The Missing Baby Bust The Consequences Of The Covid 19 Pandemic For Contraceptive Use Pregnancy And Childbirth Among Low
Summary
A peer-reviewed article, The Missing Baby Bust: The Consequences of the COVID-19 Pandemic for Contraceptive Use, Pregnancy, and Childbirth Among Low-Income Women, by Martha J. Bailey, Lea Bart and Vanessa Wanner Lang, published in Population Research and Policy Review (2022) 41:1549–1569. The paper uses survey and administrative data from the Michigan Contraceptive Access, Research, and Evaluation Study (M-CARES), conducted at Planned Parenthood health centers in Michigan, to examine how reduced access to reproductive health care in 2020 affected contraceptive efficacy. The abstract states that the results predict a decline in births of 1.1 percent in 2021 for low-income women, and that accounting for reduced abortion access implies birth rates may rise. The article reviews Title X funding changes and prior studies, and it 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
Cite this document
“The Missing Baby Bust The Consequences Of The Covid 19 Pandemic For Contraceptive Use Pregnancy And Childbirth Among Low.”
Original at doi.org. Archived at Pandemic Darlings, https://pandemicdarlings.com/source-documents/the-missing-baby-bust-the-consequences-of-the-covid-19-pandemic-for-co-516b052ec6/
, SHA-256 80527cf188a13754ba91eb1df3f5e9f76a487de7e54ed4232937e50df483e2b8.Full text
Population Research and Policy Review (2022) 41:1549–1569
https://doi.org/10.1007/s11113-022-09703-9
ORIGINAL RESEARCH
The Missing Baby Bust: The Consequences of the COVID‑19
Pandemic for Contraceptive Use, Pregnancy, and Childbirth
Among Low‑Income Women
Martha J. Bailey1 · Lea Bart2 · Vanessa Wanner Lang3
Received: 26 May 2021 / Accepted: 2 February 2022 / Published online: 2 March 2022
© The Author(s), under exclusive licence to Springer Nature B.V. 2022
Abstract
Multiple episodes in US history demonstrate that birth rates fall in response to reces-
sions. However, the 2020 COVID-19 recession differed from earlier periods in that
employment and access to contraception and abortion fell, as reproductive health
centers across the country temporarily closed or reduced their capacity. This paper
exploits novel survey and administrative data to examine how reductions in access
to reproductive health care during 2020 affected contraceptive efficacy among low-
income women. Accounting for 2020’s reductions in access to contraception and the
economic slowdown, our results predict a modest decline in births of 1.1 percent in
2021 for low-income women. Further accounting for reductions in access to abortion
implies that birth rates may even rise for low-income women. These results also sug-
gest that already economically disadvantaged families disproportionately affected by
the COVID-19 economy will experience a large increase in unplanned births.
Keywords Contraception · Long-acting reversible contraceptives (LARCs) ·
Fertility rates · Pregnancy · COVID-19 · Pandemic · Recession
Introduction
Multiple episodes in US history demonstrate that pregnancies and birth rates fall in
response to economic uncertainty and downturns (Autor et al., 1990; Black et al.,
2013; Buckles et al., 2019, 2021; Cherlin et al., 2013; Currie et al., 2014; Gemmill &
* Martha J. Bailey
marthabailey@g.ucla.edu
1
Department of Economics, University of California Los Angeles, 315 Portola Plaza, Bunche
Hall 9349, Los Angeles, CA 90095, USA
2
Department of Economics, University of Michigan, Ann Arbor, USA
3
Population Studies Center, Institute for Social Research, University of Michigan, Ann Arbor,
USA
1 3
Vol.:(0123456789)
1550 M. J. Bailey et al.
Hartnett, 2020; Kearney & Levine, 2020a; Lindo, 2010; Schaller et al., 2020; Schnei-
der & Hastings, 2015). Using this historical relationship between recessions and birth
rates, researchers projected early in the 2020 pandemic that changes in the demand
for children would produce a massive baby bust, reducing US births by 300,000 to
500,000 (Kearney & Levine, 2020a, b).
Consistent with this projection, a 2020 survey showed that many US women
planned to delay or avoid childbearing as the economy slowed during the COVID-
19 pandemic (Lindberg et al. 2020a). Plans to reduce or delay childbearing were
especially pronounced among lower-wage and minority women, who were dispro-
portionately affected by the employment contraction (Bernstein & Jones, 2020;
Chetty et al., 2020), with 37 percent of women with incomes below 200 percent of
the federal poverty line wanting to delay or avoid childbearing compared to 32 per-
cent among women with higher incomes (Lindberg, Kost, et al. 2020b). Nearly half
of Black and Hispanic women reported plans to delay or avoid childbearing, com-
pared to just 28 percent of White women (Lindberg, Kost, et al. 2000b).
But the 2020 recession differed from previous recessions. One especially relevant
difference is that access to contraception and reproductive health services fell, as
health centers canceled or limited appointments in accordance with social distancing
guidelines and patients limited in-person interactions. According to the 2020 Gutt-
macher Survey of Reproductive Health, 33 percent of women had to delay or cancel
visiting their health care provider or had trouble getting their birth control due to
the pandemic. These challenges were also more acute among poorer and minority
women: 32 percent of women with incomes less than 200 percent of the poverty
line reported worrying about affording or obtaining contraceptives during the pan-
demic relative to 23 percent of women with higher incomes, and 29 percent of Black
women and 38 percent of Hispanic women reported increased worry, compared to
only 23 percent of White women (Lindberg, Kost, et al. 2000b). In short, at the same
time changes in the economy reduced the demand for children, the supply of con-
traceptives and access to abortion fell and likely moderated the baby bust (Bailey,
2010; Easterlin et al., 1980; Willis, 1973).
This paper quantifies the consequences of reduced access to reproductive health
care due to COVID-19 restrictions using novel data from the Michigan Contracep-
tive, Access, Research, and Evaluation Study (M-CARES). When the COVID-19
pandemic reduced appointments at Planned Parenthood of Michigan (PPMI) health
centers in March 2020, M-CARES was in the field and had been recruiting partici-
pants since August 2018. Like women hardest hit by the pandemic, patients at PPMI
have lower rates of private health insurance, lower incomes, and are more likely to
be racial and ethnic minorities (Planned Parenthood of Michigan, 2021; Planned
Parenthood Federation of America, 2019, 2020). We refer to women seeking care
at PPMI as “Title X patients,” because they received subsidized services for repro-
ductive health care through the Title X Public Health Service Act when M-CARES
began.1 Participants in M-CARES reported which birth control methods they used
in the month prior to their visit and consented to release their billing records at
1
Title X of the Public Health Service Act allocates federal funding to health centers to provide sub-
sidized reproductive health services and contraception to low-income and uninsured women. In 2018,
Title X served around 3.4 million women nationally. Planned Parenthood was one of Title X’s largest
providers in 2018, serving over 70 percent of Michigan’s Title X patients. In 2019, Planned Parenthood
13
The Missing Baby Bust: The Consequences of the COVID‑19 Pandemic… 1551
PPMI. This combination of survey and administrative data provide a unique per-
spective on contraceptive method use before and after a typical Title X visit. These
unique data allow us to calculate the changes in contraceptive efficacy and reduc-
tions in expected pregnancies resulting from a typical Title X visit. We then use this
information to project changes in 2021 childbearing for the Title X population due
to reductions in access to Title X health centers.
The results show that a typical Title X visit resulted in contraceptive method
changes expected to result in eight fewer pregnancies for every 100 women seek-
ing care. Translating this change in pregnancies into expected births, we find that
reductions in access to contraception due to COVID-19 restrictions should largely
offset decreases in childbearing due to the COVID-19 economic slowdown for low-
income women. This methodology projects a much smaller decline in births of 1.1
percent in 2021 for Title X women than when accounting for economy alone. Fur-
ther accounting for reductions in access to abortion due to COVID-19 restrictions
implies that birth rates could even rise among low-income women in 2021. Taken
together, these results anticipate that already disadvantaged families disproportion-
ately affected by the COVID-19 economy will not experience a large baby bust but
an increase in unplanned births.
