 per month in increased net benefits on labor market engagement. When

using the third specification, it represents the effect of a given percent change in the return to work

on labor market engagement.

Because we are unable to identify whether households actually received CTC payments in

the CPS, β3 in Equation (1) provides the intent-to-treat effect—the effect of the treatment on the

full treatment group, regardless of actual CTC receipt—of the CTC expansion for this sample. In

addition to estimating Equation (1) for the Pulse sample, we also utilize the information on CTC

receipt in the survey to provide estimates of the treatment effect on the treated, or the local average

treatment effect. To do so, we estimate two-stage least squares models (2SLS) using the treatment

group identifier as an instrumental variable and observed receipt of the treatment as the

endogenous variable. When applying our binary treatment measure, observed receipt of the

treatment reflects whether a family in the Pulse reports having received the monthly CTC

payment(s). When applying our continuous treatment indicator, the observed treatment in the

2SLS model is a family’s projected net benefit increase from the CTC interacted with their report

of receipt. In other words, because previous and current size of the CTC benefit received are not

directly measured in the Pulse, we apply our projected value of the net CTC benefit increase based

on the family’s income, number of children, and number of adults (as defined above) as the

12
observed treatment; however, we convert the projected benefit value to zero for families reporting

that they did not receive the CTC payment.

In addition to our main analytical specification, we also estimate effects separately for

families in different pre-period annual income bins, from less than $25,000 to over $200,000 in

total family income. These subgroup estimates allow us to test the fundamental prediction of prior

simulation work, namely that parental labor force participation and employment will decline

(Corinth et al., 2021). That work posits that the decline in the return to employment due to the

elimination of the phase-in will cause some parents to drop out of the labor force. Because the

percentage drop in the return to work is larger the lower the earnings of the parent, their simulations

predict that the lowest-income families will exhibit the largest reduction in employment after

expansion of the CTC. Evidence inconsistent with the lowest-income families responding

differently from higher-income families thus represents prima facie evidence against the validity

of the assumptions driving recent simulations’ predictions that labor force decline will offset anti-

poverty and anti-material hardship effects of the CTC expansion.

FINDINGS

Descriptive Findings

Figure 2 shows the trend in employment and labor force participation from April 2021 to

December 2021 using the monthly CPS. This window is in the middle of the COVID-19 pandemic

and begins after the initial macro-level drop in employment and participation. Each point

represents the weighted average employment rate or the labor force participation rate among

respondents between the age of 18 and 65. These results are split by whether the respondent

reported at least one child in the household.

13
Figure 2 highlights the baseline difference in labor market engagement between adults in

households with and without children, wherein adults in childless households tended to have lower

levels of employment and labor force participation throughout this sample window. Employment

and labor force participation increased among adults in both household types from April through

June 2021. For adults in families with children, employment continued to increase even after the

first CTC payment in July 2021. Labor force participation among adults in families with children

in December 2021 was higher than its pre-CTC value, in contrast to the labor force participation

rate of adults in childless families.

Patterns are similar when evaluating differences in labor market engagement by household

income (Figure 3). Across three income bins from $0 to $34,999, $35,000 to $74,999, and $75,000

or more, adults in families with children were consistently more likely to be employed and

participate in the labor market compared to adults in childless households. Nevertheless, in all

three income bins, employment and labor force participation rates among adults in households

with children trended similarly to rates among adults in childless households from April to July.

Estimation Results

Table 2 presents the results from difference-in-differences estimates on employment status

and labor force participation using data from the CPS sample. Columns 1 and 2 provide estimates

from models using our binary treatment measure, while Columns 3 and 4 provide estimates from

our continuous treatment measure as discussed above. Our results suggest, in short, that the CTC

expansion had no significant impact on employment outcomes.

Specifically, estimated effects using the binary model of treatment show that relative to

adults in childless households, adults in households with children had a statistically insignificant

0.2 percentage point drop in employment (s.e.=0.005) and 0.1 percentage point increase in labor

14
force participation (s.e.=0.005) after CTC expansion. Using the continuous treatment measure—

capturing variation based on pre-tax income and household size—we find that an additional $100

in monthly CTC led to a statistically insignificant 0.1 percentage point increase in employment

(s.e.=0.1 percentage points) and a statistically insignificant 0.2 percentage point increase in labor

force participation (s.e.=0.1 percentage point).

