
Lawyer Pay
Think new BigLaw associates are all making $225K? Think again, new NALP report says
Despite reports of $225,000 starting salaries for BigLaw associates, most law firms are paying less, according to a report released Wednesday by the National Association for Law Placement. (Image from Shutterstock)
Despite reports of $225,000 starting salaries for BigLaw associates, most law firms are paying less, according to a report released Wednesday by the National Association for Law Placement.
As of Jan. 1, the median base salary for first-year associates was $200,000 in all firms surveyed and $215,000 in firms with more than 700 lawyers, according to a May 28 press release summarizing the 2025 U.S. Associate Salary Survey report.
The pay figures are based on responses from 437 firms, 87% of which had more than 250 lawyers.
Median salaries have remained flat since 2023, an indication that the market has softened, said Nikia L. Gray, executive director of the NALP, in the press release. She attributed the lack of movement to waning “talent wars,” which means that there is less pressure for firms to raise salaries to remain competitive.
Only 32% of reporting law offices paid $225,000 to first-year hires, although the percentage paying that figure was higher—45%—among firms of more than 700 lawyers.
“Still, $225,000 emerged as the most frequently reported salary figure, suggesting a potential trend,” according to the press release.
In six of 19 markets analyzed in the survey, the median is already $225,000. Those markets are Austin, Texas; Boston; Houston; New York City; San Francisco; and the Washington, D.C., area.
Median salaries for all associates are $200,000 for first-year associates, $210,000 for second-year associates, $220,625 for third-year associates, $238,125 for fourth-year associates, $255,000 for fifth-year associates, $271,175 for sixth-year associates, $286,700 for seventh-year associates and $330,000 for eighth-year associates.
Write a letter to the editor, share a story tip or update, or report an error.