By Michael Pusic, J.D. candidate, Harvard Law School
STUDENT VOICES: The views expressed below are those of the student author and do not necessarily reflect the position of the Access to Justice Lab.
During his candidacy, President Trump promised to launch “the largest deportation program in American history.” While this campaign has been slow to start, Congress seems likely to appropriate $150 billion for border enforcement by March, and the Department of Homeland Security will strip 350,000 Venezuelans of protected status in April.
Once large-scale deportations begin, immigration courts will struggle to deliver due process. Already, the backlog of deportation cases has grown from 744,000 to 3.7 million over the past decade, leaving each judge with a docket of 4,500 cases. It is hard to believe that each case, each human being, will receive an individualized determination on its own merits. Compounding this problem is the fact that two-thirds of people facing deportation lack legal representation. Eighty-six percent of immigrants being detained do not have a lawyer. Without counsel, many will be deported to face persecution or violence – regardless of their legal right to stay.
One partial solution may lie in empowering non-lawyers to take a larger role in representing individuals facing deportation. This post examines promising evidence for this approach and proposes a randomized control trial to evaluate its effectiveness.
Accredited representatives could expand access to justice
Since 1975, U.S. law has allowed nonlawyers to represent immigrants in immigration court – a rare exception to legal practice rules. To participate, nonprofits must first be recognized by the Department of Justice, then apply to accredit particular employees as representatives. Once accredited, representatives can handle cases before immigration courts, appeals boards, and the DHS. An individual representative’s accreditation is tied to their employer, and they must renew every three years, but the program dramatically expands affordable legal help. There are 2,000 authorized representatives nationwide, and their efforts show significant promise.
Research suggests that non-lawyers may be able to match lawyers’ performance in certain contexts. In an early study, political scientist Herbert Kritzer compared the efficacy of lawyers and non-lawyers in four different contexts: Social Security Disability appeals, state tax appeals, unemployment compensation claims appeals, and labor grievance arbitrations. His study has since been criticized for a lack of statistical sophistication – Professor Kritzer applies statistical probability tests to only one of the four groups – but his qualitative analysis suggests that lawyers who are unprepared for hearings often do fail, and non-lawyers who are familiar with specialized proceedings “do not suffer from a lack of advocacy training.” A later observational study reached a similar conclusion. It examined outcomes for clients represented by lawyers and nonlawyers in over 5,000 unemployment appeals cases while modeling for some relevant factors (e.g., judge reviewing the case, case complexity). There, the authors found that “experienced nonlawyers can help parties through their expertise with common court procedures.”
Building on these findings, Moorhead, Sherr, and Paterson studied 82,705 cases assisted by lawyers and non-lawyers in the United Kingdom. Using multivariate analysis to model for key variables, they found that non-lawyer agencies statistically outperformed solicitors in civil legal aid cases across multiple measures, including peer review evaluations and case outcomes. In welfare benefits cases, solicitors were about a quarter as likely as non-lawyers to achieve positive financial results, while in employment cases, they were approximately half as likely. This evidence, though still limited by the fact that it all stems from observational (not randomized) studies, suggests that specialized training and experience may matter more than formal legal credentials in certain legal settings.
While there has been no comprehensive study on the efficacy of non-lawyers in the immigration context, anecdotal evidence from a number of programs using accredited representatives is promising. Catholic Charities of Santa Clara County employs six partially and four fully accredited representatives alongside just two attorneys, allowing them to assist more clients at lower cost. Program Director Robert Yabes notes that beyond cost savings (accredited representatives earn $40,000-55,000 compared to attorneys at $75,000), there’s little practical difference in capability: “It’s just a title difference. Pretty much they do the same thing.” The organization’s collaborative approach, where representatives discuss cases in team meetings and identify best practices, has resulted in impressive outcomes, with USCIS rarely rejecting their applications. As Yabes explains, this success stems from combining thorough client education, careful case selection, and staff who are “always engage[] in conversation” about their work.
Even so, there are concerns about expanding non-lawyer representation. Evidence about the efficacy of lawyers versus trained non-lawyers is sparse in any context, and evidence in the immigration context is merely anecdotal. The success of Catholic Charities is not necessarily representative of the hundreds of other programs using accredited representatives. Even if those programs are currently effective, they may break when scaled to a point at which attorneys cannot provide as much oversight and guidance. And all of this assumes good faith actors – the prevalence of immigration fraud, often by “notarios” falsely claiming legal expertise, might suggest a need for formal training and accreditation. All this leads many to worry that expanding non-lawyer practice could blur professional boundaries and leave immigrants more vulnerable to exploitation.
The twinned promise and risk of expanding these programs underscores the need for a credible study on the efficacy of non-lawyers in immigration cases. With lives at stake, policymakers and legal services organizations need evidence before either expanding or restricting these programs.
A randomized control trial (RCT) could validate this transformative tool
A RCT offers the ideal tool for resolving uncertainty in this area. The scarcity of adequate representation creates natural conditions for study: with most immigrants receiving no counsel, randomly assigning cases to either lawyer or trained non-lawyer representation would be both ethical and illuminating.
A study could track several dependent variables related to the outcome of the case: whether a client is detained, the length of their time in detention, and/or their sense of procedural justice. The key treatment variable would be type of representation they receive (lawyer, trained non-lawyer, or none), with randomization controlling for case complexity, client background, and other potential confounding factors.
Accredited representative programs like those at Catholic Charities could provide the experimental framework. Their intake systems already assess cases that, with informed consent, could be randomly assigned to different representation types. Such an effort would generate the first credible evidence on whether trained non-lawyers can effectively protect immigrants’ rights in removal proceedings.
The cost-benefit calculation is compelling. The funding invested in an RCT could unlock the potential of non-lawyer representation, potentially saving millions while expanding access to justice. Given constrained legal aid budgets, optimizing resource allocation through evidence-based policy is crucial.
With deportations accelerating and courts overwhelmed, the status quo is prone to widespread rights violations. Expanding qualified non-lawyer representation could ensure basic due process at low cost. The research framework exists to test this solution. The only question is whether we’ll gather the evidence before it’s too late for those facing removal without legal assistance.
If you’re interested in more details of this project, listen to our Proof Over Precedent podcast episode on the topic.