Table 3 displays the estimated effect of the CTC on being employed using data from the

Pulse. Again, we present estimated effects using our binary and continuous treatment measures.

Due to data limitations in the Pulse (which does not ask about labor force participation), Table 2’s

estimates are restricted to employment effects. Our analyses presented in Columns 1 and 3 estimate

the effect of the policy intending to treat (ITT) all households with children, regardless of whether

they report receiving the benefit. Columns 2 and 4 present the 2SLS estimates of the treatment

effect on the treated (TOT), using the intent-to-treat as an instrument for the actual treatment.

Similar to analyses using the CPS data, we find that CTC expansion had no statistically significant

impacts on being employed: the estimated increase in employment ranges from 0.1 percentage

points to 0.5 percentage points, with standard errors ranging from 0.2 to 0.9 percentage points.

Figures 4 and 5 visualize the effect of the CTC on employment outcomes across the income

distribution using the CPS and Pulse data respectively. Each point in Figures 3 and 4 represents

the coefficient from the interaction terms for our binary treatment (black circle) and continuous

treatment (gray triangle) when subsetting the sample to only include households with total incomes

under $25,000, then between $25,000 to $35,000, $35,000 to $50,000, $50,000 to $100,000, and

$100,000 to $200,000.

Figure 4 and Figure 5 highlight the insignificant and inconsistently signed relationship

between the CTC expansion and employment across the income distribution, using both data sets

15
and both the binary and continuous treatment measures. In the CPS, where we can also observe

labor force participation, we also find insignificant and inconsistently signed relationships between

the CTC expansion and engagement in the labor force at nearly every point in the income

distribution. The one exception is the binary treatment for adults with children in the $50,000 to

$100,000 bin (-0.1 percentage point decline), but our continuous treatment estimate for this income

group is not statistically significant.

Notably, we find no significant differences between the effects of expansion on the lowest

income groups and higher income groups, nor do we find any evidence of a gradient in effects by

income—both of which would be predicted by the elasticity assumptions that drive Corinth et al.’s

(2021) predictions of labor force decline. In fact, in some cases, estimates for the lowest income

groups are more positive than those for other groups.

In Table 4, we estimate models using the change in the return to work rather than the change

in income. Due to the fully refundable status of the CTC expansion, the relative return to work is

reduced in comparison to the CTC payments from the Tax Cuts and Jobs Act. Table 5 defines

treatment as the percent change in the net effective wage (difference between new earnings plus

work-dependent transfers and old earnings plus work-dependent transfers divided by old income

plus work-dependent transfers) under the expanded CTC. Again, the results are statistically

insignificant and inconsistently signed.

Robustness Checks

In Table 5, we present alternative models using March 2021, the month when the policy

was introduced (but before CTC payments were distributed), as our treatment timing. Similar to

our actual estimates, we find small and insignificant effects that suggest that parents did not

respond to the announcement of the policy by reducing their labor supply. In Table 6, we replicate

16
our primary findings on 2020 CPS data to assess the possibility of seasonality affecting our

findings. The results suggest that this is not the case.

Callaway and Sant’Anna Group-Time Treatment

Recent work has cast doubt on both the validity and robustness of the two-way fixed effect

estimator when more than two treatment groups and periods are included in analytical models,

treatment timing varies, and treatment effect is heterogeneous (Borusyak, Jaravel & Spiess, 2021;

Callaway & Sant’Anna 2021; Goodman-Bacon, 2021; Sun & Abraham, 2021). While the use of

the CTC expansion as a treatment does not fall within these critiques, the two-way fixed effects

estimator still reports a weighted average treatment on the treated estimate across treatment groups

because treatment size varies across households. Fortunately, however, treatment size varies on

observable characteristics—the number and age of children in the household.

A further complication in our analytic design is the continuous nature of CTC payments.

The traditional difference-in-differences framework operates on a binary treatment, identifying the

treatment and control groups and clean comparisons. When using a continuous treatment, the

comparison groups become more difficult to identify (Callaway, Goodman-Bacon, & Sant’Anna,

2021). For continuous treatments, both the level and slopes of the dose-response relationship are

required to gain a complete understanding of the treatment effects of a group-varying continuous

treatment.

