Early childhood is a crucial period in a child’s development. It’s a time of rapid brain development, with early experiences having long-term effects on a person’s success, from risk of obesity and other health problems, to likelihood of going to college and finding gainful employment. Adequate nutrition is a vital component for development, and for infants, the American Academy of Pediatrics (AAP) recommends exclusive breastfeeding for the first six months.

While the Department of Health and Human Services put forth recommendations in 2010 intended to increase the prevalence and duration of breastfeeding among U.S. mothers, there have been only modest improvements. One major barrier to breastfeeding is the lack of paid family leave in the U.S., which often forces women to go back to work (often at jobs where pumping is difficult, if not impossible) soon after birth.

MIP network member Jane Waldfogel and her co-authors Jessica Pac, Ann P. Bartel, and Christopher Ruhm wanted to know whether access to paid family leave (PFL) would have a positive effect on breastfeeding for infants. The authors study this in their recent HCEO working paper, using California’s enactment of a 2004 PFL policy as a natural experiment.

California’s policy ensured eligible mothers up to six weeks of leave at a 55 percent wage replacement rate. “California was the first state in the U.S. to roll out a paid family leave policy that extended the unpaid leave that was already available [through the Family Medical Leave Act (FMLA), which allows up to 12 weeks of job-protected unpaid leave],” Pac says.

To calculate the effects of PFL, the authors used the restricted-use 2003-2014 National Immunization Survey, a nationally-representative data set containing detailed information on breastfeeding behaviors. The final sample consisted of 271,309 child-mother pairs for births occurring between 2000 and 2012, which makes this study “the first U.S. evidence on the relationship between PFL and breastfeeding using a large, representative sample combined with appropriate econometric techniques.”

The authors looked at questions pertaining to breastfeeding initiation and duration. They use difference-in-difference models using synthetic control methods to compare the pre- versus post-law differences in outcomes of mother-child pairs in California to those outside the state. To apply the synthetic control method, states were “weighted based on their statistical similarity to California in terms of pre-treatment trends.” The authors also vary the pre-law outcome variables and covariates, along with estimating other models, to test the strength of their results.

“Our results suggest that PFL significantly increases overall breastfeeding duration by nearly 18 days (from a base of 221 days),” the authors write. It also increases the likelihood that children will be breastfed for at least six months by five percentage points, though it had little effect on the probability of initiating breastfeeding.

“But we find much larger effects among women who are disadvantaged,” Pac says.

PFL was shown to increase breastfeeding for families who had recently experienced an interruption in phone service. Similar effects were found for families with lower levels of education, and for those who reported receiving assistance from the Women, Infants, and Children program (WIC) in the prior year. Black/non-Hispanic respondents were “nearly 7.5 percentage points more likely to initiate breastfeeding under increased PFL access.” The authors note that “mothers with less than a high school education were predicted to breastfeed 63 additional days as the result of PFL, compared to around 35 days for high school graduates or those with some college and no change for college graduates.” The policy was also associated with a 14 percentage point increase in breastfeeding for at least six months for respondents living below fifty percent of the federal poverty line.

“This finding of larger effects among disadvantaged women is especially important, because parental investments are often prioritized among high-earning families,” Pac says. “Yet, they’re much harder to obtain among more disadvantaged families. So the fact that there are distributional effects of this policy is very encouraging and, I think, merits consideration from other states examining ways to benefit lower-earning children and families.”

It is important to note that this study does have some limitations. For one, the researchers were unable to observe women’s labor market participation and actual eligibility for PFL. Thus, future research that could “more fully integrate pre-birth employment and eligibility” is needed to “reduce any remaining residual bias.” In addition, mothers in California were already eligible for a temporary disability insurance policy, which made available “up to four weeks of paid pre-birth leave and six weeks of paid post-birth leave (eight weeks for children born by Cesarean section).” California already had a high pre-PFL rate of breastfeeding, 85 percent, possibly due to this temporary disability insurance policy, which may partially explain why PFL had little effect on breastfeeding initiation rates.

Despite these limitations, California’s PFL policy seems to be successful in increasing breastfeeding rates, a longstanding goal in the U.S., and for clear reasons. As the authors note, breastfeeding “has been linked to strengthened immunity, reduced likelihood of post-neonatal and SIDS mortality, and decreases in hospitalizations and deaths from infectious diseases, diarrhea, and respiratory infections.” Breastfed children are also at lower risk of developing obesity, type II diabetes, and asthma. For mothers, breastfeeding is associated “with decreases in maternal breast and ovarian cancer, type II diabetes, and postpartum depression.” Indeed, the paper states that, “California’s PFL program was associated with reductions in obesity, ADHD, and hearing related problems among elementary school children.”

While AAP recommends continued “breastfeeding until the baby’s first birthday or longer,” the government’s lack of systemic support for new parents makes that difficult, particularly for disadvantaged families. As the authors point out, less than 60 percent of employees are estimated to even be eligible for unpaid leave under FMLA. The U.S. “is the only developed country without a statutory national PFL entitlement.” The evidence from this paper could be a guiding light for policy, as it suggests that “extending PFL to families in states without current mandates may increase breastfeeding durations, possibly leading to health improvements for children and mothers in the longer-term.”

The effects on disadvantaged families are even more promising, showing that extending PFL “may also reduce disparities in breastfeeding and associated outcomes.” The authors note that“[t]his finding makes sense, given that employer paid leave is tilted toward more advantaged workers and given that low-income mothers may be less able to afford to use the unpaid leave offered under the FMLA.”

“That we find such large effects among disadvantaged mothers is very encouraging, considering the fact that these benefits are low relative to other countries and to some other states,” Pac says. “Our findings imply that women do in fact take paid family leave, and that when they do, their children benefit.”