Background: How Access to Contraception Affects Use, Pregnancy,
and Childbearing Among Low‑Income Women
In most high-income countries, governments make contraceptives free or inexpen-
sive. In contrast, out-of-pocket costs for contraception in the US are closely linked
to health insurance. Before the Patient Protection and Affordable Care Act (ACA)
passed in 2010, US insurers often required patients to pay all or a significant part
of the cost of birth control out-of-pocket. These costs for the highly effective, long-
acting reversible contraceptives (LARCs) such as IUDs and implants were prohibi-
tively high. Before the ACA, women with insurance could be charged upwards of
$1,000 out-of-pocket for an IUD. The 2010 ACA included a “contraceptive coverage
mandate” which required insurance plans to cover FDA-approved forms of contra-
ception with no cost-sharing.
However, the ACA did not reduce prices for the roughly one third of non-elderly
US women without private insurance (Kaiser Family Foundation, 2020). In some
states, prices for contraception have increased over the last two decades as political
initiatives increasingly link funding for contraception to the hot-button issue of abor-
tion (Shorto, 2006). Since the ACA, state legislatures have reacted both by reducing
Footnote 1 (continued)
withdrew from the Title X program in response to Trump Administration requirements. Throughout this
paper we use 2018 estimates of the Title X population as the best estimates for the Title X population
and Planned Parenthood clients, and references to the Title X clinics include Planned Parenthood health
centers, although Planned Parenthood is no longer in the program. See section III for more detail on how
PPMI adjusted its fee scale and services after its withdrawal from Title X.
13
1552 M. J. Bailey et al.
their public spending on reproductive health and limiting the types of organizations
eligible to receive public funds (NYT Editorial Board, 2019).
In 2019, the Trump Administration rewrote national funding guidelines to impose
additional restrictions on the recipients of federal Title X funds, including a fund-
ing ban for providers referring patients to abortion providers and a requirement that
recipients of federal funds physically separate sites that provide non-abortion repro-
ductive health services from those that provide abortion. The new guidelines suc-
ceeded in causing large Title X providers like Planned Parenthood to withdraw from
the Title X program in 2019. Even with the Trump Administration guidelines in
effect for only five months, 844,803 fewer patients received care through Title X
than in the previous year—2.7 million women in 2019 versus the 3.4 million served
in 2018 (Fowler et al., 2020). Throughout 2020, Title X served just 1.3 million
women, with 63 percent of the decrease relative to 2018 attributed to the new Title
X guidelines (Fowler et al., 2021).
Several quasi-experimental studies suggest that changes in access to subsidized
reproductive health services should affect the take-up of contraceptives and subse-
quent childbearing. Bailey’s (2012) analysis of the expansion in federally funded
family planning programs in the early years of Title X finds a reduction in US birth
rates overall by 1.4 to 2.1 percent and by 19–30 percent among the most disadvan-
taged women who took up Title X services. Kearney and Levine (2009) show that
state-level expansions in Medicaid eligibility for family planning services in the
1990s and 2000s increased the use of contraception and reduced childbearing by
8.9 percent among newly eligible women. After Colorado—in collaboration with a
donor—made LARCs free for all women in 2009, Lindo and Packham (2017) find
that the teen birth rate fell by 6.4 percent within five years. Most recently, the ACA’s
requirement that private health insurance cover contraceptives has been shown to
have decreased out-of-pockets costs and increased use of the most expensive and
effective contraceptive methods (Becker, 2018; Carlin et al., 2016; Dalton et al.,
2020; Heisel et al., 2018). Finally, and especially relevant for this study, Lu and
Slusky (2019) show that closures of women’s health centers in Texas increased
birthrates by 1.2 percent. Similarly, Venator and Fletcher (2021) find that clinic clo-
sures in Wisconsin reduced the number of abortions by 30.7 percent and increased
births by 3.2 percent.
While these studies show that increasing access and reducing the costs of con-
traceptives tends to increase take-up and reduce birth rates, their implications for
the current pandemic are more difficult to gauge. One reason is that, even before the
pandemic, many more low-income women are using LARCs than in the past. LARC
use among low-income women using some type of contraception grew from 5.6 per-
cent in 2008, just before the Colorado policy began, to 16.4 percent by 2014 (Bailey
& Lindo, 2018). Higher use of LARCs allows women to delay their reproductive
health visits during the pandemic with little consequence for pregnancy, because
LARCs require fewer check-ups and no refills. A second reason is that the demand
for children is different today than in the periods considered in earlier work. Today,
women’s market work is compensated at much higher rates than when the Title X
program began in the 1970s. The booming economy of the 1990s and 2000s and
revisions in public assistance programs likely imply differences in the demand for
13
The Missing Baby Bust: The Consequences of the COVID‑19 Pandemic… 1553
children. Finally, the Great Recession, which unfolded concurrently with Colorado’s
free LARC program and the implementation of the ACA, very likely had different
effects on low-income workers than the 2020 COVID-19 economy, which lowered
employment and increased women’s childcare burdens simultaneously (Adams-
Prassl et al., 2020; Alon et al., 2020a, b; Bick & Blandin, 2020; Collins et al., 2020).
Data and Methods: Quantifying Changes in Contraceptive Use at Title
X Health Centers
This paper uses novel data from the Michigan Contraceptive Access, Research, and
Evaluation Study (M-CARES). M-CARES is a randomized control trial designed to
test how financial access to contraception affects the lives of low-income women in
today’s policy environment. This study was conducted at Planned Parenthood health
centers in Michigan (PPMI) starting on August 20, 2018. M-CARES randomized
half of study participants to receive vouchers to reduce the cost of contraceptives,
while a control group received the PPMI standard of care and paid out-of-pocket
according to PPMI’s sliding scale. Eligibility for M-CARES required that women
(1) were between ages 18 and 35, inclusive; (2) were at risk of unintended preg-
nancy (not currently pregnant, did not want to become pregnant in the next year,
fecund, and had sex with men); and (3) had out-of-pocket costs for contraceptives
(i.e., uninsured, or their insurance did not cover contraception). The study did not
require that participants be visiting PPMI to obtain contraception—participants
were recruited if they came to PPMI for any reason and met the eligibility criteria.
After Planned Parenthood withdrew from the Title X program in November 2019,
PPMI used private donations and its own resources to subsidize its services using a
slightly modified sliding scale that was around 20 percent less generous than before
the withdrawal.2 Although women below the poverty line had no out-of-pocket costs
previously, this changed on November 4, 2019, for those ages 22 or older. Because
this group was only recruited after November 2019, they are underrepresented in the
M-CARES sample. As described in subsequent sections, our analysis uses weights
to address this data limitation.
2
To the extent that withdrawal from Title X reduced women’s ability to afford more expensive contra-
ceptives (i.e., decreased contraceptive efficacy gains from a patient visit), our approach should understate
the changes in efficacy due to COVID-19 restrictions relative to the fully funded Title X period. To test
how much this change in fee scale affects our estimated efficacy rates, we compute changes in contracep-
tive efficacy for our sample for individuals recruited before and after November 4, 2019. During the Title
X period, the average contraceptive efficacy increased by 0.10 after a PPMI visit (unweighted). After the
Title X period, average contraceptive efficacy increased by 0.09 after a PPMI visit, which is not statisti-
cally different from the estimate in the Title X period (p = 0.851). We, therefore, conclude that the impact
of the withdrawal from Title X on our estimates is minimal.
13
1554 M. J. Bailey et al.
Analytic Sample
The goal of this paper is to quantify the effects of a typical Title X reproductive
health visit on contraceptive efficacy. To that end, this analysis uses data only from
the 1,405 women assigned to the control group who were recruited between August
20, 2018, and March 13, 2020, when recruitment was suspended due to safety con-
cerns during the COVID-19 pandemic. These women received the Planned Par-
enthood standard of care and paid on the sliding scale, allowing us to character-
ize changes in the use of contraception that occur in the current policy environment
without an experimental intervention.3 After excluding a handful of individuals who
withdrew from the study, could not be linked to billing records, or who were missing
responses about pre-visit birth control use or other key demographic characteristics,
our final sample is 1,326 women. (See Appendix Figure A1 for details). In short, the
sample is restricted to data collected prior to the COVID-19 pandemic and includes
four and a half months of data collected after Planned Parenthood withdrew from
Title X on November 4, 2019.