We address these issues by leveraging the Callaway and Sant’Anna (2021) estimator at

both the aggregate and group treatment levels. We first treat the introduction of the CTC as a

simultaneous binary treatment among all eligible households. We use the binary treatment coding

to create an event study design, presented in Figure 6, exploring how the CTC expansion’s effect

17
changed in the months following its adoption. This also allows us to estimate an aggregate effect

across all eligible individuals. The binary treatment coding is expanded in three ways. First, we

create separate household groups by household composition (i.e., the number of children and their

ages). Second, we create $25 bins of additional monthly CTC payments and present those

treatment bins which are represented by at least 1% of the sample. Third, we create 1% bins of the

change in the relative wage ranging from no change in the control group to a reduction of 9% or

more. Repeating the analysis at the treatment-group level in Tables 7, 8, and 9 allows us to estimate

a different ATT(d/x) for CTC dosage d in treatment group x. This estimation does not allow us to

make the case that each ATT(d/x) can be linked as a single dosage response curve, but rather that

each is a different point on each group x’s dosage response curve.

Results from our analysis using the Callaway and Sant’Anna (2021) estimator reveal no

evidence of an effect of the CTC expansion on employment or labor force participation. These

results are in line with results from our main analytical specification. When exploring the CTC

expansion by groups, we find no evidence of significant deviation from the aggregate result. The

CTC expansion did not appear to affect parents’ employment or labor force participation regardless

of household composition.

DISCUSSION AND CONCLUSION

The temporary expansion of the Child Tax Credit in July 2021 into a more generous and

inclusive monthly payment represented a large deviation from the direction of the U.S. welfare

state throughout the past three decades. Recent studies have shown that the CTC expansion had

reduced poverty rates (Parolin, Collyer, et al., 2021) and food insufficiency (Parolin, Ananat, et

al., 2021) among households with children. Some researchers, however, suggest that the expansion

18
may generate negative employment effects. While recent simulation research has reached varying

conclusions—with some scholars finding negligible negative effects and others significant

reductions in labor force participation—little research to date has empirically evaluated the impact

of the CTC expansion on employment outcomes using real-world data.

This study uses two nationally-representative datasets covering the six months of payments

(April to December 2021) to assess the effect of the expanded CTC on employment outcomes

among adults with children. To identify policy effects, we implemented a series of difference-in-

differences models leveraging the variation in policy eligibility and the structure of CTC payment

amounts. In addition to our main analytical specification, we also estimated effects separately for

families in different pre-period annual income bins, from less than $25,000 to $100,000-$200,000

in total pre-period family income.

Our analyses reveal that the CTC expansion had no meaningful impact on employment and

labor force participation after six months of benefit distribution. Among adults in the CPS sample,

we find statistically insignificant differences in employment (with inconsistent signs) and

statistically insignificant increases in labor force participation. Using the Pulse data, in which we

estimate treatment effects on the treated, we find a statistically insignificant increase in

employment among adults in eligible households in the six months following the CTC expansion.

A variety of robustness checks also consistently show insignificant effects on employment and

labor force participation.

Results from analyses exploring heterogeneity across income groups also reveal that the

CTC expansion had no significant effect on employment and labor force participation for the

lowest-income groups, which benefited most from the CTC expansion. Of note, we find no

significant differences in policy effects between the lowest income groups and high-income

19
groups, or any evidence of a gradient in effects by income. Evidence inconsistent with the lowest-

income families responding differently from higher-income families represents prima facie

evidence against the validity of the assumptions driving recent simulations’ prediction that labor

force decline will offset anti-poverty and anti-material hardship effects of the CTC expansion

(Corinth et al., 2021). Moreover, evidence from a temporary change in the cost of non-

employment, given the scholarly consensus that the income effect of the CTC on parent

employment is zero, should provide a lower bound for the effect of a permanent expansion of the

CTC on employment. Evidence that even the temporary wage cut in the 2021 CTC expansion did

not reduce parent employment thus strongly suggests that a permanent expansion would not

meaningfully reduce parent employment.

