The Lawyer Pipeline: Where Applicants, Enrollment, Bar Passage, and Jobs Diverge

Abstract

We examine the pipeline that produces new lawyers in the United States, from the point of application through licensure and into the labor market. Using publicly documented series from the Law School Admission Council (LSAC), the American Bar Association Section of Legal Education and Admissions to the Bar (ABA 509 disclosures and bar passage reports), the National Conference of Bar Examiners (NCBE), the National Association for Law Placement (NALP), and the Bureau of Labor Statistics (BLS), we describe how the population of prospective lawyers contracts at successive stages. The applicant pool has moved through a documented boom near 2004, a long contraction that reached a trough in the mid 2010s, and renewed volatility in the pandemic and post pandemic cycles. First year enrollment tracked that curve with a lag. Bar passage removes a further share, and first time pass rates vary widely by jurisdiction and by school. Employment outcomes ten months after graduation show that the category of full time, long term, bar passage required work absorbs most but not all graduates. BLS projections point to modest occupational growth and a replacement driven flow of annual openings. We argue that the pipeline narrows unevenly, and that the narrowest points shift across cycles rather than sitting at a single fixed stage.

Background

The path to becoming a practicing lawyer in the United States runs through a small number of well marked gates. A person decides to apply. They take an admissions test, historically the LSAT and now also the GRE at many schools. They enroll, usually in a three year Juris Doctor program at an ABA approved law school. They graduate. They sit for a bar examination in one or more jurisdictions. They pass, or they do not. They look for work as a lawyer, or they take a job that a law degree helped them get without requiring bar admission.

Each gate is measured by a different body, and the bodies do not always agree on definitions, timing, or population. That fragmentation is the central methodological problem of studying the pipeline. LSAC counts applicants and applications. The ABA counts enrolled students and reports bar passage. The NCBE builds and scores the national components of the exam and reports aggregate performance. NALP surveys graduates about jobs. BLS counts employed lawyers in the broader economy and projects future demand. No single agency owns the full sequence.

The stakes are practical. Prospective students weigh three years of tuition and lost income against uncertain employment. Schools manage enrollment against accreditation standards that tie their status to graduate outcomes. Regulators weigh licensing rules that affect both consumer protection and access to the profession. And the labor market absorbs, or fails to absorb, whatever the pipeline delivers.

Two structural facts frame everything that follows. First, the number of ABA approved law schools has stayed close to 196 in recent years, a slow moving denominator against a volatile flow of applicants. Second, the profession is large and mature. BLS estimates the employed lawyer population in the neighborhood of 850,000 to 860,000 in its recent releases, which means the annual class of new graduates, on the order of tens of thousands, is small relative to the standing stock. That ratio matters when we interpret projected openings.

Data and method

This is a descriptive study. We do not run a causal model. We assemble published series from the primary bodies that measure each stage and we read them together, aligning by graduation cohort and admission cycle where the timing allows.

For applications and applicants we rely on LSAC end of cycle volume summaries, which report the number of persons who applied to at least one ABA approved school and the total count of applications submitted. For enrollment we use the ABA 509 required disclosures, which each school files and which the Section aggregates into national totals for first year JD enrollment and total JD enrollment. For licensure we use two sources: NCBE statistics on examinee volume and national mean scaled scores on the Multistate Bar Examination, and the ABA bar passage disclosures, which report first time and ultimate pass outcomes tied to graduating classes. For employment we use the NALP Jobs and JDs reports and the ABA employment summary data, both of which classify jobs by whether bar passage is required, whether a JD gives an advantage, and whether the position is full time and long term. For occupational demand we use the BLS Occupational Outlook Handbook entry for lawyers and the underlying Employment Projections program.

Where a series reports an exact number we cite it and name the source in text. Where the most recent figure is provisional, or where different sources define the population differently, we describe direction and magnitude in words rather than asserting a false precision. We treat the ABA disclosures as the authoritative count of enrolled students and graduates, because accreditation compels their filing. We treat NALP and ABA employment data as complementary; they cover overlapping cohorts using similar but not identical categories.

