The following comments were submitted by Richard Besser, MD, Robert Wood Johnson Foundation (RWJF) President and CEO, in response to the Office of Management and Budget's (OMB) request for comments on the Consumer Inflation Measures produced by Federal statistical agencies.

RWJF is the nation’s largest philanthropy dedicated to improving health and well-being in the United States. Since 1972, we have worked with public and private sector partners to advance the science of disease prevention and health promotion, train the next generation of health leaders, and support the development and implementation of programs and policies to foster better health across the country, including high-quality health care coverage for all. We are working alongside others to build a Culture of Health that provides everyone in America a fair and just opportunity to live the healthiest life possible.

A key to this vision is ensuring that every person has the necessary supports to meet their own and their families’ basic needs, such as food, shelter, health care, and education. For people living at or near poverty, employment income is essential in paying for these necessities; however, even full-time work at the low end of the wage scale leaves many with financial gaps. Other groups—such as workers with part-time or unstable employment, people with physical or mental disabilities, and the elderly—are even more vulnerable to material hardship. In these cases, government plays a crucial role in providing a safety net that fosters financial stability, supports work and caregiving, and promotes individual and community well-being. Robust research demonstrates that public benefit programs like Medicaid, Affordable Care Act (ACA) subsidies, the Supplemental Nutrition Assistance Program (SNAP), the Special Supplemental Nutrition Program for Women, Infants, and Children (WIC), and Head Start improve health, educational, and economic outcomes, particularly for children, both during childhood and throughout their lives.

Recommendation

It is in this context that we submit comments to OMB as it considers the strengths, weaknesses, and best practices for applying various measures of inflation in calculating the Official Poverty Measure (OPM). The OPM is used in setting poverty guidelines by which numerous federal agencies determine eligibility for critical safety net programs. Overall, we believe OMB should reject the use of alternative indices that would result in lower rates of inflation—such as the Chained Consumer Price Index for All Urban Consumers (C-CPI-U) and the Personal Consumption Expenditures Price Index (PC-EPI)—and that, in turn, would reduce poverty thresholds. First, lower estimates of inflation do not accurately reflect the changes in costs that people with low and moderate incomes actually experience, contradicting the purpose of the proposed reform. Second, lower poverty thresholds would result in fewer low-income individuals and families qualifying for critical public benefits. Notably, this comes at a time when numerous other regulatory reforms have decreased eligibility for and/or enrollment in safety net programs among people facing economic insecurity. Third, decreased access to these programs would likely result in poorer health and well-being and widening disparities by income and race. Fourth, changes to inflation adjustment overlook the more fundamental concern that current measures of poverty as a whole are outdated and inadequate, resulting in significant underestimates of people living in poverty and facing serious financial hardships.

Rationale

1.    Alternative indices that result in lower rates of inflation would not accurately reflect the changes in costs that people with low and moderate incomes actually experience.

The Census Bureau determines poverty status by using the OPM, which compares pre-tax income to a threshold set at three times the cost of a minimum food diet in 1963. This threshold is updated annually for inflation and adjusted for family size, composition, and age of householder. In 2019, the OPM poverty threshold for a family of four was $25,750. Currently, the Census Bureau utilizes the Consumer Price Index for All Urban Consumers (CPI-U) as its inflation measure, which captures the average change over time in the prices paid by urban consumers for a market basket of more than 200 goods and services, such as food, housing, utilities, and medical care. Since 2000, the average annual CPI-U has been 2.0 percent.¹

Independent estimates demonstrate that alternative measures of inflation, such as the C-CPI-U and the PC-EPI, would be lower than the CPI-U on an annual basis and result in a lowering of the poverty threshold over time. According to the Congressional Budget Office, use of the C-CPI-U would reduce the poverty threshold by approximately 0.2 percent annually and by 2.0 percent after ten years; use of the PC-EPI would reduce it by 3.4% after ten years.^{2, 3}

Theoretically, the C-CPI-U would make for a more accurate measure of inflation because it accounts for consumers changing what they purchase when relative prices are changing. For example, a household might shift from buying milk to juice if milk prices were rising more quickly than juice prices. However, the ability of low-income households to make such changes may be different from that of the general population because of fewer choices about where and when they shop.⁴ Those living in rural areas may have particularly constrained choices. In addition, a lower rate of inflation, regardless of the theoretical advantages of the underlying index, likely does not reflect the experience of low-income households. A recent study demonstrates that inflation rises faster for people with low incomes than the population as a whole: over the course of nine years, cumulative inflation was eight percentage points higher for low-income households.⁵ A number of other studies confirm this pattern.⁶ Looking specifically at housing costs provides additional clarity. Compared to renters with higher incomes, low-income renters are twice as likely to spend fifty percent or more of their income on rent, and the national median rent rose 20 percent faster than inflation between 1990 and 2016.^7,8 So, a substantial component of spending by low-income households is growing more quickly than the CPI-U; a lower inflation measure would only exacerbate this problem.   