Table 1 shows the characteristics of the M-CARES sample (column 1) relative
to the national population (column 2, 2017–2019 NSFG). The M-CARES sample is
more likely to be in their early 20s, less likely to be a racial or ethnic minority, and
more likely to have lower income (below 200 percent FPL) and be uninsured relative
to the NSFG sample of 18- to 34-year-old women. Because the M-CARES sam-
ple was recruited at Planned Parenthood, its participants are (unsurprisingly) more
likely to use contraception than the national population. The M-CARES sample dif-
fers less from the national population of Title X patients (column 3). While similar
in age to Title X patients, the M-CARES sample is less likely to be Hispanic, have
health insurance, and have incomes lower than the federal poverty line (FPL). The
underrepresentation of very low-income and insured individuals reflects the study
eligibility criterion 3—women below the FPL or with insurance were less likely to
have out-of-pocket costs for contraception and, consequently, were less likely to be
eligible to participate in M-CARES. The underrepresentation of Hispanic women
reflects their underrepresentation in Michigan (see Appendix Table A1 for compari-
sons of Michigan to the US using data from the American Community Survey). The
M-CARES sample also uses a different mix of contraception than the national sam-
ple, which likely reflects to some extent the fact that we measure contraceptive use
in this table prior to their appointment, as opposed to after they see their provider.
Later in the paper, we discuss reweighting our sample to resemble the national Title
X population.
Although the sample for this paper is entirely drawn from one state, Michigan’s
reproductive health policies are shared with many other states. Michigan’s Medicaid
expansion program, Healthy Michigan, supplements the state’s Medicaid program
3
The usual policy environment applies the Title X sliding scale to patient costs. Patients with incomes at
101–150 percent of the federal poverty line (FPL) pay 25 percent; with incomes 151–200 percent pay 50
percent; with incomes 201–250 percent pay 75 percent; and with incomes above 250 percent pay 100 of
PPMI prices.
13
The Missing Baby Bust: The Consequences of the COVID‑19 Pandemic… 1555
Table 1 Characteristics of analysis sample, women ages 18–34
(1) (2) (3) (4)
M-CARES 2017–9 2018 Title X MCARES
Sample NSFG Patient Charac- Reweighted to
teristics Title X Char-
acteristics
Observations 2665 6141 ~ 3.4 million 2665
Age 18–19 0.096 0.105 0.135 0.135
Age 20–24 0.393 0.274 0.354 0.354
Age 25–29 0.326 0.319 0.304 0.304
Age 30–34 0.186 0.302 0.207 0.207
Non-Hispanic White 0.673 0.565 0.333 0.333
Non-Hispanic Black 0.135 0.154 0.195 0.195
Hispanic any race 0.099 0.212 0.340 0.340
Other/Not Reported 0.093 0.069 0.132 0.132
Less than 100% FPLa 0.080a 0.236 0.667 0.667
101–150% FPL 0.398 0.129 0.149 0.149
151–200% FPL 0.246 0.115 0.073 0.073
201–250% FPL 0.124 0.080 0.035 0.035
251 + % FPL 0.151 0.439 0.076 0.075
Insured 0.073 0.863 0.583 0.583
Uninsured 0.927 0.137 0.417 0.417
b b
Sterile 0.071 0.028
LARC 0.137 0.135 0.171 0.085
Birth control pills 0.298 0.198 0.251 0.245
Condoms 0.202 0.108 0.162 0.122
Withdrawal 0.016 0.046 0.025 0.010
Other hormonal 0.105 0.045 0.179 0.306
Other non-hormonal 0.003 0.039 0.010 0.001
No method 0.218 0.358 0.174 0.218
M-CARES participants recruited from August 20th, 2018, to March 13th, 2020 ages 18–34. For
M-CARES participants, contraceptive method is for the period prior to the Planned Parenthood visit.
LARC includes implant and IUD. Condoms refer to male condoms. Other hormonal methods include the
shot, patch, and ring. Other non-hormonal methods include vasectomy, female condoms, NFP, sponge,
diaphragm, and spermicide. No method includes no reported method and abstinence. Estimates of the
2018 Title X participants are derived from Fowler et al. (2019), Exhibits 4, 7, 15, 16, and 18. The age dis-
tributions are the distribution of the ages of female Title X users between ages 18 and 34, calculated using
Exhibit 4. Race distributions are for female Title X users of all ages, calculated using Exhibit 7. Income
distributions are for all Title X users with income above 100% of FPL, calculated using Exhibit 15
a
This group is excluded from the M-CARES sample before November 2019 and so is underrepresented
b
This group is excluded from the M-CARES sample because these women are not at risk of pregnancy
and covers women who are not pregnant who earn up to 133 percent of the federal
poverty line. Michigan offers a 6 percent state EITC and has per capita spending on
public welfare close to the national average (US Census Bureau, 2019). As of 2019,
13
1556 M. J. Bailey et al.
Michigan’s childcare subsidy assistance had the lowest initial eligibility threshold
nationally (Dwyer et al., 2020).
Computing Changes in Expected Childbearing from Reductions in Title X Services
This paper computes the expected change in contraceptive use before and after a
typical PPMI appointment in several steps. First, we calculate the expected change
in contraceptive efficacy during a typical Title X visit. This allows us to compute the
change in the expected number of pregnancies, holding sexual activity fixed. Second,
we compute the pregnancy-to-birth transition rate using the NSFG, which accounts
for the fact that not all pregnancies result in live births. Third, we use estimates from
the literature to approximate changes in the demand for children that covary with the
business cycle. Our last step combines the estimated increase in childbearing (due to
reductions in reproductive health care) with the estimated decrease in childbearing
(due to the economic downturn) to predict the change in births for Title X women
in 2021 relative to 2020. The following discussion describes each of these steps in
more detail.
Computing Changes in Contraceptive Efficacy
A key outcome in this study is the change in contraceptive efficacy, which describes
how patients’ choices of contraception changed during a typical visit to PPMI. We
calculate this change using two data sources. A screening survey determined eligi-
bility for M-CARES and asked eligible women seeking care at PPMI about their use
of contraception in the month before they came to PPMI. All study participants com-
pleted the screening survey. The screening survey is complemented by PPMI billing
records, which all study participants consented to allow PPMI to release records to
M-CARES for research purposes. These records characterize any transactions that
the patient had with PPMI during and after her visit and can be linked to survey
responses for over 99.9 percent of participants. Combining these data sources pro-
vides a highly accurate description of birth control method use before and after the
visit to PPMI.
We compute contraceptive efficacy as 1—the CDC’s expected one-year method
failure rate with typical use. The expected contraceptive failure rates represent the
share of women who are expected to experience a pregnancy within one year with
typical use (Trussell, 2011). CDC failure rates range from 0.0005 for a subdermal
implant, 0.09 for birth control pills, and 0.18 for the male condom. Women who
report using multiple forms of contraception are assigned the efficacy rate corre-
sponding to their most effective method. For example, if the respondent reported
using both condoms and birth control pills, we assign contraceptive efficacy as 0.91
(1–0.09), which uses the failure rate of typical use associated with birth control pills.
Pre-visit efficacy is calculated using the contraceptive methods reported by the
patient in the pre-visit survey, which is taken by all women who enroll. Post-visit
efficacy is calculated using PPMI billing records. If women purchased contracep-
tives from PPMI up to 100 days after their visit, we code the post-visit method to
13
The Missing Baby Bust: The Consequences of the COVID‑19 Pandemic… 1557
be the method purchased. We use purchases up to 100 days after a visit, because
some methods require a follow-up. For instance, switching to an IUD requires both a
pregnancy test and typically a follow-up visit. For all women with no contraceptive
purchases in PPMI data, we assume they continued using the method reported in the
pre-visit survey. For instance, if a woman reported using birth control pills before
her visit in the survey but did not purchase any pills, we assume she continued using
birth control pills (presumably purchased before her visit). Women who report using
abortion, miscarriage, Plan B, or abstinence as a method are coded as using “no pre-
ventative method.”
Empirical evidence supports the assumption that if respondents are missing data
on contraceptive purchases in the post-period, they are using their pre-visit method.