A note on timing. An admission cycle that produces enrollment in the fall generates graduates roughly three years later, bar results a few months after that, and NALP employment measurement about ten months after graduation. A single cohort therefore appears in different series across a four year span. We keep that lag explicit when we connect stages.

Findings

Applicants: a long cycle with sharp turns

The applicant pool is the widest and the most volatile part of the pipeline. LSAC data show a peak in the mid 2000s, with the number of applicants to ABA approved schools reaching roughly 100,000 around the 2004 cycle. That was a high water mark. The pool then entered a long decline that accelerated after the 2010 cycle, as the aftermath of the financial crisis and a wave of reporting on weak legal employment discouraged applications.

By the mid 2010s the pool had contracted to a trough. LSAC counts placed applicants near 56,000 in the 2015 cycle, a level not seen in decades. That is roughly a 40 percent contraction from the peak in about a decade. The decline was persistent rather than a single bad year, which is why it reshaped the enrollment picture that followed.

Then the direction reversed. The 2018 cycle showed an uptick sometimes attributed to renewed interest in law and public affairs. The pandemic cycles brought a sharp surge: the 2021 cycle recorded a large year over year jump in applicants, on the order of a double digit percentage increase, as remote testing lowered friction and a weak entry level job market pushed people toward graduate education. That surge did not hold at the same level. The following cycles gave some of it back, then the 2024 and 2025 cycles again showed strong increases in applicant volume reported by LSAC, restoring the pool toward levels last seen in the early 2010s.

Two features of the applicant series deserve emphasis. The swings are large and fast. And the pool responds to signals about jobs and the broader economy with a lag, which means applicant counts often move against the labor market that graduates will actually face three years later. People apply into a strong signal and graduate into a different one.

Enrollment: the flow that schools control

First year JD enrollment, reported through the ABA 509 disclosures, tracks the applicant curve but with damping. Schools admit from whatever pool exists, and they adjust selectivity and class size to hit their own targets. When the applicant pool collapsed, enrollment fell, but by less in percentage terms, because schools accepted a larger share of a smaller pool.

First year enrollment had exceeded 52,000 around the 2010 cycle at the top of the boom. It then fell through the first half of the 2010s, settling into a range near 37,000 to 38,000 first year students in recent years. Total JD enrollment across all three years has moved in a band roughly from the high 110,000s into the 120,000s depending on the cycle, with the most recent disclosures showing renewed growth consistent with the applicant rebound.

The composition of that enrollment changed alongside the count. The share of students admitted with the GRE rather than the LSAT rose as more schools accepted it. The demographic composition of entering classes shifted, with the ABA disclosures showing growth in the enrolled share of students of color over the decade. And credential distributions moved: during the trough years, median LSAT scores at many schools declined as competition for a shrinking pool intensified, then firmed again as the pool recovered.

The enrollment stage is where institutional behavior is most visible in the data. The applicant pool is a market. Enrollment is a set of decisions by 196 schools operating under accreditation standards, revenue pressure, and ranking incentives. That is why enrollment is smoother than applications, and why it can rise even when the eventual job market is uncertain.

Attrition and graduation: a quieter contraction

Between enrollment and the bar exam sits a less discussed narrowing: academic and other attrition during the three years of study. The ABA disclosures report attrition, and while most enrolled students graduate, a meaningful minority do not complete on schedule or at all. Attrition is concentrated in the first year and varies substantially across schools, with higher rates at institutions that admitted classes with lower entering credentials during the trough years.

The result is that the graduating class is smaller than the entering class three years earlier, and the gap is not uniform across institutions. When we align an entering cohort with its graduation, the pipeline has already lost a slice before anyone reaches a bar exam. This stage rarely draws headlines, but it is a real narrowing, and it falls hardest on the students who were admitted at the margins.

