The State-by-State Geography of the American Legal Profession

Abstract

This study examines the geographic distribution of licensed lawyers in the United States, drawing primarily on the American Bar Association National Lawyer Population Survey, supplemented by federal population estimates and Bureau of Labor Statistics occupational data. We describe the national total, which sat above 1.3 million active lawyers in the early 2020s, and we trace its modest movement across the decade from roughly 2013 through 2023. Three jurisdictions dominate the count. New York and California each host well over 150,000 active lawyers, and the District of Columbia records a density that is an order of magnitude above any state. We then convert raw counts into lawyers per capita, which reveals a wide gap between dense coastal and metropolitan states and sparsely served interior states. The final section addresses rural access to justice, where county-level analysis has documented that a large share of United States counties contain fewer than one lawyer per thousand residents, and a smaller set contain none at all. We read the evidence as showing a profession that has grown slowly in aggregate while concentrating spatially, leaving a persistent and measurable gap between where lawyers practice and where clients live. We close with the limits of licensing-based counts and the reforms some states have tried.

Background

The question of how many lawyers a country has is older than any survey. What has changed is our ability to answer it with some precision, state by state, year over year.

In the United States, admission to practice is a state function. There is no national bar license. A lawyer is admitted by the highest court of a state or by a state bar acting under that court’s authority, and the count of admitted lawyers is therefore a sum of separate jurisdictional registries. That structure matters for anyone trying to measure the profession, because it means the national figure is assembled from more than fifty independent sources, each with its own definition of active and inactive status, its own treatment of retired and judicial members, and its own reporting calendar.

The American Bar Association has published a National Lawyer Population Survey for decades to standardize this assembly. Each year the ABA requests membership counts from the mandatory and voluntary bars of every state, the District of Columbia, and the territories, and it reports the resulting totals. The survey is the most widely cited single source for state-level lawyer counts, and we treat it as our primary instrument here.

Why does the geography matter? Two reasons run through the legal literature.

The first is economic. Legal services are a large sector, and lawyers cluster where the demand for high-value legal work clusters, which is to say near financial centers, corporate headquarters, seats of government, and the courts that hear the disputes those institutions generate. The distribution of lawyers is, in part, a map of where money and regulation meet.

The second reason is about access. Courts have long recognized that the presence or absence of counsel changes outcomes. In criminal matters the constitutional baseline is settled. Gideon v. Wainwright, 372 U.S. 335 (1963), held that the Sixth and Fourteenth Amendments require states to provide counsel to indigent defendants in felony prosecutions. On the civil side the picture is far less generous. The Supreme Court declined to recognize a categorical right to appointed counsel in civil contempt proceedings in Turner v. Rogers, 564 U.S. 431 (2011), and it had earlier refused a per se rule for parental termination cases in Lassiter v. Department of Social Services, 452 U.S. 18 (1981). In civil matters, then, whether a person has a lawyer often turns on whether a lawyer is available and affordable nearby. Geography becomes a determinant of access. A county with no practicing attorney is a place where the civil right to be heard is thinner than the words of the statute suggest.

Those two forces, concentration by demand and scarcity by place, sit in tension. This study tries to measure both.

Data and method

Our core dataset is the ABA National Lawyer Population Survey, which reports active and resident lawyer counts by jurisdiction. From it we take national totals and per-state totals for the years available in the modern series, roughly 2013 through the mid-2020s. We use the survey’s own definitions and do not attempt to re-classify inactive, retired, or judicial members across states.

For population denominators we rely on the United States Census Bureau, using intercensal and postcensal state population estimates and the 2020 decennial count. Dividing lawyer counts by population yields lawyers per capita, which we express as lawyers per one thousand residents to keep the numbers legible.

We bring in a second occupational measure from the Bureau of Labor Statistics. The BLS Occupational Employment and Wage Statistics program counts employed lawyers, meaning people paid to work as lawyers, and its national figure sits well below the ABA licensed count. The gap is expected and instructive. A person can hold an active license and not practice, or practice part time, or work in a role that the BLS classifies under a different title. We use the BLS figure as a lower bound on working lawyers and the ABA figure as an upper bound on licensed capacity, and we say so wherever the distinction affects a conclusion.

For the rural analysis we draw on county-level work that maps attorney counts against county population. The most cited body of that work comes from the ABA Profile of the Legal Profession and from academic studies of legal deserts, which we name in the references. County data is coarser and older than state data, so we characterize its findings by direction and magnitude rather than reporting a single precise year figure where the underlying counts are uncertain.

Three cautions govern everything that follows. Licensing is not the same as practice. Residence of record is not the same as place of work, which matters enormously for the District of Columbia. And a state total tells you nothing about the internal distribution within that state. We return to each caution in the Limitations section, but the reader should hold them from the start.

