The Change in Poverty from 1995 to 2016 Among Single Parent Families
- Date Posted:
- Is Database:
- Database
Single parent family poverty declined by 62% btw 1995 and 2016, according to the Comprehensive Income Dataset (CID). @KevinCorinth
Core finding, "...This paper is the first to examine changes in poverty over time using a comprehensive set of linked survey and administrative data, implementing recommendations of the Interagency Technical Working Group on Evaluating Alternative Measures of Poverty. Using the Comprehensive Income Dataset (CID), we correct for measurement error in survey-reported incomes, focusing on single parent families from 1995 to 2016. Our preferred estimates indicate that single parent family poverty declined by 62% over time, while it fell by only 45% using survey data alone. Moreover, survey-reported deep poverty among single parent families increased over time, while it fell using the CID...."
Their Poverty Thresholds, "...To construct absolute poverty thresholds in 1995 and 2016, we use as a starting point the official threshold in each year for a family of two adults and two children multiplied by the SPM 3-parameter equivalence scale to account for family size and composition. Note that the official thresholds are updated for inflation using the CPI-U, which has a widely documented upward bias (see Moulton 2018 for a recent review and additional citations). We therefore also construct alternative poverty thresholds in 2016, which use an adjusted inflation measure that subtracts 0.8
percentage points off the annual growth in the CPI-U-RS each year starting from the 1995 baseline (Meyer and Sullivan 2012). We also show poverty estimates that use the Personal Consumption Expenditure (PCE) price index to adjust for inflation, which has been used by several prior studies to update thresholds over time and corrects for some but not all of the bias in the CPI-U. While the level of the baseline thresholds is fundamentally arbitrary,we use the 1995 official thresholds because they allow us to show how the poverty rate in 1995 falls after using a broader income definition and correcting for survey misreporting. We focus on poverty (income below 100% of the threshold) and deep poverty (income below 50% of the threshold). We also focus on single parent families rather than single mother families, although - as Appendix Table A2 shows - the percentage of single parent families headed
by a female remains largely consistent over time (85% in 1995 and 84% in 2016). We define single parent families as sharing units with a single adult (aged 18 or over) and at least one child (under age 18), and we estimate that there were 7.2 million single parent families in 1995 and 6.4 million single parent families in 2016..”
The Evidence
“…Table 2 shows poverty rates in 1995 and 2016 Focusing first on 1995 survey levels, we find that the poverty rate for single parent families using survey-reported pre-tax money income is 47.4% (Column 1). While this rate is calculated using the reweighted sample of individuals in PIKed and non-whole imputed families, it is virtually identical to the rate (47.0%) calculated using the full survey sample and original survey weights (Appendix Table A3). Incorporating tax liabilities and credits in the resource measure leads to a reduction of 3.8 percentage points (8%), while the combination of taxes and non-medical in-kind transfers reduces survey poverty by 17.1 percentage points (36%) to a rate of 30.3%. Switching to the CID further reduces the poverty rate (relative to the survey baseline) for each income concept, as it falls by 5% for pre-tax money income, 2% for post-tax money income, and 14% for post-tax/in-kind transfer income - yielding a final rate of 26.1%. In 2016, using thresholds adjusted by the CPI-U, the poverty rate for single parent families is 39.9% using survey-reported pre-tax money income (Column 2). The poverty rate falls by 7.6 percentage points (19%) after incorporating taxes, and by 18.3 percentage points (46%) after incorporating both taxes and in-kind transfers. Again, switching to the CID causes the poverty rate to fall for each income concept, with post-tax/in-kind transfer poverty falling the most by 28% to a final rate of 15.5%. Not only is the percent decline in poverty due to changes in the income concept larger in 2016 (reflecting increased spending on tax credits and in-kind transfers over time), but the percent decline due to employing the CID is also larger in 2016 than in 1995. The latter result is consistent with increased underreporting of survey incomes, indicating that survey estimates of changes in poverty - regardless of income concept - will be biased upward. Changes in poverty from 1995 to 2016 are explicitly shown in Column 5, 6, and 7 of Table 2. When updating thresholds based on the CPI-U, the percent decline in poverty is greater using the CID than using the survey alone. Poverty based on pre-tax money income fell by 21% using the CID, compared to 16% using the survey. Poverty based on post-tax money income fell by 38% using the CID, compared to 26% using the survey. And poverty based on post-tax/in-kind transfer income fell by 41% using the CID, compared to 29% using the survey. The poverty reductions are even larger after correcting for the upward bias in the CPI-U by updating thresholds using either the PCE price index or the adjusted CPI-U-RS. Using the PCE price index, CID poverty fell by 28% using pre-tax money income, 44% using post-tax money income, and 49% using post-tax/in-kind transfer income. Using the adjusted CPI-U-RS (our preferred inflation adjustment), CID poverty fell by 37% using pre-tax money income, 51% using post-tax money income, and 56% using post-tax/in-kind transfer income. Further accounting for administrative SNAP in 2016 leads to a 62% decrease in poverty over time and a 2016 poverty rate among single parent families of 9.8%.This final estimate of the change in poverty is our preferred estimate because it reflects the broadest set of income sources available for consumption, corrects for survey misreporting using the CID, and corrects for bias in the inflation measure used to adjust official poverty thresholds. Given the large increase in aggregate SNAP underreporting and its bias toward poverty rate increases, we prefer estimates that include administrative SNAP data for 2016 even though we do not have administrative data available for 1995. Table 2 also reports results for deep poverty. Based on pre-tax money income, deep poverty between 1995 and 2016 fell by 8 to 21% (depending on the inflation measure used to update thresholds) when relying only on the survey and by 31 to 46% when relying on the CID. After incorporating both taxes and in-kind transfers into the resource measure, survey-reported deep poverty rose between 1995 and 2016 by 9 to 23% (depending on the inflation measure used to update thresholds). In contrast, CID deep poverty fell either slightly (when using the CPI-U) or by more than 20% when using the adjusted CPI-U-RS…”
Kevin Corinth, Bruce D. Meyer, and Derek Wu, "The Change in Poverty from 1995 to 2016 Among Single Parent Families,"Becker Friedman Institute, March 2022, https://bfi.uchicago.edu/wp-content/uploads/2022/03/BFI_WP_2022-45.pdf