Bar passage: the gate with the widest variance

Licensure is where the pipeline narrows most visibly and most unevenly. The bar examination has historically combined a national multiple choice component, the Multistate Bar Examination, with essay and performance components, scored and weighted differently across jurisdictions. Passing scores differ by state, which means an identical performance can pass in one jurisdiction and fail in another.

NCBE data show that national MBE mean scaled scores are not fixed. They dipped through the mid 2010s, drawing considerable attention and debate about whether declining entering credentials during the enrollment trough were feeding through to weaker exam performance. Scores recovered somewhat later in the decade and then moved again through the pandemic administrations, when testing conditions themselves were disrupted and several jurisdictions offered remote or modified exams.

First time bar pass rates, reported in the ABA disclosures, generally sit well above overall pass rates because repeat takers pass at lower rates. Nationally, first time pass rates in recent cycles have commonly landed in a range from the mid 70s to the low 80s in percentage terms, but that aggregate hides enormous spread. At the strongest schools, first time passage approaches or exceeds the high 90s. At schools that struggled during the trough, first time rates have fallen well below the national figure, sometimes under 50 percent in a given administration.

The ABA responded to this variance with an accreditation standard that ties a school’s status to bar outcomes. Under the revised Standard 316, an approved school must show that at least 75 percent of its graduates who sat for a bar exam passed within two years of graduation. That is an ultimate pass rate requirement, and it converted bar performance from a reputational concern into an accreditation risk. Some schools with persistent weak outcomes have faced ABA scrutiny, remediation requirements, or, in a handful of documented cases, loss of accreditation or closure.

The regulatory picture is shifting again. The NCBE has developed a redesigned examination, the NextGen bar exam, scheduled for first administration in July 2026, which restructures how knowledge and skills are tested. Because that first administration falls after the period this study covers, we describe it only as a scheduled change, not as an outcome. Its arrival means that bar passage series across the 2025 to 2027 window will span two different examination designs, which will complicate direct comparison for anyone reading the numbers later.

Bar passage narrows the pipeline in two ways. It removes those who fail outright and never pass. And it delays those who pass on a later attempt, pushing them into the job market off cycle, often after their classmates have already been placed. Both effects are concentrated at the same schools, which compounds the disadvantage for their graduates.

Employment: what the degree buys ten months out

The NALP and ABA employment data measure outcomes about ten months after graduation, a timing chosen to allow for bar results and the fall hiring that follows. The central metric is the share of graduates in full time, long term positions that require bar passage. A second category captures jobs where a JD provides an advantage but bar admission is not required. A third captures other professional and non professional work, part time or short term positions, and those still seeking employment.

For recent classes, the overall employment rate reported by NALP has been strong, commonly in the low 90s in percentage terms when all employment categories are combined. The more demanding measure, full time long term bar passage required employment, has run lower, generally in the neighborhood of the high 70s to low 80s depending on the class. The gap between those two numbers is the part of the pipeline that ends in something other than the traditional lawyer job: JD advantage roles, business and government positions, and a residual of graduates who are underemployed or still searching.

The classes that graduated into the post pandemic hiring years benefited from a period of unusually strong demand at large law firms, which lifted the bar passage required share and pushed reported salaries upward at the top of the distribution. NALP salary data show a persistently bimodal structure: a cluster of high starting salaries at large firms, historically anchored by a widely followed market rate that has climbed in steps to 225,000 dollars for first year associates at the largest firms, and a much larger cluster of far lower salaries across smaller firms, government, and public interest work. The mean sits between the two peaks and describes almost no one.

That bimodal shape matters for how we read the pipeline’s exit. A graduate’s outcome depends heavily on which school they attended and how they placed within their class, because large firm hiring concentrates at a subset of schools and among students near the top of their cohorts. The same national employment rate can describe a graduate earning 225,000 dollars and a graduate earning a third of that, and the pipeline delivers both from the same starting gate.