The other proposed inflation indexes are, for varying reasons, inappropriate or unnecessary for adjusting the OPM. The PC-EPI—another index resulting in lower inflation—is calculated based on price surveys of businesses and includes the spending patterns of consumers and non-profit institutions serving consumers, so it is as not as well-suited as the CPI-U in capturing actual consumer spending. The Consumer Price Index for Urban Wage Earners and Clerical Workers (CPI-W) is based on the same sample as the CPI-U but is weighted differently to reflect the spending among a subset of urban consumers who work.⁹ Over the past 11 years, it has been very closely aligned with the CPI-U and would, therefore, offer no distinct advantage. The Consumer Price Index Research Series (CPI-U-RS) is a research version of the CPI-U, which adjusts historical inflation levels to reflect the current methodology for calculating the CPI-U.¹⁰ Because it is a backward-looking adjustment, it would not yield a different measure of inflation than the CPI-U in real time from one year to the next. The Consumer Price Index for Urban Elderly Consumers (CPI-E) is an experimental index focused on households headed by those at least 62 years of age.¹¹ It does not reflect the costs experienced by households headed by people of younger ages and is influenced substantially by changes in health care costs.

2.    Lower poverty thresholds would result in fewer low-income individuals and families qualifying for critical public benefits.

With a lower poverty threshold, fewer people would qualify over time for numerous means-tested public benefit programs, including, but not limited to: Medicaid, the Children’s Health Insurance Program (CHIP), Medicare Part D subsidies, the ACA cost-sharing subsidies and premium tax credits, SNAP, WIC, the National School Lunch Program, the School Breakfast Program, Head Start, the Low-Income Home Energy Assistance Program, and the Weatherization Assistance Program for Low-Income Persons.¹²

The Center on Budget and Policy Priorities estimates that using the C-CPI-U instead of the CPI-U would, after ten years, result in:

• More than 300,000 children losing comprehensive health coverage through Medicaid and CHIP;
• More than 250,000 adults losing Medicaid coverage expanded via the ACA;
• More than 250,000 seniors and people with disabilities losing eligibility for, or receiving less help from, the Medicare Part D Low-Income Subsidy Program;
• More than 150,000 consumers losing cost-sharing assistance when purchasing health insurance through the ACA marketplaces;
• Approximately 6 million marketplace consumers experiencing reductions in their premium tax credits; and
• Accordingly, billions of dollars in cuts to means-tested federal health care programs.¹³

The Urban Institute estimates that 579,000 individuals who received SNAP in 2016 would have been ineligible had the C-CPI-U been in place over the last 15 years. Children make up 242,000 of the recipients who would have been ineligible for SNAP. States such as Florida, Ohio, New York, and California would have experienced the greatest reductions.¹⁴

Notably, these potential reductions in safety net access come at a time when numerous other regulatory reforms—including new work requirements in Medicaid, tightening of work requirements in SNAP, and the public charge rule—have unjustifiably decreased eligibility for and/or enrollment in public benefit programs.

3.    Reduced access to public benefit program would likely result in poorer health and greater disparities.

Here, we focus on the health and health equity effects of five major public benefit programs that would be affected by changes to the poverty thresholds—Medicaid, ACA subsidies, SNAP, WIC, and Head Start.

