Skip to main content
. 2023 Dec 5;23:2422. doi: 10.1186/s12889-023-17166-6

Table 2.

Effects of a Ten-Percentage-Point Increase in Refundable State EITC (as % of Federal Credit) on Birth Outcomes Born to Single Low-Educated Women Aged 18–46 Years, Natality Files 1989–2018, All County Sample by Birth Order

Outcomes EITC Effect
(95% CI)
Outcome Mean
first order birth order
Birth weight (grams) 4.262 3196.27
[-0.374,8.89]
Low birth weight rate (%) -0.002* 0.096
[-0.003, -0.00009]
Gestational weeks (week) 0.042* 38.90
[0.007,0.078]
Preterm birth rate (%) -0.001 0.123
[-0.004,0.001]
Fetal growth rate (grams/week) 0.024 81.92
[-0.05,0.10]
second birth order
Birth weight (grams) 7.823** 3226.15
[2.87,12.77]
Low birth weight rate (%) -0.003** 0.089
[-0.005, -0.001]
Gestational weeks (week) 0.045** 38.67
[0.014,0.077]
Preterm birth rate (%) -0.002 0.135
[-0.005,0.001]
Fetal growth rate (grams/week) 0.114* 83.25
[0.02,0.21]
third or higher birth order
Birth weight (grams) 12.411** 3200.94
[5.48,19.34]
Low birth weight rate (%) -0.005** 0.108
[-0.008, -0.002]
Gestational weeks (week) 0.061** 38.40
[0.024,0.099]
Preterm birth rate (%) -0.005* 0.160
[-0.009, -0.001]
Fetal growth rate (grams/week) 0.201** 83.09
[0.067,0.33]

Note: Each cell represents the effect of a 10%-point increase in refundable state EITCs (relative to federal credits) in tax year t-m on an outcome in birth year t. The EITC measure, each outcome and demographic characteristics are the average value for each county per year (one observation for each county per year). The model includes one EITC variable, the refundable percentage of federal credit (states with no EITC and states with non-refundable ETIC have 0 on this variable as the control group). The demographic controls include indicators for maternal age, child sex, race/ethnicity, education, and county and year fixed effects. State level controls include average minimum wage (2018$) from the past 12 months, average state cigarette tax from the past 12 months, and average maximum Medicaid income eligibility for pregnant women as % of FPL from the past 12 months. The regressions are run separately for each birth order. The sample size is 81,589 for first birth order, 79,536 for second birth order, and 79,221 for third or higher birth order. The regressions are weighted using the count of outcomes at the county level. Standard errors (SE) are clustered at state level and shown in paratheses. *p < 0.05, **p < 0.01