We conclude that real-world data on employment during the CTC expansion do not support

claims that the elimination of the phase-in portion of the CTC discouraged work among parents in

any meaningful way, much less that such effects are large enough to offset decreases in poverty

and material hardship driven by the expansion’s increased generosity and inclusivity. Our results

are, instead, consistent with other real-world analyses of the expansion’s effects, which show

observed declines in child poverty (Parolin, Collyer, et al., 2021) and strongly reduced food

insufficiency (Parolin, Ananat, et al., 2021). Our results suggest that the mechanism through which

these reductions occurred is that the increase in income that was both intended and accomplished

through the expansion, combined with a lack of unintended effects on parental work, led to

improved well-being for families with children.

20
References

Baker, M., Messacar, D., & Stabile, M. (2021). The effects of child tax benefits on poverty and
labor supply: Evidence from the Canada Child Benefit and Universal Child Care Benefit
(No. w28556). National Bureau of Economic Research.
Bastian, J. (2022). Investigating the effects of the 2021 Child Tax Credit expansion on poverty
and maternal employment. Retrieved from
https://drive.google.com/file/d/1H5iNZZO_YFRIDz-3Tip4C-BpnD85bUjH/view
Bauer, L., Schanzenbach, D. W., & Shambaugh, J. (2018). Work requirements and safety net
programs. The Hamilton Project.
Biddle, J.E. (1988). Intertemporal substitution and hours restrictions, The Review of Economics
and Statistics, MIT Press 70(2), 347-351.
Bitler, M., Hoynes, H. W., & Schanzenbach, D. W. (2020). The social safety net in the wake of
COVID-19 (No. w27796). National Bureau of Economic Research.
Borusyak, K., Jaravel, X., & Spiess, J. (2021). Revisiting event study designs: Robust and efficient
estimation. arXiv preprint arXiv:2108.12419.
Brill, A., Pomerlau, K., & Seiter, G. (2021). Model introduction: Estimating the labor supply
response to a permanent Child Tax Credit expansion. Retrieved from
https://grantseiter.com/CTC-Labor-Response/intro.html
Cai, C., Woolhandler, S., Himmelstein, D. U., & Gaffney, A. (2021). Trends in anxiety and
depression symptoms during the COVID-19 pandemic: Results from the US Census
Bureau’s Household Pulse Survey. Journal of General Internal Medicine, 36(6), 1841-
1843.
Callaway, B., Goodman-Bacon, A., & Sant'Anna, P. H. (2021). Difference-in-differences with a
continuous treatment. arXiv preprint arXiv:2107.02637.
Callaway, B., & Sant’Anna, P. H. (2021). Difference-in-differences with multiple time
periods. Journal of Econometrics, 225(2), 200-230.
Collyer, S., Harris, D., & Wimer, C. (2019). Left behind: The one-third of children in families who
earn too little to get the full Child Tax Credit. Poverty & Social Policy Brief, 3(6).
Corinth, K., Stadnicki, M., Meyer, B. D., & Wu, D. (2021). The anti-poverty, targeting, and labor
supply effects of the proposed Child Tax Credit expansion. University of Chicago, Becker
Friedman Institute for Economics Working Paper, No. 2021-115.
Crandall-Hollick, M. L. (2018). The Child Tax Credit: Current Law. CRS Report R41873.
Washington, DC: Congressional Research Service
Crandall-Hollick, M. L. (2021). The Child Tax Credit: Proposed expansion in the American
Rescue Plan Act of 2021 (ARPA; H.R. 1319). CRS Report IN11613. Washington, DC:
Congressional Research Service
Curran, M. (2015). Catching up on the cost of raising children: Creating an American child
allowance. In Big Ideas: Pioneering Change, Innovative Ideas for Children and Families.
Washington DC: First Focus.
Curran, M., & Collyer, S. (2020). Children left behind in larger families: The uneven receipt of
the federal Child Tax Credit by children’s family size. Poverty & Social Policy Brief, 4(4).
Retrieved from
https://static1.squarespace.com/static/5743308460b5e922a25a6dc7/t/5e605fe34cbc0979d
6dd5ac8/1583374310078/Child-Tax-Credit-Larger-Families-CPSP-2020.pdf