Occupational demand: a mature market with replacement flow

The BLS Occupational Outlook Handbook entry for lawyers frames the demand side. BLS reports a median annual wage for lawyers around 145,000 dollars in its recent release, well above the median for all occupations, and it projects employment growth for the occupation that is modest rather than rapid, in the single digit percentage range over its ten year projection horizon.

The more informative number for pipeline purposes is annual openings. BLS projects on the order of 40,000 openings for lawyers per year across the projection decade. That figure is dominated by replacement needs, meaning openings created as current lawyers retire or leave the occupation, rather than by net new positions from growth. In a mature profession with a large standing stock, replacement demand is the engine, and net growth is a smaller add on.

Placed against roughly 38,000 first year enrollees and a somewhat smaller number of eventual graduates who pass a bar and seek lawyer work, the projected 40,000 annual openings look, in aggregate, roughly commensurate with supply. But aggregate balance conceals mismatch. Openings are distributed by geography, practice area, and employer type in ways that do not line up neatly with where graduates want to work or where they are willing to live. A rough national parity between openings and new entrants can coexist with local shortages in rural practice and local gluts in saturated urban markets.

Reading the stages together

When we align the stages by cohort, the pipeline is best described as a series of narrowings whose relative severity moves across cycles. Consider a stylized progression using the orders of magnitude documented above. An admission cycle draws somewhere between 60,000 and 100,000 applicants depending on where in the cycle it sits. Those applicants generate roughly 38,000 first year enrollees. Attrition and non completion trim that before graduation. Bar passage removes a further share, concentrated at particular schools, and delays others. Employment measurement then finds most graduates working, but sorts them into bar passage required roles, JD advantage roles, and a residual.

The point we want to make is that no single stage is the permanent bottleneck. During the trough years of the mid 2010s, the widest narrowing was at the application stage, where the pool itself collapsed and schools scrambled to fill seats. In the same period, bar passage tightened as the accreditation standard bit and pass rates fell at weaker schools. During the strong post pandemic hiring years, the employment exit widened for graduates of well positioned schools while the applicant pool surged. The narrowest point is not fixed. It migrates with the economic cycle, with regulatory change, and with the lagged relationship between when people apply and when they graduate.

Discussion

Three patterns emerge from reading the series together, and each has consequences for how the profession replenishes itself.

The first is the timing mismatch. Applicants respond to the labor market they observe when they apply, but they graduate into the market that exists three years later. When entry level jobs are weak, as during the early pandemic, applications rise. When those applicants graduate, conditions may have changed. This lag builds a structural instability into the pipeline. It also means that the strong applicant cycles of 2024 and 2025 will deliver larger graduating classes near the end of the decade, into a labor market whose condition we cannot read from current data.

The second is the concentration of risk at particular schools. Attrition, bar failure, and weak employment outcomes are correlated, and they cluster at the same institutions. A student who enrolls at a school with low entering credentials faces a higher probability of not graduating, a higher probability of failing the bar if they do graduate, and a lower probability of a bar passage required job if they pass. These are not independent risks stacked randomly across the pipeline. They compound along the same path for the same students, and the accreditation standard on bar passage was a direct regulatory response to that concentration.

The third is the divergence between aggregate balance and distributional reality on the demand side. BLS projections suggest that the total flow of openings and the total flow of new lawyers are, at a national level, in rough proportion. Yet the bimodal salary structure documented by NALP, and the geographic distribution of openings noted by BLS, mean that the experience of individual graduates varies enormously around that aggregate. A national figure that looks balanced can sit on top of a market where some graduates command high salaries at large firms while others struggle to find full time legal work, and where some rural jurisdictions report difficulty attracting lawyers at all.