Findings

National totals and the decade trend

The headline number is large and slow moving. The ABA National Lawyer Population Survey placed the total count of active lawyers in the United States above 1.3 million in the early 2020s. That is a substantial professional class by any measure, and it dwarfs the working count reported by the BLS, which has hovered near three quarters of a million employed lawyers in the same period. The spread between those two numbers, on the order of half a million people, is the population of license holders who are not, at any given moment, counted as employed lawyers by the federal survey.

The decade trend is the part that surprises people who expect a flood. Between the early 2010s and the early 2020s the licensed total grew, but modestly. The increase across the ten-year window was in the low single digits as a percentage, which averages to well under one percent per year. Set that against United States population growth over the same window, which ran at a broadly similar low rate, and the ratio of lawyers to people held roughly flat at the national level. The profession did not shrink. It also did not surge. It expanded at about the pace of the country it serves.

Several mechanics sit under that flat line. Law school enrollment fell sharply after its 2010 peak and then stabilized at a lower level, which reduced the inflow of new admittees relative to the boom years. At the same time a large cohort of lawyers admitted in the 1970s and 1980s reached retirement age, which increased the outflow. New entrants and departing seniors roughly offset one another. The aggregate number moved little because two large flows were close to balanced.

We note this because the public conversation often assumes a lawyer glut that keeps growing. The data through 2023 does not support a story of relentless expansion. It supports a story of a mature profession near a plateau, with the interesting variation hiding not in the total but in where those lawyers sit.

The concentration in New York, California, and the District of Columbia

Distribution is where the survey earns its keep. A handful of jurisdictions hold a share of the national total far larger than their share of the national population.

New York and California are the two largest by a wide margin. Each reports well over 150,000 active lawyers in the ABA series, and between them they account for roughly a quarter of every licensed lawyer in the country. New York’s count reflects the gravitational pull of the New York City legal market, which combines the largest concentration of corporate finance work in the world with the state and federal courts that adjudicate it. California’s count reflects a different mix, spread across the technology economy of the Bay Area, the entertainment and media economy of Los Angeles, and a large and litigious general economy statewide.

Behind the top two the counts fall off but remain large. Texas and Florida each report totals in the neighborhood of tens of thousands, reflecting their size and rapid growth. Illinois, anchored by Chicago, sits in a similar band. Together the five or six largest states hold roughly half of all United States lawyers, a concentration that mirrors the concentration of the national economy in those same states.

Then there is the District of Columbia, which breaks every scale.

In raw count the District is not the largest jurisdiction. In density it is not close to any rival. The District records active lawyers at a rate that works out to something on the order of dozens per thousand residents, roughly one active lawyer for every handful of people who live there. No state approaches that figure. The explanation is structural and it is important for interpretation. The District is a small federal city whose lawyer population is defined by the national government, not by its resident base. Lawyers admitted in the District practice before federal agencies, federal courts, Congress, and the national headquarters of trade associations and firms. Many of them do not live in the District at all. They live in Maryland or Virginia and commute, but they are counted where they are admitted or where their office of record sits.

This makes the District’s per capita figure a genuine outlier that must be read with care. It is not evidence that residents of the District enjoy unmatched personal access to counsel. It is evidence that the seat of the federal government concentrates a specialized national legal workforce inside a small geographic boundary. We keep the District in our density tables because it is real and documented, and we flag it every time as a jurisdiction whose numerator belongs to the nation while its denominator belongs to a single city.

Strip the District out and the concentration story becomes a story about states. The northeastern corridor and the large coastal metropolitan states hold a disproportionate share of the profession. The interior and the rural South and Mountain West hold less. That pattern is stable across the years of the survey. It is the base map on which every other finding sits.

Lawyers per capita across the states

Raw counts favor big states by construction. To compare a small state fairly with a large one, we divide by population.

The national ratio sits near four active lawyers per thousand residents. That single figure hides a range that is wide at both ends.

At the high end, once the District is set aside as the extreme outlier it is, the densest states cluster in the Northeast and among the older industrial and financial centers. New York, Massachusetts, and Connecticut consistently report the highest per capita ratios among the states, comfortably above the national average. Illinois and Colorado tend to appear near the top as well. These are states with large legal markets attached to financial services, insurance, higher education, and state capitals that generate steady regulatory and litigation work. Their density reflects a deep base of institutional clients.