Using available survey responses for women who report using the pill before their
PPMI visit and did not purchase birth control pills in PPMI billing records, 98 per-
cent report still being on the pill immediately after their appointment in our post-
visit survey. This is consistent with Planned Parenthood’s policy that patients are
permitted to purchase up to 13 packs of pills when they receive a new prescrip-
tion. Women using the pill prior to their PPMI visit may have a supply of pills at
home and continue using the pill despite not filling a new prescription. While we do
not know the continuation rate for women who do not have billing records and did
not respond to our survey after their appointment, the high rate of pill continuation
among women using the pill at enrollment for the women we do observe suggests
our assumption is reasonable.
Given the differences between our study population and the Title X population
nationally, we use entropy balancing to reweight the sample such that the age, race/
ethnicity, insurance, and income characteristics of M-CARES participants match
those in the national Title X population captured in the 2018 Health and Human
Services (HHS) Annual Report (Fowler et al., 2019; Hainmueller, 2011).4 We gener-
ate 95-percent confidence intervals by bootstrapping our entropy-balanced estimates
using replacement (1000 replications). Identifying the 0.025 and 0.975 percentiles
from the distribution of 1000 estimates provides the 95-percent confidence interval
(Efron & Tibshirani, 1993). Table 1 shows that the reweighted M-CARES sample
(column 5) is similar in key characteristics to the 2018 Title X reports (column 4).
This reweighting ensures that the M-CARES sample resembles the age, race, insur-
ance, and income characteristics of the Title X patients, although it does not guar-
antee that unobserved Title X patients characteristics are also balanced. Importantly,
we do not reweight the M-CARES sample to resemble the Title X patients in terms
of contraceptive use, because contraceptive use of Title X patients results from their
appointment at a Title X provider. Accordingly, the reweighted M-CARES sam-
ple is less likely to use birth control and LARCs, as measured prior to their visit,
than the 2018 Title X patient population. See Online Appendix B for more details
about reweighting the M-CARES sample to resemble national Title X patient
characteristics.
4
Unfortunately, we do not have microdata for all Title X patients, which means that we cannot use
inverse propensity score reweighting.
13
1558 M. J. Bailey et al.
Translating Expected Pregnancies into Expected Births
In 2008, an estimated 65 percent of pregnancies resulted in a live birth, 17 percent
in miscarriage, and 18 percent in abortion (Ventura et al., 2012; Zolna & Lindberg,
2012). However, the likelihood that a pregnancy will end in a live birth may differ
for the M-CARES sample of low-income women and also have changed between
2008 and today.
The 2017–2019 NSFG allows us to estimate the share of pregnancies that result
in childbirth (pregnancy-to-birth transition rate) for the Title X population. For every
woman with a pregnancy that began at least 10 months before the NSFG interview
and no earlier than 2015, the NSFG’s pregnancy history identifies her contraceptive
method at the time of conception and the outcome of the pregnancy (e.g., live birth,
miscarriage, abortion),5 as well as information on age, race, poverty status, and
health insurance at the time of the interview. For women with multiple completed
pregnancies in this time period we use the contraceptive method from their most
recent pregnancy. For women without a completed pregnancy during this period, we
use contraceptive method as of the January of the year before they were interviewed.
We then use entropy balancing to reweight the NSFG sample to match the Title X
population in terms of contraceptive method type, age, race/ethnicity, income dis-
tributions, and insurance status. In particular, reweighting to match the distribution
of contraceptive use accounts for the fact that Title X patients are more likely to
use birth control and more likely to desire avoiding pregnancy and childbirth than
the average woman in the population. A final step in this process is that we inflate
reported abortions and miscarriages to account for well-known underreporting in
the NSFG (Lindberg et al., 2020a; Linnakaari et al., 2019). Appendix Table C1
shows balance in characteristics after this reweighting and a resulting pregnancy-
to-birth transition rate for Title X patients of 0.479, which is shown in Appendix
Table C2. The reweighted estimate indicates that 47.9 out of every 100 pregnancies
to Title X clients resulted in a live birth. This rate is slightly lower than what we esti-
mate for the overall population in the 2017–2019 NSFG (49.5 after adjustments for
misreporting).6 The true transition rate may also differ because access to abortion
changed during the pandemic. Many southern and midwestern states declared abor-
tion a non-essential service in 2020, which effectively created temporary abortion
bans.7 In addition, 19 states required medication abortion to be issued in person by
5
We exclude pregnancies that did not begin at least 10 months before the interview because the out-
comes are unknown.
6
Reweighting on contraceptive method type is primarily responsible for the lower the transition rate for
the Title X population relative to the overall population. If we reweight the NSFG so that all characteris-
tics except for birth control use (income, insurance, age, and race) correspond to the Title X population,
we estimate a pregnancy-to-birth transition rate of 58.7 percent, reflecting a younger, lower-income, less
white, and more uninsured Title X population relative to national levels. However, if we reweight using
these characteristics and pre-visit contraceptive use, our estimate falls to 47.9 percent, which we report.
7
See https://www.kff.org/coronavirus-covid-19/issue-brief/state-action-to-limit-abortion-access-during-
the-covid-19-pandemic (accessed October 22, 2021).
13
The Missing Baby Bust: The Consequences of the COVID‑19 Pandemic… 1559
a clinician, which was difficult or impossible to arrange during the pandemic.8 If the
true Title X pregnancy-to-birth transition rate is higher than we estimate, reflecting
lower access to abortion care during the pandemic, our results will underestimate
the increase in births to Title X women due to changes in the supply of contracep-
tion during COVID-19. Under the assumption that COVID-19 shutdowns affected
abortion access similarly to contraceptive access, the pregnancy-to-birth transition
rate could be as high as 73.4 percent, which would reflect all pregnancies ending in
either birth or miscarriage. (See Appendix C for details).
Computing Changes in the Demand for Children
Up to this point, our calculations of efficacy, pregnancy, and childbirth implicitly
hold the demand for contraception and childbirth constant. That is, they assume
that women make no other adjustments in their sexual activity or contraceptive
use within the year aside from what takes place during their PPMI visit. However,
research documents that recessions could both increase and decrease childbearing
in theory. On the one hand, childbearing may increase because the opportunity cost
of having a child falls as employment opportunities or wages contract. On the other
hand, reductions in income and increases in economic uncertainty that come dur-
ing recessions tend to decrease childbearing. Empirically, the latter effect tends to
dominate, and fertility rates vary positively with the business cycle in the short run
(Currie & Schwandt, 2014; Lindo, 2010; Schneider, 2015). Improvements in labor
market conditions for women specifically have small or no effects on childbearing,
suggesting that the opportunity cost of childbearing decreases fertility rates around
as much as the income effect increases it for this group (Schaller, 2016). Overall, we
expect that the COVID-19 pandemic reduced the demand for children.
To account for likely shifts in the demand for contraception and children, our
analysis follows Kearney and Levine’s (2020a) methodology, which estimates child-
bearing responses based on previous recessions. The purpose of following their
methodology is to account for the extent women may have changed other behaviors
to reduce childbearing, even if they could not obtain their desired contraceptives
(e.g., abstained from sex). While Kearney and Levine’s methodology was devel-
oped for the US population overall and is not specific to this paper’s sample of low-
income women, it is a transparent and straightforward approximation to measure
shifts in the demand for children on childbearing.
Their calculation assumes that a one-percentage-point increase in the unemployment
rate will reduce birth rates by one percent. Because the unemployment rate for the Title
X population increased by approximately 8 percentage points during the COVID-19
downturn (Chetty et al., 2020), this methodology anticipates an 8 percent decrease in
births for the Title X population due to reductions in the demand for children.9 Given the
8
See https://www.guttmacher.org/state-policy/explore/medication-abortion (accessed October 22, 2021).
9
Chetty et al. (2020) estimate a 20 percent decline in employment for low wage workers due to COVID-
19 as of October 2020. Using an estimated base employment rate of 39.5 percent for Title X workers (see
Online Appendix D), calculated using the March 2020 CPS, a 20 percent decline in employment implies
13
1560 M. J. Bailey et al.
baseline estimate of 178,874 births to Title X patients,10 an 8 percent reduction implies
about 14,350 fewer births to Title X patients nationally as a result of the COVID-19
employment contraction for this group. In addition, Kearney and Levine (2020a) assume
that uncertainty and anxiety due to the pandemic will reduce births by another 15 per-
cent. With this further adjustment, our estimate of the reduction in childbearing due to
demand factors rises to 16,251 fewer births. Online Appendix D provides all details
underlying this calculation.
