The New Practice Lab’s Racial Equity Framework

How we’re thinking about unpacking inequities in policies and programs, and our commitment to anti-racist policy making.
Blog Post
Sept. 24, 2020

Earlier this month, the Trump administration announced that it would cancel all race-related trainings at federal agencies. As the federal government becomes increasingly hostile toward addressing racial inequality, it is more important than ever that we bring an equity lens to all of the work we do at the New Practice Lab. Our deep dive into Unemployment Insurance (UI) uncovered a series of compounding racial inequities, both within the UI system and beyond, that impacted Black and Latinx workers who were seeking UI benefits. The important insight from this work was not that racist systems exist, but that they intersect and magnify each other, making the UI system unworkable for many recipients.


This finding has prompted us to create a framework to help us identify, unpack, and provide recommendations to rectify inequality in policies and programs moving forward. Understanding the origins of policies, who they were designed to serve, and the way in which racism has emerged either by design or though implementation, is crucial to designing an equitable safety net that serves the needs of all Americans. The following framework will provide an overview of the types of questions we will ask as we approach our work, and we hope that others will find them useful as they develop and implement policy. 

  1. Understanding the historical context and who this policy was meant to serve. Many programs were designed with a particular beneficiary in mind and purposefully excluded certain populations. For example, at its inception, Social Security did not include farm workers or domestic workers, who were more likely to be Black and Brown workers, and women. Today, UI still excludes many jobs held disproportionately by workers of color. Understanding that the program was exclusionary from its start can help us understand potential challenges in its implementation today. What coded language were policymakers using to exclude certain groups, explicitly or not? By exploring who a policy favored and who it did not, who it was meant to serve and who it was not, and how it has evolved over time, we gain leverage to understand which populations must have their needs addressed for the program to become more responsive and equitable. 
  2. Who benefits today? To better understand how a policy functions today, we need to ask who the policy is designed to provide the largest and smallest benefits to, who is excluded, and who actually receives the benefits. For example, who receives the largest benefit, and what proportion of people eligible actually receive the benefit? Conversely, who receives the smallest benefit, and how many people who are eligible actually receive it? Or who is eligible but chooses not to even apply, and why? Examining who is left out, both explicitly and by the program's implementation, can provide insights into who is most marginalized by policy decisions. This allows us to see which parts of the policy may be inequitable, and how the implementation may or may not exacerbate inequality. 
  3. What larger institutions are these policies embedded within? Policies are not created in a vacuum - they are the product of institutions, and they interact with them. For example, the UI program exists within the larger institution of the workplace. Black and Brown workers are discriminated against in hiring, and are often “first fired, last hired.” A UI program that requires a recipient to search for work will automatically be more punitive to Black and Brown workers compared to white workers, as it collides with the racist preferences of employers. Taking into account these dynamics helps to expand our understanding of the problem and the need for a comprehensive solution. 
  4. What other systems does this policy engage with? Each policy system intersects with other systems that may also be discriminatory. In our UI work, one important source of overlapping disadvantages was unequal access to broadband. UI systems already excluded many workers and provided meagre benefits, undermined by inaccurate and harmful beliefs about low-income workers and workers of color and claims that access to UI would undermine work. As states have made UI forms available online and prioritized processing these applications over ones submitted by mail, this discriminatory design intersected with another racially unequal system: access to broadband. Workers of color and low-wage workers are less likely to have reliable access to the internet or a computer at home, so a discriminatory UI system collided with a lack of investment in infrastructure to make benefits even harder to access. This dynamic may be present in many other policies, and has important impacts on inequality. 
  5. Examining the role of technology. Sometimes, technical solutions can improve access to government programs and expand who benefits from a program. However, applying a technical solution can also hardcode existing inequities, making the program even less responsive and accessible to the needs of certain populations. Applying technical solutions must be sure to ask what problem the technology is fixing, what bias might be carried over from the previous process and hardcoded into the new one, who has access to the technology and who does not, and what type of impact that might have. 
  6. What can we learn from both qualitative and quantitative methods? Quantitative methods can provide information on scope, highlight inequality between groups, and provide a big picture understanding of a program. Qualitative research, including user research, and ensuring recipients are included in the design of a program, is necessary to understand recipient needs and how to meet them. Relying on a variety of methods and types of expertise provides the most comprehensive and inclusive approach to policy design. 
  7. Designing for the 10%. An important principle we consider is to design policy for the 10% of the population that is hardest to reach. We believe that by designing for those who are hardest to serve and most in need, the policy will work better for everyone. This means approaching policy design with an equity lense from the outset. 

This framework is still evolving, and our plan is to create something that is shareable and usable for policymakers and program administrators when evaluating a new or existing policy. But we are committed to actively identifying anti-Blackness, racial biases, and inequities that exist in our policies and harm communities of color, immigrant communities, indigenous communities, and other vulnerable people. 

We invite others to join us in thinking through these issues by emailing us at