At the low end sit a group of fast-growing and heavily rural states. Arizona, South Carolina, and North Carolina have appeared repeatedly among the states with the fewest lawyers per resident, and several Mountain West and Deep South states join them near the bottom. Some of this reflects rapid population growth outrunning the growth of the bar. Some of it reflects economies with fewer of the corporate and financial clients that concentrate lawyers. Whatever the cause, a resident of a bottom-ranked state lives in a place with roughly half the lawyer density, per person, of a resident of a top-ranked state.

We want to be precise about what this ratio does and does not measure. A high state ratio does not mean a resident can easily find and afford a lawyer, because the lawyers may all sit in one metropolitan county and charge rates aimed at corporate work. A low state ratio does not mean the state is uniformly underserved, because its one large city may be perfectly well supplied while its countryside is empty. The per capita figure is a useful comparison across states. It is a poor description of access within a state. That within-state distribution is the subject of the next subsection, and it is where the access problem actually lives.

One further pattern deserves mention. The states that host a major national or regional financial center pull their whole state ratio upward, because the lawyers who serve that center are counted against the entire state population. New York State’s high ratio is really New York City’s ratio spread thin across upstate. The same effect operates, at smaller scale, wherever a single dominant metropolis anchors a state’s legal economy. The per capita map, read carefully, is often a map of where the largest cities are.

Rural counties and access-to-justice deserts

State averages conceal the sharpest finding in this study. When analysts descend from the state to the county, the distribution of lawyers turns out to be far more uneven than any state ratio suggests.

County-level work associated with the ABA Profile of the Legal Profession and with academic studies of rural practice has documented a consistent pattern. A large share of United States counties, on the order of two in five, contain fewer than one lawyer per thousand residents. A smaller but real set of counties, numbering in the dozens, contain no actively practicing attorney at all. These are the places the literature calls legal deserts. They are overwhelmingly rural. They are concentrated in the Great Plains, the Mountain West, parts of the rural South, and stretches of Appalachia and the upper Midwest.

The mechanics of a legal desert are not mysterious. Rural populations are older and shrinking in many counties, which reduces the client base and the fee base. The rural bar itself is aging, and when a long-established country lawyer retires there is frequently no younger attorney to take over the practice. Law graduates carry debt that pushes them toward higher-paying urban and suburban work. A young lawyer who might prefer rural life often cannot make the arithmetic work against student loans and the thin, dispersed demand of a rural county. The result compounds over time. Each retirement without a replacement widens the desert.

The consequences are practical and legal at once. A person in a county with no lawyer must travel, sometimes far, to file a probate matter, defend a foreclosure, seek a protective order, resolve a land dispute, or handle the countless civil matters that require competent counsel. The constitutional floor for civil cases is low, as Turner v. Rogers and Lassiter make clear, so there is no general right to have a lawyer appointed. Availability is left to the market, and in a legal desert the market has failed to supply the good. The gap between the formal right to be heard and the practical ability to be represented is widest exactly where lawyers are scarcest.

Several states have treated this as a policy problem rather than an unavoidable fact. South Dakota created the first statewide rural attorney recruitment program in 2013, an incentive scheme administered through its Unified Judicial System that offers a payment, funded jointly by the state, the county, and the state bar, to attorneys who commit to practice in an eligible rural county for a set term. Nebraska, North Dakota, and other states have developed rural practice initiatives, law school pipeline programs, and loan-related incentives aimed at the same shortage. These efforts are small relative to the size of the gap, and their measured effect has been incremental rather than transformative, but they represent an explicit recognition that geography, not headcount, is the binding constraint.

We read the county evidence as the most consequential part of the whole distribution. The national total is large. The state ratios are unequal. But the county map is where the inequality becomes a denial of service, because a county with zero lawyers is a place where the average national figure of 1.3 million license holders means nothing at all to the person who needs one.

Discussion

Pull the findings together and a coherent picture emerges. The United States legal profession is numerically large, demographically mature, and spatially concentrated. It grew slowly over the decade we examined, roughly keeping pace with population, so the national ratio of lawyers to people barely moved. The action was never in the total. It was in the geography.

Concentration is the first through-line. New York, California, and a short list of other large states hold the bulk of the profession because they hold the bulk of the client base that supports high-value legal work. The District of Columbia holds an extraordinary density because the federal government sits inside its borders and staffs a national legal workforce within a single small city. None of this is accidental, and none of it is likely to reverse, because the underlying economic geography that produced it is stable. Lawyers follow clients, and the largest clients cluster.

Scarcity is the second through-line, and it is the one with public consequences. The same market logic that concentrates lawyers in profitable metropolitan practice pulls them away from rural counties where demand is thin and dispersed. The legal desert is the shadow cast by the corporate skyline. The two phenomena are the same phenomenon seen from opposite ends. A profession that allocates itself by paying demand will overserve the places with the most paying demand and underserve the places with the least, and the county data shows exactly that allocation.