Our analysis adjusts for changes in the demand for births, including changes in
preferences, sexual activity, or behavior in response to the pandemic economy, by
subtracting the 16,251 expected decrease in births due to reductions in the demand
for children from the expected increase in births due to the change in access to
contraception.
Results: How Title X Health Centers Affect Contraceptive Efficacy
and Birth Rates
Table 2 presents the expected changes in contraceptive use during a typical visit to
PPMI (for comparison, unweighted frequencies and distributions appear in Appen-
dix Tables A2 and A3, and weighted frequencies underpinning calculations in
Table 2 are presented in Table A4). Pre-visit methods are presented in the first col-
umn and ordered from the most to least effective according to CDC’s expected one-
year method failure rate with typical use (presented in parentheses), and post-visit
methods are captured in columns. The final three rows show the share of women
who switched to more effective methods by their pre-visit method. For instance,
among women using an implant before their visit to PPMI, 51 percent continued this
method and 49 percent switched to a less effective method. For women who did not
purchase another method, we conservatively assume they continue to use their pre-
visit method. Aggregating across all methods, roughly 79 percent of women stayed
on the same method after their visit, and 19 percent switched to a more effective
method. Only 2 percent of women switched to a less effective method.
Table 3 summarizes the average change in contraceptive efficacy during a typi-
cal visit to PPMI as captured by 1—the CDC failure rate. We present estimates for
the M-CARES sample as well as estimates of this sample weighted to match the age,
race, income, and insurance characteristics of the Title X patient population nationally.
Footnote 9 (continued)
a 7.9 percentage-point decrease in the employment rate. Assuming the share of the Title X population in
the labor force remains constant, this translates to a 7.9 percentage-point increase in the unemployment
rate for the Title X population due to COVID-19.
10
The number of female Title X and Planned Parenthood users (about 3.4 million in 2018) multiplied
by the estimated birth rate for Title X clients obtained from the NSFG (51.9 births per 1,000 women)
implies that Title X and Planned Parenthood clients gave birth to about 180,000 children per year on
average. See Online Appendix D.
13
Table 2 Method transitions at a typical title X visit
Most effective method within 100 days after enrollmenta
Most effective birth control Implant Vasec- IUD Shot Pill Ring Patch Dia- Condom With- Rhythm No Total
pre-visit method (CDC failure tomy phragm drawal Method
rate, %)b Purchased
Implant (0.02%) 0.44 0.00 0.03 0.44 0.02 0.00 0.00 0.00 0.00 0.00 0.00 0.07 1.00
Vasectomy (0.15%) 0.00 1.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 1.00
IUD (0.20%) 0.00 0.00 0.96 0.03 0.01 0.00 0.00 0.00 0.00 0.00 0.00 0.00 1.00
Shot (6%) 0.00 0.00 0.00 0.99 0.01 0.00 0.00 0.00 0.00 0.00 0.00 0.00 1.00
Pill (9%) 0.01 0.00 0.19 0.00 0.79 0.01 0.00 0.00 0.00 0.00 0.00 0.00 1.00
Ring (9%) 0.00 0.00 0.00 0.00 0.00 0.99 0.00 0.00 0.00 0.00 0.00 0.00 1.00
Patch (9%) 0.00 0.00 0.00 0.00 0.35 0.00 0.65 0.00 0.00 0.00 0.00 0.00 1.00
Condom (18%) 0.02 0.00 0.07 0.14 0.18 0.00 0.00 0.00 0.57 0.00 0.00 0.02 1.00
Withdrawal (24%) 0.00 0.00 0.01 0.44 0.38 0.00 0.00 0.00 0.00 0.16 0.00 0.01 1.00
Rhythm (24%) 0.90 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.10 0.00 1.00
No Methodc (85%) 0.06 0.00 0.02 0.06 0.15 0.08 0.00 0.00 0.00 0.00 0.00 0.62 1.00
Total 0.03 0.00 0.11 0.33 0.30 0.03 0.02 0.00 0.05 0.00 0.00 0.13 1.00
The Missing Baby Bust: The Consequences of the COVID‑19 Pandemic…
Pre-visit method Implant Vasec- IUD Shot Pill Ring Patch Dia- Condom With- Rhythm No Method Total
tomy phragm drawal Purchased
Switched to more effective N/A 0.00 0.00 0.00 0.20 0.01 0.35 N/A 0.41 0.83 0.90 0.38 0.19
Stayed on same method 0.51 1.00 0.96 0.99 0.79 0.99 0.65 N/A 0.59 0.17 0.10 0.62 0.79
Switched to less effective 0.49 0.00 0.04 0.01 0.01 0.00 0.00 N/A 0.00 0.00 N/A N/A 0.02
Cells contain the weighted percentages of individuals remaining on their pre-visit birth control or transitioning to new methods. aPost-enrollment birth control methods
uses data from the PPMI billing records. bPercent of women experiencing an unintended pregnancy within the first year of typical use (Trussell, 2011). cNo reported
method includes abstinence, Plan B, abortion, miscarriage, and no method reported
1561
13
1562 M. J. Bailey et al.
Table 3 Changes in contraceptive efficacy and expected pregnancies at a typical Title X visit
Unweighted Weighted
Average efficacy of contraceptives before visit 0.70 (0.68, 0.72) 0.73 (0.64, 0.79)*
Average efficacy of contraceptives after visit 0.80 (0.78, 0.82) 0.81 (0.75, 0.87)*
Change in efficacy during visit 0.10 (0.07, 0.12) 0.08 (0.05, 0.15)*
Expected new pregnancies if Title X health centers close for 12 months
Female title X patients in Michigan (N = 54,981) 5498 4398
All female title X patients in the US (N = 3,446,504) 344,650 275,720
Sample includes women 18–34 recruited by M-CARES who were randomly assigned to the control
group. Observations missing race, age, income, or pre-visit method use are excluded from the analy-
sis. *Confidence intervals for weighted estimates are estimated using a bootstrap with 1000 iterations.
Expected pregnancies are obtained by multiplying the population by the change in average efficacy. See
text for more details on M-CARES sample and reweighting procedure
Prior to their PPMI visit, women’s contraceptive efficacy averaged 0.73, indicating
that pregnancies are expected for 27 out of 100 women within a year. After a visit to
PPMI, contraceptive efficacy increased to an average of 0.81, indicating that only 19
pregnancies are expected to occur within one year for every 100 women. 95-percent
confidence intervals in parentheses capture the uncertainty associated with each esti-
mate. Combining the before and after efficacy measures shows that a typical visit to
PPMI is expected to reduce pregnancies by 8 per 100 women within the year using the
weighted estimates. Closing all Title X health centers in Michigan for one year would,
therefore, be expected to result in an additional 4,398 pregnancies for the state’s
54,981 Title X patients. Alternatively, scaling these estimates by the 3,446,504 female
Title X patients nationally in 2018, these estimates predict that a closure of all Title X
health centers for one year would lead to 275,720 additional pregnancies. Using the
pregnancy to childbirth conversion rate of 47.9 percent from section III.B.2 implies
that a one-year shutdown of Title X health centers would increase births in Michi-
gan by 2170 and births nationally by 132,070. Similar calculations based on efficacy
changes by subpopulations can be found in Appendix Tables A5 and A6.
Table 4 uses several data sources to refine these estimates to reflect the actual
magnitude and duration of COVID-19 health center reductions in capacity. Accord-
ing to PPMI administrative records for health centers in M-CARES, PPMI had 60
percent fewer patient visits in April 2020 than in February 2020 due to COVID-
related capacity reductions and social distancing requirements. Assuming this 60
percent reduction in capacity was similar at Title X health centers across the country
in states with lockdowns similar to Michigan’s for the first 3 months of the pan-
demic, we expect that reduced health center capacity during the first 3 months of
COVID-19 shutdowns should result in 27,090 additional pregnancies among the
Title X population and 12,976 more live births.11 Assuming half the reduction in
11
This calculation is as follows: 3,446,504 Title X clients * 0.08 efficacy change * 3/12 of a year *
0.6 reduction * 0.665 of US population under strict business closure and stay-at-home orders similar to
Michigan.