There is a regulatory layer moving underneath all of this. The bar examination is being redesigned, with the NextGen exam scheduled to begin in July 2026. Several jurisdictions have experimented with or adopted alternative licensure pathways that reduce or replace the traditional examination, and the debate over whether the bar exam is the right gate at all has grown more prominent. Because the documented outcomes of these changes fall largely after the period we study, we flag them as sources of future discontinuity rather than as measured effects. Anyone comparing pipeline data across the second half of this decade will need to account for the fact that the licensure gate itself is being rebuilt while people pass through it.

What we do not find is evidence for the simplest stories. The pipeline is not uniformly oversupplied, because aggregate openings and new entrants are roughly commensurate and demand is replacement heavy. It is not uniformly undersupplied, because saturated markets and a large residual of graduates outside bar passage required work both persist. And it does not narrow at one predictable place, because the narrowest point moves. The honest description is a system with multiple gates, each of which can tighten independently, whose combined effect on any individual depends heavily on where that person entered.

Limitations

Several constraints bound what we can claim. The sources define their populations differently. LSAC counts persons who applied to at least one ABA approved school, which excludes non ABA and foreign trained candidates who may still enter the profession through other routes. The ABA counts enrolled JD students at approved schools. NALP surveys graduates through their schools with high but incomplete response, and the ABA employment data use categories that overlap with but do not perfectly match NALP’s. BLS counts employed lawyers across the whole economy using occupational codes that capture people whose day to day work is legal practice, which is a different population from recent graduates. Stitching these together requires accepting that the seams are imperfect.

Timing is a second constraint. Because each stage is measured at a different point relative to graduation, aligning them requires assumptions about lag. We have tried to make those assumptions explicit, but a graduate who passes the bar on a second attempt, or who takes a job after the ten month measurement window, will be recorded differently than the timing model assumes.

Third, we have deliberately used ranges and directional language where the most recent figures are provisional or where sources disagree. Readers who want a single authoritative number for a specific class year should consult the underlying disclosure for that year rather than the ranges we give, which are chosen to be robust across cycles rather than precise for one.

Fourth, this is a national study. It cannot capture the state level variation in bar passing scores, licensure rules, and labor markets that shapes individual outcomes. A pipeline that looks balanced nationally can be tight in one state and slack in another.

Finally, the regulatory environment is changing during the window we study. The redesigned bar exam and alternative licensure experiments mean that series which were comparable in the past may not be comparable going forward. We have restricted our factual claims to documented data through late 2025 and have described scheduled future changes as scheduled, without projecting their results.

Conclusion

The lawyer pipeline is often discussed as though it had a single chokepoint, a place where the profession either lets people in or keeps them out. The data do not support that framing. Applications swing by tens of thousands across a decade. Enrollment damps those swings but follows them. Attrition quietly removes a slice before graduation. Bar passage removes and delays another slice, unevenly, concentrated at the same schools where entering credentials are lowest. Employment measurement then finds most graduates working, but sorts them into a bimodal structure where the same national rate describes very different lives. And BLS projections describe a mature occupation whose openings are driven mainly by replacement, roughly commensurate with new supply in the aggregate while diverging sharply by place and practice.

The narrowest point in this pipeline is not fixed to one stage. It migrated from the application gate during the mid 2010s trough, tightened at the bar passage gate as accreditation standards took hold, and eased at the employment exit for well positioned graduates during the strong post pandemic hiring years. The lag between applying and graduating guarantees that the recent surge in applicants will deliver larger classes into a labor market we cannot yet read. Anyone trying to forecast the supply of lawyers from any single stage will be wrong, because the stages move independently and the people who pass through them face compounding rather than independent odds.

References

Cite this study

VerifiedLawFirms Research Desk (2026). The Lawyer Pipeline: Where Applicants, Enrollment, Bar Passage, and Jobs Diverge. VerifiedLawFirms. https://verifiedlawfirms.com/blog/study-legal-employment-pipeline/

Link back to /blog/study-legal-employment-pipeline/ when citing.