21
Fisher, J. D., Johnson, D. S., Smeeding, T. M., & Thompson, J. P. (2020). Estimating the marginal
propensity to consume using the distributions of income, consumption, and wealth. Journal
of Macroeconomics, 65, 103218.
Garfinkel, I., Harris, D., Waldfogel, J., & Wimer, C. (2016). Doing more for our children:
Modeling a universal child allowance or more generous child tax credit. New York: The
Century Foundation.
Goldin, J., Maag, E., & Michelmore, K. (2021). Estimating the net fiscal cost of a Child Tax Credit
expansion. NBER Chapters.
Goodman-Bacon, A. (2021). Difference-in-differences with variation in treatment timing. Journal
of Econometrics, 225(2), 254-277.
Hoynes, H. (2019). The earned income tax credit. The Annals of the American Academy of Political
and Social Science, 686(1), 180-203.
Morales, D. X., Morales, S. A., & Beltran, T. F. (2020). Racial/ethnic disparities in household food
insecurity during the COVID-19 pandemic: a nationally representative study. Journal of
Racial and Ethnic Health Disparities, 1-15.
National Academies of Sciences Engineering and Medicine. (2019). A roadmap to reducing child
poverty. Washington, DC: National Academies Press
Pac, J., Nam, J., Waldfogel, J., & Wimer, C. (2017). Young child poverty in the United States:
Analyzing trends in poverty and the role of anti-poverty programs using the Supplemental
Poverty Measure. Children and Youth Services Review, 74, 35-49.
Parolin, Z., Ananat, E., Collyer, S. M., Curran, M., & Wimer, C. (2021). The initial effects of the
expanded Child Tax Credit on material hardship (No. w29285). National Bureau of
Economic Research.
Parolin, Z., Collyer, S., Curran, M., & Wimer, C. (2021). Monthly poverty rates among children
after the expansion of the Child Tax Credit. Poverty & Social Policy Brief 5(4). Retrieved
from https://www.povertycenter.columbia.edu/s/Monthly-Poverty-with-CTC-July-CPSP-
2021.pdf
Perez-Lopez, D. (2021). Economic hardship declined in households with children as Child Tax
Credit payments arrived. U.S. Census Bureau. Retrieved from
https://www.census.gov/library/stories/2021/08/economic-hardship-declined-in-
households-with-children-as-child-tax-credit-payments-arrived.html
Rachidi, A. (2021). New survey data raises questions about the expanded Child Tax Credit.
American Enterprise Institute. Retrieved from https://www.aei.org/poverty-studies/new-
survey-data-raises-questions-about-the-expanded-child-tax-credit/
Roll, S., Hamilton, L., & Chun, Y. (2022). Expanded Child Tax Credit payments have not reduced
employment. Evidence from Census data. Retrieved from https://cpb-us-
w2.wpmucdn.com/sites.wustl.edu/dist/a/2003/files/2021/10/CTC-and-Employment-
10212021.pdf.
Schanzenbach, D., & Pitts, A. (2020). Food insecurity in the census household pulse survey data
tables. Institute for Policy Research, 1-15.
Shaefer, H. L., Collyer, S., Duncan, G., Edin, K., Garfinkel, I., Harris, D., . . . Yoshikawa, H.
(2018). A universal child allowance: A plan to reduce poverty and income instability
among children in the United States. RSF: The Russell Sage Foundation Journal of the
Social Sciences, 4(2), 22-42.
Shaefer, H. L., & Edin, K. (2013). Rising extreme poverty in the United States and the response
of federal means-tested transfer programs. Social Service Review, 87(2), 250-268.

22
Sun, L., & Abraham, S. (2021). Estimating dynamic treatment effects in event studies with
heterogeneous treatment effects. Journal of Econometrics, 225(2), 175-199.
Twenge, J. M., & Joiner, T. E. (2020). US Census Bureau‐assessed prevalence of anxiety and
depressive symptoms in 2019 and during the 2020 COVID‐19 pandemic. Depression and
Anxiety, 37(10), 954-956.
U.S. Department of Treasury. (2021). Treasury and IRS announce families of nearly 60 million
children receive $15 billion in first payments of expanded and newly advanceable Child
Tax Credit [Press release]. Retrieved from https://home.treasury.gov/news/press-
releases/Treasury-and-IRS-Announce-Families-of-Nearly-60-Million-Children-
Receive-%2415-Billion-Dollars-in-First-Payments-of-Expanded-and-Newly-
Advanceable-Child-Tax-Credit
Ziliak, J. P. (2021). Food hardship during the COVID‐19 pandemic and Great Recession. Applied
Economic Perspectives and Policy, 43(1), 132-152.