13
The Missing Baby Bust: The Consequences of the COVID‑19 Pandemic… 1563
Table 4 Expected deviations in 2021 pregnancies and births from reductions in Title X health center
operations
A. Deviation in 2021 pregnancies
Michigan Nationally
Title X health centers shutdown for one year 4398 275,720
Estimate of actual Title X capacity reductions 660 29,964
B. Deviation in 2021 births
No change in abortion access No abortion access
Michigan Nationally Michigan Nationally
Title X health centers 2107 132,070 3228 202,379
shutdown for one year
73.8%* 73.8%* 113.1%* 113.1%*
Estimate of actual Title X 316 14,353 484 21,994
capacity reductions
11.1%* 8.0%* 17.0%* 12.3%*
C. Predicted changes in 2021 births after accounting for reductions in the demand for children
No change in abortion access No abortion access
Michigan Nationally Michigan Nationally
Changes in births net of 57 − 1898 225 5743
shifts in demand
2.0%* − 1.1%* 7.9%* 3.2%*
Each cell in Panel A represents how many fewer pregnancies are expected if Title X health centers saw
the corresponding reduction in capacity, assuming that the average improvement in contraceptive efficacy
of going to a Title X health center appointment is 0.08. This increase in efficacy comes from our nation-
ally representatively weighted estimate from the efficacy table. Panel B represents the expected number
of births that would result from these pregnancies, which is obtained by multiplying the expected preg-
nancies by a pregnancy-to-birth transition rate. We present estimates for two such rates: 47.9%, which is
directly estimated from the NSFG as described in the text, and 73.4%, which additionally assumes that
abortion became temporarily unavailable during the COVID-19 shutdowns and all pregnancies ended in
birth or miscarriage. Panel C subtracts the anticipated reduction in births due to the pandemic slowdown
in the economy from the increases in births in panel B
*Numbers in italics are percent changes relative to the estimated number of baseline births to the Title X
population, which was 2854 in Michigan and 178,874 nationally
capacity at Title X health centers was in counties with less stringent lockdowns (14
percent of the US population) for the first 3 months of the pandemic, we expect
that reduced health center capacity resulted in 2874 additional pregnancies for
the Title X population and 1377 more births.12 Together, Table 4 shows that these
12
This second calculation is identical to the first except for that it uses information that 13.9 percent
of US residents were in safer-at-home order counties which experienced smaller reductions in capacity.
Information on safer-at-home orders and lockdowns comes from https://www.naco.org/resources/featu
red/counties-and-covid-19-safer-home-orders (downloaded January 14, 2021).
13
1564 M. J. Bailey et al.
figures imply 29,964 more pregnancies (panel A, row 2, column 2). Using the NSFG
estimate of the pregnancy-to-birth transition rates of 47.9 percent, we expect the
COVID-19 reductions in reproductive health services to result in 14,353 more births
for the Title X population in 2021 (panel B, column 2), an increase of 8.0 percent
from the baseline birthrate. At most, if the shutdown also eliminated access to abor-
tions and all pregnancies ended in birth or miscarriage, this would raise the estimate
to 21,994 additional births, an increase of 12.3 percent over the baseline birth rate
(panel B, column 4).
The estimates assume contraceptive efficacy and abortion access are the only fac-
tors changing during the COVID-19 shutdowns. However, Kearney and Levine’s
(2020a) historical estimates show that the COVID-19 economy should reduce the
demand for children through other channels and, potentially, through lower rates of
sexual activity. Subtracting the expected decrease in births of 16,251 due to demand
changes from employment, uncertainty and anxiety (see section III.B.3) from the
14,353 expected increase in births from changes in the supply of contraception, the
net change in births to the Title X population due to COVID-19 is a decrease of
1.1 percent, or 1898 births below a baseline of 178,874 (panel C, column 2). An
alternative calculation assumes that the COVID-19 shutdowns eliminated access to
abortion and that all pregnancies result in childbirth or miscarriage. In this case,
we subtract the expected decrease in births due to demand changes from the 21,994
additional births expected increase in births from changes in the supply of contra-
ception, which results in a net change in births to the Title X population of 3.2 per-
cent, or 5743 births above the baseline (panel C, column 4).
How much of these changes in childbearing represent mistimed versus unwanted
births? Although this cannot be directly computed, the M-CARES baseline survey
asks women about their ideal number of children using a series of questions devel-
oped by Coombs (1974).13 For the 83 percent of women who respond to the baseline
survey, around one quarter had reached their ideal number of children. As is well
documented, ideals are not always realized, but these seem like reasonable num-
bers to use as a guide. These responses suggest that around 75 percent of births due
to reductions in Title X health center operations are at least one year earlier than
desired, while 25 percent of the births that occur due to reductions in Title X health
center operations would not have occurred in the future.
Study Limitations
This analysis has several important limitations. First, our sample only represents
women who elected to participate in the M-CARES randomized control trial.
Although 75 percent of the women M-CARES approached chose to participate,
13
The Coombs scale introductory question is, “For you personally, what would be the ideal number of
children you would like to have?” The respondent is then asked systematically about what—if the ideal
number is not available—the next choice would be in order to assess preferences not captured in the first
response.
13
The Missing Baby Bust: The Consequences of the COVID‑19 Pandemic… 1565
this study’s conclusions could be altered if the typical visit for the 25 percent of
women who opted out resulted in different changes in contraceptive efficacy.
Second, we assume that changes in contraceptive efficacy from a visit to
PPMI are comparable to those at other types of Title X health centers by using
the term, “typical Title X visits,” throughout the paper. However, Planned Par-
enthood health centers have been found to provide more timely and accessible
contraceptive care than other Title X providers, such as health departments and
federally qualified health centers (FQHCs). Planned Parenthood is more likely to
provide same-day appointments, on-site pill dispersal, same-day IUD or implant
insertion, and the full range of contraceptive methods than health departments or
FQHCs (Hasstedt, 2017). If these differences result in larger contraceptive effi-
cacy changes for a typical PPMI visit than to other Title X providers, our cal-
culations could overestimate the average changes in efficacy and the resulting
increase in births due to reductions in Title X health centers operations. Because
Planned Parenthood accounted for 41 percent of all contraceptive clients at Title
X funded health centers prior to their exit from the Title X program, the differ-
ences between Planned Parenthood visits alone and the average of Planned Par-
enthood and Title X visits is likely minimal (Frost et al., 2017).
Finally, we follow Kearney and Levine’s (2020a) methodology in using previ-
ous recessions to quantify the change in the demand for children. To the extent
that the childbearing response among low-income women during the COVID-19
economy (e.g., due to differential changes in women’s employment, childcare
availability, and school closures) differs from responses to past recessions, this
methodology may misrepresent shifts in demand during the COVID-19 downturn.
Conclusions: Slight or No Decline in 2021 U.S. Birth Rates and Large
Increases in Unplanned Childbirth
Using a novel dataset, this paper describes changes in contraceptive efficacy that
occurs at a typical Title X visit and predicts changes in birth rates for Title X
women in 2021. Accounting for the magnitude and duration of the COVID-19
reductions in health center access, we expect only around 2000 fewer births, or a
1.1 percent decline, for this disadvantaged population—much less than the pro-
jected 8% or 300,000 births for the US population overall (Kearney and Levine’s
(2020a, b). Unless couples use non-medical contraceptive methods like absti-
nence, withdrawal, or condoms with the same effectiveness as medical/hormo-
nal alternatives, the year 2021 could experience a smaller baby bust than pro-
jected. Another conclusion is that shifts in both the demand for children and the
availability of reproductive health services amid the pandemic should increase
the share of pregnancies and births that are unplanned for low-income, uninsured
women.
Although our estimates are specific to the Title X population, the magnitudes
of changes in birth rates in this paper align closely with recent figures released for
the entire US population. Provisional data from the National Vital Statistics System
13
1566 M. J. Bailey et al.
show that US birth rates from the first half of 2021 were down by 1.9 percent rela-
tive to the first half of 2020—a figure only slightly larger than our estimated 1.1
percent for the Title X population and much smaller than anticipated.14 Although
outside the scope of our data, a reasonable expectation is that changes in access to
contraception and abortion may moderate the anticipated 2021 COVID-19 baby bust
for the broader population as well.