What should we make of the reform efforts? The rural recruitment programs are honest attempts to counteract market allocation with a modest subsidy. They work at the margin. They are not scaled to the size of the problem, and nothing in the data suggests they will close the desert by themselves. The larger levers, if states want to pull them, sit elsewhere: in the economics of legal education and debt, in the regulation of who may deliver legal services and how, and in the technology that lets a lawyer in one place serve a client in another. Several states have loosened rules on remote practice and on limited-license or paraprofessional providers, and those changes may bear more on rural access over time than any single recruitment bonus. We report this as a direction, not as a settled result, because the evidence on those reforms is still accumulating.

There is one interpretive trap we want to close. It is tempting to read a high national count and conclude that the country has more than enough lawyers. The distribution data refutes that inference. A country can have too many lawyers in one county and none in the next, and the sum can look comfortable while the median experience is one of scarcity. Aggregate sufficiency and local availability are different questions. The ABA total answers the first. The county map answers the second, and its answer is less reassuring.

Our overall reading is that the geography of the profession is stable, unequal, and only weakly responsive to policy so far. The concentration in a few states is durable. The rural deficit is durable. The reforms are real but small. Anyone who wants the distribution to change should expect to work against a market that has arranged itself this way for reasons that are economic and persistent.

Limitations

This study inherits every limit of licensing-based measurement, and readers should weigh each one.

Licensing is not practice. The ABA counts active license holders, not working lawyers, and the gap between that count and the BLS employment figure is roughly half a million people. Some hold licenses while working outside law, in business, government, academia, or retirement. A state’s licensed count therefore overstates its supply of practicing counsel, and it does so unevenly, because states differ in how they classify inactive and retired members.

Residence of record is not place of work. The District of Columbia is the clearest case. Its density is inflated by lawyers admitted there who live and often work elsewhere, and by a national federal workforce counted against a single city’s population. We flag the District as an outlier for exactly this reason, but the same distortion operates at smaller scale in any state whose bar admits many nonresidents.

State totals hide internal distribution. Every per capita ratio we report is a state average, and state averages can be dominated by a single metropolitan county. A high state ratio can coexist with rural counties that have no lawyers at all. The within-state variation is the part that matters most for access, and it is the part a state-level survey cannot show. We addressed it with county-level literature, but that literature is coarser and older than the annual state survey, so our rural findings are best read as direction and magnitude rather than as precise current counts.

Finally, the survey depends on voluntary and mandatory bars reporting consistently year over year. Definitional drift, late reporting, and changes in how a given bar classifies its members can introduce noise into the trend, particularly at the state level. We have described the decade trend in aggregate terms for that reason, and we have avoided treating small year-over-year state movements as meaningful.

Conclusion

The map of the American legal profession is not the map of the American population. Lawyers cluster where paying legal demand clusters, which means a small number of large coastal and metropolitan states, plus a federal city whose density belongs to a different logic entirely, hold most of the profession. The national total grew slowly across the decade and the national ratio held roughly flat, so the story is not one of a growing glut. It is one of a stable and unequal allocation.

The counties tell the harder truth. A large share of United States counties are thinly served, and a real set of them have no practicing lawyer at all, which turns the formal civil right to be heard into a practical problem of distance and cost. The reforms aimed at that gap are genuine and modest, and the market forces working against them are strong. We expect the geography to remain concentrated at the top and thin at the bottom until something larger than a recruitment bonus changes the economics of where a lawyer can afford to practice.

References

• American Bar Association, National Lawyer Population Survey. https://www.americanbar.org
• American Bar Association, Profile of the Legal Profession. https://www.americanbar.org
• U.S. Bureau of Labor Statistics, Occupational Employment and Wage Statistics (Lawyers, 23-1011). https://www.bls.gov
• U.S. Census Bureau, Population Estimates and 2020 Decennial Census. https://www.census.gov
• Administrative Office of the U.S. Courts, court and caseload data. https://www.uscourts.gov
• National Center for State Courts, Court Statistics Project. https://www.courtstatistics.org
• Gideon v. Wainwright, 372 U.S. 335 (1963). https://supreme.justia.com
• Turner v. Rogers, 564 U.S. 431 (2011). https://supreme.justia.com
• Lassiter v. Department of Social Services, 452 U.S. 18 (1981). https://supreme.justia.com
• South Dakota Unified Judicial System, Rural Attorney Recruitment Program. https://ujs.sd.gov
• State Bar of California, membership and admissions data. https://www.calbar.ca.gov
• New York State Unified Court System, attorney registration data. https://www.nycourts.gov