23
TABLES AND FIGURES

Table 1: Descriptive Statistics for Analytical Samples
CPS Pulse
Mean St. Dev. Mean St. Dev.
Employment rate 0.746 0.435 0.795 0.404
Labor force participation
0.781 0.413 - -
rate
Age 36.28 10.47 35.98 6.42
Percent female 0.508 0.5 0.628 0.483
Family size 3.17 1.63 3.21 1.56
Number of children 0.918 1.22 1.10 1.23
N 504,364 478,026
Note: Sample: adults between ages 18–65.

24
Table 2: Difference-in-Differences Estimates of the Effect of Expanded CTC on Employment
Outcomes (Current Population Survey, April 2021 through December 2021)
Binary Treatment Continuous Treatment
($100s of Net Monthly Benefit Value)
1: Employed 2: Active in 1: Employed 2: Active in
Labor Force Labor Force
Household with Children 0.042*** 0.035*** 0.000 -0.000
(0.004) (0.004) (0.001) (0.001)

Household with Children -0.002 0.001 0.001 0.002
X Post-July 15 (0.005) (0.005) (0.001) (0.001)

Pre-Treatment Mean 0.727 0.768 0.727 0.768
among Households with
Children

Observations 504,364 504,364 502,112 502,112
Note: Sample: adults between ages 18–65. All models include state and month fixed effects and controls for age,
education, and sex of individual. Robust standard errors in parentheses. * p < 0.05, ** p < 0.01, *** p < 0.001.

25
Table 3: Difference-in-Differences Estimates of the Effect of Expanded CTC on Employment
Outcomes (U.S. Census Household Pulse Survey, April 14 through December 13, 2021)
Binary Treatment Continuous Treatment
($100s of Net Monthly Benefit Value)
1: Employed 2: Employed 3: Employed 4: Employed
(Intent to Treat) (Treatment on (Intent to Treat) (Treatment on
Treated) Treated)
Household with Children 0.005 0.005 -0.008*** -0.008***
(0.004) (0.004) (0.001) (0.001)

Household with Children 0.003 0.005 0.001 0.001
X (0.006) (0.009) (0.002) (0.002)
Post-July 15

Pre-Treatment Mean 0.701 0.701 0.701 0.701
among Households with
Children

Observations 478,026 478,026 416,071 416,071
Note: Sample: adults between ages 18–65. All models include state and week fixed week effects and control for age,
education, and sex of respondent. Robust standard errors in parentheses. * p < 0.05, ** p < 0.01, *** p < 0.001.

26
Table 4: Difference-in-Differences Estimates of the Effect of the Change in Relative Wage on
Employment Outcomes (Current Population Survey and U.S. Census Household Pulse Survey)
CPS Pulse
1: Employed 2: Active in 3: Employed 4: Employed
Labor Force (Treatment on
Treated)
% Change in Relative -1.037*** -0.845*** -0.979*** -1.142***
Wage (0.086) (0.089) (0.132) (0.144)

% Change in Relative 0.025 -0.045 -0.036 -.241
Wage X Post-July 15 (0.087) (0.083) (0.131) (.227)

Pre-Treatment Mean 0.752 0.785 0.761 0.761
among Households with
Children

Observations 504,364 504,364 478,026 478,026
Note: Sample: adults between ages 18–65. All models include state and month fixed effects and controls for age,
education, and sex of individual. Robust standard errors in parentheses clustered at the state level. * p < 0.05, ** p <
0.01, *** p < 0.001.

27
Table 5: Difference-in-Differences Estimates of the Effect of Expanded CTC on Employment
Outcomes with March 2021 as Treatment Start Month (CPS, January 2021 through December
2021)
Binary Treatment Continuous Treatment
($100s of Net Monthly Benefit Value)
1: Employed 2: Active in 1: Employed 2: Active in
Labor Force Labor Force
Household with 0.038*** 0.032*** -0.000 -0.001
Children (0.003) (0.004) (0.001) (0.001)

Household with 0.003 0.004 0.001 0.002
Children X (0.003) (0.003) (0.001) (0.001)
Post-March 15

Pre-Treatment Mean 0.714 0.762 0.714 0.762
among Households
with Children

Observations 636,401 636,401 633,490 633,490
Note: Sample: adults between ages 18–65. All models include state and month fixed effects and control for age,
education, and sex of individual. Robust standard errors in parentheses. * p < 0.05, ** p < 0.01, *** p < 0.001.

