This blog is the second in a four-part series sharing key takeaways from Making Tax Analyses More Effective and Inclusive: A Guide For Analysts. 

Tax policies affect different groups of people differently. Sometimes, this is by design. For example, a household with higher income may receive a smaller child tax credit than a household with lower income, because the policy aims to provide relatively greater support to those with limited means.  

In other instances, different treatment is unintended, or a consequence of other design choices. For example, TPC researchers Janet Holtzblatt, Swati Joshi, Nora Cahill, and Bill Gale note that marriage penalties, bonuses, or both are inherent to any progressive, family-based income tax system. Because marriage penalties are most likely when spouses have similar earnings, their research showed higher penalties among Black couples, whose earnings are more likely to be similar than those of White couples, as long suggested by Dorothy Brown and other legal scholars.  

In a new guide, we share how tax analyses can demonstrate how policies affect different populations, or subgroups, in accurate and compelling ways. For analysts seeking to provide credible, policy-relevant insights to policymakers and advocates, disaggregating data to include subgroup analyses, now increasingly feasible given new tools and data, can shed light on whether tax policies and practices are having their intended effects, versus producing unintended ones. 

Most traditional tax analyses disaggregate tax outcomes by income percentile or level. Some, such as TPC’s own distributional tables, offer additional breakdowns by tax filing status (single, head of household, married filing jointly) as well as for taxpayers with children and elderly taxpayers. Other characteristics of subgroups could meaningfully affect how their members experience tax policy. For example: 

• Race and ethnicity: Longstanding racial and ethnic disparities in income and wealth are driven by historical and contemporary structural inequalities, making race and ethnicity relevant to consider when examining many tax policies. Tax preferences tied to retirement savings, for example, tend to deliver larger benefits to households who already have more income and wealth. These provisions may disproportionately advantage White households relative to Black or Latine households, even if the tax policies are, on their surface, race-neutral.
• Disability. Many tax provisions assume steady earnings and predictable work patterns. But people with disabilities are more likely to have intermittent employment, lower average earnings, and higher out-of-pocket medical or caregiving expenses. An expansion of the earned income tax credit (EITC) may, on average, increase after-tax income for low-wage workers. Yet if much of the benefits “phase in” with income (someone with $10,000 in income gets a larger credit than someone with an income of $5,000), then individuals with disabilities may not benefit as much.
• Immigration status. Eligibility for many tax credits depends on having a Social Security Number, which undocumented immigrants (and some other lawfully present immigrants) cannot obtain. Therefore, a “mixed-status” family with both documented and undocumented members can miss out on tax credits like the EITC, even if they have a child who is a US citizen. Beyond eligibility barriers, other factors like language access, familiarity with the tax system, and fear of interacting with government agencies can also affect uptake of tax credits, so tax credits may not reach all families in need equally.
• Gender. Women are more likely to be secondary earners in married couples and more likely to take time out of the labor force for caregiving, so a change in marginal tax rates for joint filers can affect their work incentives differently than men’s. When secondary earners face relatively high tax rates, additional work may yield limited financial returns—an effect that aggregate analyses can miss but that becomes clearer when examining impacts separately for women and men.
• Geographic region. Consider the federal deduction for state and local taxes. It provides greater benefits to residents of high-tax states than those in low-tax states. Likewise, housing-related tax incentives may deliver greater dollar gains in high-cost metropolitan areas, but may have more modest effects in rural communities where home prices are lower. Analysts may miss how the same federal policy translates into very different economic realities on the ground if they don’t examine impacts across regions—such as urban versus rural areas, or high-cost versus lower-cost states.  

This is not an exhaustive list of subgroups, nor are all of these important to look at in all analyses. Analysts will have an easier time choosing subgrouping variables with an understanding of historical and structural context and how policy design choices contribute to outcome differences.  

Tax analyses that focus only on summary data and do not disaggregate by subgroup risk being incomplete or misleading, impeding the development of equitable and effective tax policy. Subgroup analysis, with robust, disaggregated data, can help ensure policymakers understand who benefits, who pays, and how behavior changes across different segments of society. 

But, importantly, subgroup analysis is not an end in itself: The analyst seeking to inform communities and policymakers alike will meaningfully highlight the design choices and contextual conditions that contribute to differences—or close gaps.

Our new guide, complete with annotated examples and actionable tools, offers a path forward.