Beyond understanding the implications of the COVID-19 economic downturn
for childbearing, these findings have implications for state and national reproduc-
tive health policy. Over the last decade, many state legislatures have limited funding
for reproductive health care through Medicaid and Title X as well as through state
programs. At the national level, the Trump administration revised 2019 guidelines
for the Title X program, which resulted in more than 1000 health centers in 34 states
withdrawing from Title X—sites that had served more than 1.5 million patients
in the year before the rule took effect (National Family Planning & Reproductive
Health Association, 2020). In the absence of new funding or providers, we expect
that reductions in funding for Title X reproductive health services should have simi-
lar implications for the pregnancies and birthrates of low-income women.
Supplementary Information The online version contains supplementary material available at https://doi.
org/10.1007/s11113-022-09703-9.
Acknowledgements The M-CARE study was supported through grants awarded by Arnold Ventures
(AV) and the NICHD (R01HD100438). Bart’s research on this project was supported by the Eunice
Kennedy Shriver National Institute of Child Health & Human Development of the National Institutes of
Health under Award Number T32HD007339. The content is solely the responsibility of the authors and
does not necessarily represent the official views of the National Institutes of Health or of Planned Parent-
hood Federation of America, Inc. We gratefully acknowledge the use of the services and facilities of
the Population Studies Center at the University of Michigan (P2CHD041028) and the California Center
for Population Research at UCLA (P2CHD041022). We are grateful for excellent research support by
Suni Jo Roberts and Mari Hashimoto. We also thank Planned Parenthood of Michigan (PPMI), includ-
ing Melissa Fuller and Martha Nokken, for helping us understand PPMI processes and facilitating access
to electronic medical records; and NORC, including Sheri Hamilton, Karen Veldman, Lauren Seward,
Chet Bowie, and Lisa Blumerman, for their expert management of the field interviewers, survey, and the
recruitment process.
References
Adams-Prassl, A., Boneva, T., Golin, M., & Rauh, C. (2020). Inequality in the Impact of the Corona-
virus Shock: Evidence from Real Time Surveys. Retrieved from www.iza.org
Alon, T., Doepke, M., Olmstead-Rumsey, J., & Tertilt, M. (2020a). The impact of COVID-19 on gen-
der equality. Retrieved from http://www.nber.org/papers/w26947
Alon, T., Doepke, M., Olmstead-Rumsey, J., Tertilt, M., Thank, W., Bardoczy, B., Zymek, R. (2020b).
This time it’s different: The role of women’s employment in a pandemic recession. Retrieved
from http://www.nber.org/papers/w27660
14
Authors’ calculations based on provisional data from NVSS on monthly live births, available at
https://www.cdc.gov/nchs/nvss/vsrr/provisional-tables.htm.
13
The Missing Baby Bust: The Consequences of the COVID‑19 Pandemic… 1567
Autor, D., Dorn, D., & Hanson, G. (1990). When work disappears: manufacturing decline and the
falling marriage market value of young men. AER Insights, 2019(2), 161–178. https://doi.org/10.
1257/aeri.20180010
Bailey, M. J. (2010). “Momma’s Got the Pill”: how anthony comstock and griswold v. connecticut
shaped US childbearing. American Economic Review, 100(1), 98–129. https://doi.org/10.1257/
aer
Bailey, M. J. (2012). Reexamining the impact of family planning programs on us fertility: Evidence from
the war on poverty and the early years of title X. American Economic Journal: Applied Economics,
4(2), 62–97. https://doi.org/10.1257/app.4.2.62
Bailey, M. J., & Lindo, J. M. (2018). Access and use of contraception and its effects on women’s out-
comes in the United States. In S. L. Averett, L. M. Argys, & S. D. Hoffman (Eds.), Oxford Hand-
book of Women and the Economy. https://doir.org/https://doi.org/10.1093/oxfordhb/9780190628
963.013.19.
Becker, N. (2018). The impact of insurance coverage on utilization of prescription contraceptives: Evi-
dence from the affordable care act. Journal of Policy Analysis and Management, 37(3), 571–601.
https://doi.org/10.1002/pam.22064
Bernstein, J., & Jones, J. (2020). The Impact of the COVID19 recession on the jobs and incomes of per-
sons of color. Retrieved from https://www.cbpp.org/research/full-employment/the-impact-of-the-
covid19-recession-on-the-jobs-and-incomes-of-persons-of
Bick, A., & Blandin, A. (2020). Real-time labor market estimates during the 2020 coronavirus outbreak.
SSRN Electronic Journal. https://doi.org/10.2139/ssrn.3692425
Black, D. A., Kolesnikova, N., Sanders, S. G., & Taylor, L. J. (2013). Are children “Normal”? The Review
of Economics and Statistics, 95(1), 21–33.
Buckles, K., Guldi, M. E., & Schmidt, L. (2019). Fertility Trends in the United States, 1980–2017: The
Role of Unintended Births. Retrieved from National Bureau of Economic Research website: http://
www.nber.org/papers/w25521
Buckles, K., Hungerman, D., & Lugauer, S. (2021). Is Fertility a leading economic indicator? The Eco-
nomic Journal, 131(634), 541–565. https://doi.org/10.1093/ej/ueaa068
Carlin, C. S., Fertig, A. R., & Dowd, B. E. (2016). Affordable care act’s mandate eliminating contra-
ceptive cost sharing influenced choices of women with employer coverage. Health Affairs, 35(9),
1608–1615. https://doi.org/10.1377/hlthaff.2015.1457
Cherlin, A., Cumberworth, E., Morgan, S. P., & Wimer, C. (2013). The Effects of the Great Recession on
Family Structure and Fertility. The Annals of the American Academy of Political and Social Science,
650, 214–231. https://doi.org/10.1177/0002716213500643
Chetty, R., Friedman, J. N., Hendren, N., Stepner, M., & Opportunity Insights Team. (2020). The eco-
nomic impacts of COVID-19: evidence from a new public database built using private sector data.
Retrieved from https://opportunityinsights.org/wp-content/uploads/2020/05/tracker_paper.pdf
Collins, C., Landivar, L. C., Ruppanner, L., & Scarborough, W. J. (2020). COVID‐19 and the gender gap
in work hours. Gender, Work & Organization, gwao.12506. https://doi.org/10.1111/gwao.12506
Coombs, L. C. (1974). The measurement of family size preferences and subsequent fertility. Demogra-
phy, 11(4), 587–611. https://doi.org/10.2307/2060472
Currie, J., & Schwandt, H. (2014). Short- And long-term effects of unemployment on fertility. Proceed-
ings of the National Academy of Sciences of the United States of America, 111(41), 14734–14739.
https://doi.org/10.1073/pnas.1408975111
Dalton, V. K., Moniz, M. H., Bailey, M. J., Admon, L. K., Kolenic, G. E., Tilea, A., & Fendrick, A. M.
(2020). Trends in birth rates after elimination of cost sharing for contraception by the patient protec-
tion and affordable care act. JAMA Network Open, 3(11), e2024398. https://doi.org/10.1001/jaman
etworkopen.2020.24398
Dwyer, K., Minton, S., Kwon, D., & Weisner, K. (2020). Key cross-state variations in CCDF policies as
of October 1, 2019. Retrieved from https://ccdf.urban.org.
Easterlin, R. A., Pollak, R., & Wachter, M. L. (1980). Toward a more general economic model of fertility
determination: endogenous preferences and natural fertility. Retrieved from https://www.nber.org/
chapters/c9664.pdf
Efron, B., & Tibshirani, R. J. (1993). An Introduction to the Bootstrap. Chapman and Hall.