28
Table 6: Seasonality Test: Difference-in-Differences Estimates of the Effect of July 2020
Treatment on Employment Outcomes (CPS, January 2020 through December 2020)
Binary Treatment Continuous Treatment
($100s of Net Monthly Benefit Value)
1: Employed 2: Active in 1: Employed 2: Active in
Labor Force Labor Force
Household with Children 0.039*** 0.034*** 0.001 0.001*
(0.003) (0.003) (0.001) (0.001)

Household with Children -0.001 -0.004 -0.000 -0.001
X Post-July 15 (0.003) (0.003) (0.001) (0.001)

Pre-Treatment Mean 0.704 0.767 0.704 0.767
among Households with
Children

Observations 705,905 705,905 702,853 702,853
Note: Sample: adults between ages 18–65. All models include state and month fixed effects and control for age,
education, and sex of individual. Robust standard errors in parentheses. * p < 0.05, ** p < 0.01, *** p < 0.001.

29
Table 7: Difference-in-differences Estimates of the Effect of the CTC Expansion on Employment
Outcomes Using the Callaway and Sant’Anna (2020) Methodology and Defining Treatment Group
by the Additional Monthly CTC Payment Received (CPS, January 2021 through December 2021)
Effect Estimate
Treatment-Group:
Monthly Additional Treated Units 1: Employed 2: Active in Labor Force
CTC Payments
0.004 0.006
All Treated Households 208,572
(0.004) (0.004)
$100 6,701 -0.009 -0.006
(0.013) (0.012)
$125 17,000 -0.011 -0.009
(0.009) (0.009)
$150 10,787 0.007 0.009
(0.014) (0.013)
$175 39,616 0.0003 0.004
(0.007) (0.007)
$200 5,355 0.006 0.011
(0.019) (0.019)
$225 7,123 0.022 0.035
(0.019) (0.017)
$250 17,930 -0.010 0.0001
(0.009) (0.009)
$325 11,419 -0.008 -0.003
(0.013) (0.011)
$350 24,882 -0.005 -0.001
(0.009) (0.008)
$375 11,787 -0.003 0.001
(0.011) (0.01)
$525 17,082 0.007 0.012
(0.010) (0.010)
$675 6,182 0.024 0.025
(0.016) (0.015)
Note: Sample: adults between ages 18–65. All models include controls for age, education, and sex of household head.
Each treatment group is defined by the additional monthly CTC payments as a result of the CTC expansion– rounded
to the nearest $25. Presented treatment groups are those with at least 1% of the sample. Robust standard errors in
parentheses are clustered at the state level. * p < 0.05, ** p < 0.01, *** p < 0.001.

30
Table 8: Difference-in-Differences Estimates of the Effect of the CTC Expansion on Employment
Outcomes Using the Callaway and Sant’Anna (2020) Methodology and Defining Treatment Group
by The Number and Age of Children (CPS, January 2021 through December 2021)
Effect Estimate
Treatment- Treatment-
Group: Group: 2: Active in Labor
Treated Units 1: Employed
Children ages: Children ages: Force
0 ≤ 𝑥𝑥 < 6 6 ≤ 𝑥𝑥 < 18

All Treated Households 208,572 0.004 0.006
(0.004) (0.004)

0 1 46,206 0.002 0.003
(0.005) (0.005)

0 2 37,703 0.001 0.002
(0.006) (0.005)

0 3 13,158 -0.0003 0.003
(0.006) (0.006)
0 4 3,523 -0.0001 0.0001
(0.007) (0.007)
1 0 31,145 -0.003 0.001
(0.006) (0.005)
1 1 21,477 0.006 0.006
(0.008) (0.007)
1 2 11,045 -0.001 0.0001
(0.006) (0.006)
1 3 3,990 -0.0001 0.002

(0.006) (0.007)
2 0 15,782 -0.003 -0.0007

(0.007) (0.006)
2 1 6,402 -0.001 0.0004

(0.007) (0.007)
2 2 2,649 0.001 0.002
(0.007) (0.007)
Note: Sample: adults between ages 18–65. All models include controls for age, education, and sex of household head.
Each treatment group is defined by the additional monthly CTC payments as a result of the CTC expansion– rounded
to the nearest $25. Robust standard errors in parentheses are clustered at the state level. * p < 0.05, ** p < 0.01, *** p <
0.001.