Medicaid is an essential source of health insurance coverage for low-income children, adults, older adults, and people with disabilities. It increases access to primary care and behavioral health services, is associated with reduced all-cause mortality, and makes it easier for adult beneficiaries to work.¹⁵ Among children, it reduces racial disparities in birth outcomes, substantially lowers income-based disparities in access to care, leads to better health in adulthood, and protects families from financial insecurity due to medical debt and bankruptcy.^16,17,18,19 Loss of Medicaid is associated with decreases in health care access and health status and increases in hospitalizations among high-risk patients.^20,21

ACA premium subsidies accounted for approximately 25 percent of the coverage gains achieved through the ACA.²² Since they took effect in 2014, the uninsured rate for non-elderly adults with incomes between 138 and 400 percent of the poverty line dropped from 19.2 percent (2013) to 12.5 percent (2017).²³ In 2014, subsidies reduced marketplace premiums by an average of 76 percent for low- and moderate-income enrollees.²⁴

Serving three in ten young children each, SNAP and WIC reduce hunger and poverty and improve birth outcomes.²⁵ The effects of WIC on dietary quality are demonstrably positive, and access to SNAP in childhood leads to lower rates of metabolic syndrome in adulthood and, among females, higher economic self-sufficiency.²⁶ A recent study demonstrated that families that lost SNAP or experienced benefit reductions were significantly more likely to experience food insecurity, and reduced benefits were associated with greater odds of fair or poor caregiver and child health.²⁷

Head Start provides approximately 40 percent of three- and four-year-olds with early education, nutrition, and health services along with parental supports. A robust body of research demonstrates that Head Start results in greater access to preventive care, improved nutrition and immunization rates, and fewer hospitalizations for accidents and injuries. Additionally, Head Start improves developmental and educational outcomes for young children, particularly children of color, and enhances social and emotional health in adolescence.^28,29

4.    Current measures of poverty as a whole are outdated and inadequate.

Changes to inflation adjustment overlook the more fundamental concern that current measures of poverty significantly underestimate the number of people facing serious financial hardships and struggling to meet their basic needs. Outside of annual inflationary adjustments, the federal government has not substantially modified the poverty threshold since the 1960s. The basic formula consists of multiplying the costs of a minimum food diet by a factor of three, which is based on decades-old research demonstrating families spend approximately one-third of their income on food.³⁰ Food now accounts for a smaller portion of family expenditures, but modern childcare, housing, and health care spending comprise much larger portions, none of which is directly accounted for in current poverty measures.^31,32 Moreover, the narrow definition of “family” does not reflect present-day family structures and cohabiting patterns.³³ Lastly, reliance on a national standard neglects the fact that costs of living vary greatly across the country.^34,35

Based on guidance from a National Academy of Science panel, federal analysts developed a Supplemental Poverty Measure (SPM), which captures contemporary spending patterns for food, shelter, clothing, and utilities; broadens the definition of family; and accounts for government cash and non-cash supports, such as SNAP and the Earned Income Tax Credit.³⁶ The SPM’s poverty threshold is higher than that of the OPM for most household types.³⁷

Another way to consider the inadequacy of the existing poverty threshold is by comparing it to a family’s basic needs budget. In Washington, D.C., in 2017, a two-parent family with one parent working full time and children three- and six-years-of-age needed $57,050 to cover their basic needs.³⁸ This is more than double the federal poverty level for a family of four in 2017—$24,600. Such a mismatch holds even in lower cost areas of the country: the West Virginia Governor’s Workforce Investment Division found that poverty-level income covers only two-thirds of the basic needs for a single-parent family with two children residing in the state.³⁹

Recent research also demonstrates that people living just above the poverty line (100 percent to 199 percent of the federal poverty line) face significant material hardship. In 2017, 38 percent reported they would have difficulty in coming up with $400 for an unexpected expense; 20 percent missed a payment for a credit card or a non-mortgage loan; and 24 percent had been contacted by a debt collector.⁴⁰ More than 40 percent were food insecure and 30 percent had trouble paying family medical bills.⁴¹ With changes in the poverty threshold, some of these people could lose coverage for Medicaid (138 percent of the federal poverty line in Medicaid expansion states) or SNAP (ranging from 130 percent to 200 percent of the federal poverty line depending on the state).

Conclusion

We believe that OMB should reject the use of alternative indices that would result in lower rates of inflation, including the C-CPI-U and the PC-EPI. Not only would these alternatives less accurately account for the rising costs faced by people with low incomes, they would also unnecessarily result in fewer people qualifying over time for critical public benefit programs, putting their well-being at risk and exacerbating health disparities based on income and race/ethnicity. OMB should instead focus its attention on updating and making more accurate the overall measures of poverty used to determine eligibility for programs such as Medicaid, ACA subsidies, SNAP, WIC, and Head Start. We would welcome the opportunity to inform such a process with an eye toward its implications for health and well-being in this country.