Fowler, C., Gabel, J., & Lasater, B. (2021). Title X family planning annual report: 2020 national sum-
mary. Retrieved from https://opa.hhs.gov/research-evaluation/title-x-services-research/family-plann
ing-annual-report/family-planning-0
13
1568 M. J. Bailey et al.
Fowler, C, Gable, J., Lasater, B., & Asman, K. (2020). Title X family planning annual report: 2019
national summary. Retrieved from https://opa.hhs.gov/research-evaluation/title-x-services-research/
family-planning-annual-report/family-planning-0
Fowler, Christina, Gable, J., Wang, J., Lasater, B., & Wilson, E. (2019). Family planning annual report:
2018 national summary. Retrieved from https://opa.hhs.gov/research-evaluation/title-x-services-
research/family-planning-annual-report/family-planning-0
Frost, J. J., Frohwirth, L. F., Blades, N., Zolna, M. R., Douglas-Hall, A., & Bearak, J. (2017). Publicly
Funded Contraceptive Services at U.S. Clinics, 2015. New York.
Gemmill, A., & Hartnett, C. S. (2020). Demographic drivers of the post-recessionary fertility decline and
the future of U.S. fertility. https://doi.org/10.31235/OSF.IO/2U78X
Hainmueller, J. (2011). Entropy balancing for causal effects: A multivariate reweighting method to pro-
duce balanced samples in observational studies. Political Analysis, 16, 25–46. https://doi.org/10.
1093/pan/mpr025
Hasstedt, K. (2017). Understanding planned parenthood’s critical role in the nation’s family planning
safety net. Guttmacher Policy Review, 20. Retrieved from https://www.guttmacher.org/sites/default/
files/article_files/gpr2001216.pdf
Heisel, E., Kolenic, G. E., Moniz, M. M., Kobernik, E. K., Minadeo, L., Kamdar, N. S., & Dalton, V. K.
(2018). Intrauterine device insertion before and after mandated health care coverage: The impor-
tance of baseline costs. Obstetrics and Gynecology, 131(5), 843–849. https://doi.org/10.1097/AOG.
0000000000002567
Kaiser Family Foundation. (2020). Health insurance coverage of females 19–64. Retrieved December 4,
2020, from https://www.kff.org/other/state-indicator/health-insurance-coverage-of-nonelderly-adult-
females/?activeTab=g raph¤tTimeframe=0&startTimeframe=11&selectedRows=%7B%
22wrapups%22:%7B%22united-states%22:%7B%7D%7D%7D&sortModel=%7B%22colId%22:%
22Location%22,%22sort%22:%22asc%22%7D
Kearney, M., & Levine, P. (2020a). Half a million fewer children? The coming COVID baby bust. Retrieved
from https://www.brookings.edu/research/half-a-million-fewer-children-the-coming-covid-baby-
bust/
Kearney, M., & Levine, P. (2020b). The coming COVID-19 baby bust: Update. Retrieved from https://
www.brookings.edu/blog/up-front/2020b/12/17/the-coming-covid-19-baby-bust-update/
Kearney, M. S., & Levine, P. B. (2009). Subsidized contraception, fertility, and sexual behavior. The
Review of Economics and Statistics, 91(1), 137–151.
Lindberg, L. D., VandeVusse, A., Mueller, J., & Kirstein, M. (2020a). Early impacts of the COVID-
19 Pandemic: Findings from the 2020a guttmacher survey of reproductive health experiences.
Retrieved from https://www.guttmacher.org/report/early-impacts-covid-19-pandemic-findings-
2020a-guttmacher-survey-reproductive-health
Lindberg, L., Kost, K., Maddow-Zimet, I., Desai, S., & Zolna, M. (2020b). Abortion reporting in the
United States: an assessment of three national fertility surveys. Demography, 57(3), 899–925.
https://doi.org/10.1007/S13524-020-00886-4
Lindo, J. M. (2010). Are children really inferior goods? Journal of Human Resources, 45(2), 301–327.
https://doi.org/10.3368/jhr.45.2.301
Lindo, J. M., & Packham, A. (2017). How much can expanding access to long-acting reversible con-
traceptives reduce teen birth rates? American Economic Journal: Economic Policy, 9(3), 348–376.
https://doi.org/10.1257/pol.20160039
Linnakaari, R., Helle, N., Mentula, M., Bloigu, A., Gissler, M., Heikinheimo, O., & Niinimäki, M.
(2019). Trends in the incidence, rate and treatment of miscarriage—nationwide register-study in
Finland, 1998–2016. Human Reproduction, 34(11), 2120–2128. https://doi.org/10.1093/humrep/
dez211
Lu, Y., & Slusky, D. J. G. (2019). The impact of women’s health clinic closures on fertility. American
Journal of Health Economics, 5(3), 334–359. https://doi.org/10.1162/ajhe_a_00123
National Family Planning & Reproductive Health Association. (2020). Key facts about title X. Retrieved
December 3, 2020, from https://www.nationalfamilyplanning.org/title-x_title-x-key-facts
NYT Editorial Board. (2019). Birth control gets caught up in the abortion wars. The New York Times.
Retrieved from https://www.nytimes.com/2019/02/26/opinion/title-x-rule.html
Planned Parenthood Federation of America. (2019). By the Numbers. Retrieved from https://www.plann
edparenthood.org/uploads/f iler_public/3a/df/3adfdb04-b7f6-420f-a8e7-7284e39f13cc/190124-
bythenumbers-v01.pdf
13
The Missing Baby Bust: The Consequences of the COVID‑19 Pandemic… 1569
Planned Parenthood Federation of America. (2020). Who we are: Black communities. Retrieved from
https://www.plannedparenthood.org/uploads/filer_public/04/da/04da4f8f-7941-4086-a9d9-1f8de
e74bddb/200225-who-we-are-black-communities-d01.pdf
Planned Parenthood of Michigan. (2021). Health Center Info 2020 Statewide: 16 Health Centers
Schaller, J. (2016). Booms, Busts, and Fertility. Journal of Human Resources, 51(1), 1–29. https://doi.
org/10.3368/JHR.51.1.1
Schaller, J., Fishback, P., & Marquardt, K. (2020). Local economic conditions and fertility from the great
depression through the great recession. AEA Papers and Proceedings, 110, 236–240. https://doi.org/
10.1257/pandp.20201109
Schneider, D., & Hastings, O. P. (2015). Socioeconomic variation in the effect of economic conditions on
marriage and nonmarital fertility in the united states: evidence from the great recession. Demogra-
phy, 52(6), 1893–1915.
Shorto, R. (2006, May 7). Contra-contraception. The New York Times Magazine. Retrieved from https://
www.nytimes.com/2006/05/07/magazine/07contraception.html
Trussell, J. (2011). Contraceptive failure in the United States. Contraception, 83, 397–404. https://doi.
org/10.1016/j.contraception.2011.01.021
US Census Bureau. (2019). State and Local General Expenditures, Per Capita. Retrieved October 13,
2021, from Annual Survey of State and Local Government Finances, Compiled by the Urban-
Brookings Tax Policy Center website: https://www.taxpolicycenter.org/statistics/state-and-local-
general-expenditures-capita
Venator, J., & Fletcher, J. (2021). Undue burden beyond texas: An analysis of abortion clinic closures,
births, and abortions in wisconsin. Journal of Policy Analysis and Management, 40(3), 774–813.
https://doi.org/10.1002/PAM.22263
Ventura, S. J., Curtin, S. C., & Abma, J. C. (2012). Estimated pregnancy rates and rates of pregnancy
outcomes for the United States, 1990–2008. In National Vital Statistics Reports (Vol. 60). Retrieved
from http://www.cdc.gov/nchs/nvss/bridged_race.htm.
Willis, R. J. (1973). A new approach to the economic theory of fertility behavior. Journal of Political
Economy, 81(2), S14–S64. Retrieved from https://www.jstor.org/stable/1840411
Zolna, M., & Lindberg, L. (2012). Unintended Pregnancy: Incidence and Outcomes Among Young Adult
Unmarried Women in the United States, 2001 and 2008. Retrieved from https://www.guttmacher.
org/sites/default/files/report_pdf/unintended-pregnancy-us-2001-2008.pdf
Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published
maps and institutional affiliations.
13
File and source
- File
- the-missing-baby-bust-the-consequences-of-the-covid-19-pandemic-for-contraceptive-use-pregnancy-and-childbirth-among-low.pdf
- Size
- 760,535 bytes
- SHA-256
- 80527cf188a13754ba91eb1df3f5e9f76a487de7e54ed4232937e50df483e2b8
- Original
- doi.org