Law Firm Associate Attrition 2026: Rates by Firm Type and Class Year

The NALP Foundation reported on 21 April 2026 that the calendar-2025 rate printed at 19 percent overall across 141 U.S. and Canadian participating firms. Firms of 100 or fewer lawyers printed 24 percent for a second consecutive year. Every larger size band in that study sat between 16 and 18 percent. The overall figure was one point below the Foundation’s 24 April 2025 print for calendar 2024 (20 percent, itself up from 18 percent in 2023) and well below the series high of 26 percent in 2021.

Participating firms in 2025 reported 6,335 associate hires and 4,442 associate departures, a net gain in associate headcount. Monthly departure levels fluctuated between 6 percent and 10 percent of the year’s exits; 52 percent of departures fell in the first half of 2025 and 48 percent in the second. None of that is a culture score. It is a firm-reported associate rate at a named sample, and it is the number a chief legal officer can put next to a replacement requisition.

Firm type is the first cut because the buyer is not replacing “an associate.” A head of legal at a company that hires from a 100-lawyer shop is buying out of a 24 percent cell. A general counsel who hires from a 500-lawyer platform is buying out of a 16-to-18 percent cell. The replacement cost sits in different markets even when the class year is the same. Sartori’s New York associate mandate telemetry, 33 closed searches over three years, is almost entirely the second cell — mid-size and large-firm laterals, not the sub-100 band.

The 2024 Foundation release (24 April 2025) covered 119 participating firms; 17 percent of respondents had 100 or fewer lawyers and 37 percent had more than 500. The 2025 sample is larger (141 firms). Neither print is a census of every U.S. firm.

Among 2024 departing associates, the next professional position was another law-firm associate seat in 41 percent of cases, up from 31 percent in 2023, and corporate in-house counsel in 18 percent, up from 15 percent, the NALP Foundation reported on 24 April 2025. The buyer who pays the replacement cost is therefore still more often another firm. When the buyer is a company, the seat is a corporate legal department reporting to a general counsel or chief legal officer, not a practice group.

The in-house gate is a class-year analogue. A Google Careers specification for Associate Safety Counsel, Devices and Services, as indexed in 2026, asked for a J.D., LL.B., or equivalent, admission or authorized practice in the seat’s state, and three years of attorney-level experience in government, in-house, or a law firm. That posting is a public role specification, not a current vacancy count. It is the shape a head of legal actually writes: the corporate desk that absorbs firm attrition is hiring the year-3-to-5 cell that the Foundation shows leaving fastest.

Across 1,675 structured interviews with New York partners and counsel, 58 percent of 196 general counsel and heads of legal over a 24-month window told Sartori they will not hire onto the in-house desk with fewer than three years of attorney experience. A general counsel at a New York-headquartered, PE-backed healthcare platform put it without romance: the in-house seat is a three-year attorney gate, not a newly-qualified conversion.

Sartori has run New York associate search for more than 10 years, for law-firm groups and for companies putting a lawyer on the payroll. Over the trailing three years we closed 33 associate searches on that line, at a 93 percent completion rate, on a typical 6-to-12-week timeline, with a median 11 working days from offer to acceptance. Counter-offer incidence on the line is 37 percent.

Of those 33 closed files, 21 were opened to replace a years-2-to-5 departure. Fourteen of the 21 asked for years 3 to 6. That matches the Foundation tenure cut, and it matches what in-house hiring managers in the same city cohort actually buy. It does not match a first-year requisition written as if the 97 percent 2L conversion rate were a company channel.

The uncomfortable number on our own book: 8 of 33 closed New York associate searches still required a second shortlist after the first-choice year-4 and year-5 laterals failed a conflicts screen or reversed on a competing lockstep match. Completion at 93 percent counts those files as closed. It does not count them as clean. The mid-level litigation and corporate cells are where we place slowly, even when the Foundation rate at large firms sits in the 16-to-18 percent band. A general counsel who opens a year-4 hire as a six-week project is writing the wrong clock.

Two anonymized composites from that book, described by category only. An Am Law 100 litigation group opened a year-4 replacement after an unwanted departure in the five-year window. The first-choice laterals failed a conflicts screen against a financial-services panel; the file closed in the twelfth week on the second shortlist, inside the envelope, not inside the plan. A PE-backed healthcare platform instructed its general counsel to hire a Corporate Counsel with a three-year attorney gate. The competing number was a $320,000 class-2022 lockstep base, not the $100,000 business-sector graduate median. Offer to acceptance ran 11 working days; a competing firm counter-offer sat on the table and was declined. Neither file is a named employer.

What a chief legal officer should do with the table is narrower than a retention program. Read firm type first (24 percent versus 16 to 18 percent). Read class year second (the five-year window, the three-year in-house gate). Price the lockstep the leaver currently earns. Decide whether the company is buying a years-3-to-5 attorney or whether the work still belongs on a firm panel. Then open in-house counsel recruiting or associate recruiting on that cell, not on a burnout narrative. The non-financial levers and the burnout brief are separate pages, on purpose.