31
Table 9: Difference-in-Differences Estimates of the Effect of the CTC Expansion on Employment
Outcomes Using the Callaway and Sant’Anna (2020) Methodology and Defining Treatment Group
by the Change in the Relative Wage (CPS, January 2021 through December 2021)
Effect Estimate
Treatment-Group:
Change in the Relative Treated Units 1: Employed 2: Active in Labor Force
Wage
All Treated Households 208,572 0.004 0.006
(0.004) (0.004)
-1% Δ in Relative Wage 18,263 -0.0001 0.003
(0.007) (0.006)
-2% Δ in Relative Wage 27,455 0.001 0.004
(0.006) (0.006)
-3% Δ in Relative Wage 41,780 -0.001 0.002
(0.005) (0.005)
-4% Δ in Relative Wage 16,830 0.003 0.006
(0.007) (0.006)
-5% Δ in Relative Wage 22,265 -0.002 0.002
(0.006) (0.006)
-6% Δ in Relative Wage 16,686 0.003 0.004
(0.007) (0.007)
-7% Δ in Relative Wage 8,735 0.0003 0.003
(0.006) (0.007)
-8% Δ in Relative Wage 3,869 0.002 0.004
(0.007) (0.007)
≤ -9% Δ in Relative Wage 2,174 -0.0001 0.002
(0.007) (0.007)
Note: Sample: adults between ages 18–65. All models include controls for age, education, and sex of household head.
Each treatment group is defined by the additional monthly CTC payments as a result of the CTC expansion– rounded
to the nearest $25. Robust standard errors in parentheses are clustered at the state level. * p < 0.05, ** p < 0.01, ***
p < 0.001.

32
Figure 1: Mean Net Gain in CTC Benefits per Month by the Number of Children in Household
and Family Type

33
Figure 2: Trends in Employment and Labor Force Participation by Presence of Child in Household
from April 2021 to December 2021, Current Population Survey

34
Figure 3: Trends in Employment and Labor Force Participation by Presence of Child in Family
and Income Group from April 2021 to December 2021, Current Population Survey

Note: This figure presents both the employment and labor force participation rates from the monthly CPS among
respondents between the ages of 18 and 65 from April 2021 to December 2021. These rates are split between childless
households and those with at least one child present in the household, as well as three income groups, those with
household incomes less than $35,000, those with household incomes between $35,000 and $74,999, and those with
household incomes greater than or equal to $75,000.

35
Figure 4: Heterogeneous Effects by Income Bin (U.S. Current Population Survey, April –
December 2021)

Note: Continuous treatment indicator represents the effect of a $100 net increase in monthly benefit value.

36
Figure 5: Heterogeneous effects by income bin (U.S. Census Household Pulse Survey, April –
December 2021)

Note: Continuous treatment indicator represents the effect of a $100 net increase in monthly value. “Treated”
represents the treatment effect (effect of the CTC) on the treated (respondents reporting receipt of the CTC).

37
Figure 6: Event Study of the Effect of CTC Expansion on Labor Force Participation and
Employment using the Current Population Survey from April 2021 to December 2021.

Note: Periods define the number of months since treatment occurred. The first month of treatment is in period zero.
Bars represent a 95% confidence interval. Treatment is defined as eligibility for the CTC after the expansion occurred.

38
39

File and source

File: effects-of-the-expanded-child-tax-credit-on-employment-outcomes-evidence-from-real-world-data-from-april-to-december-202.pdf
Size: 539,105 bytes
SHA-256: 978eed4f674dfae374f7f75de0f0d0db3e6120c0ecc9bda73ea5d7fe6b69cf26
Our copy: effects-of-the-expanded-child-tax-credit-on-employment-outcomes-evidence-from-real-world-data-from-april-to-december-202.pdf
Original: www.nber.org