Wall Street banks ask elite law firms to cut hourly billings — arguing that AI is speeding legal work

Goldman Sachs, Morgan Stanley, and Citigroup are asking Big Law firms to cut hourly fees, arguing AI is reducing labor for research, document review, and contract analysis. Citi says if AI cuts hours per matter, costs should fall “significantly per transaction,” and it has started requiring proof of AI savings; Morgan Stanley plans to shift more outside work to competitive bids and use fixed fees rather than hourly billing. While AI-driven efficiency could compress Big Law’s billable-hour economics (associate rates up 33% since 2023 to ~$798, per Persuit), the impact is likely most concentrated in pricing negotiations rather than immediate market-wide financial effects.

Analysis

The market implication is less about near-term earnings than about bargaining power. These banks are using AI as a procurement lever to rebase third-party pricing, which can extend well beyond legal bills into accounting, advisory, litigation support, and eventually broader corporate services spend. For GS, MS, and C, any direct P&L benefit is likely immaterial this quarter, but the strategic value is in forcing vendors to share productivity gains rather than keep them as margin.

The second-order effect is a compression of the “human-hours = revenue” model in high-end services. That does not break elite firms overnight, but it does threaten the mix of work that historically subsidized partner economics: commoditized diligence, document review, and routine drafting. If banks succeed, other concentrated buyers will copy the playbook, which could gradually lower pricing power across professional services even if billable rates stay headline-stable.

Consensus may be overrating the disruption speed and underrating the buyer concentration. A handful of large banks can push through fixed-fee or outcome-based pricing faster than the rest of the market, so the first real evidence will show up in negotiated fee structures before it shows up in industry-wide margin data. The key falsifier is simple: if bank legal spend per transaction does not decline over the next 1-2 reporting cycles, this is mostly narrative rather than economics.

Key Decisions for Investors

• No immediate directional trade: the EPS impact for C/GS/MS is too small to underwrite a standalone position; wait for evidence in Q3/Q4 expense disclosures that legal spend per transaction is actually falling.
• Tactical buy-the-dip in GS and MS only if they underperform XLF by >1% on this headline; target a 1-3 month relative rebound as the market reads this as SG&A discipline rather than a growth issue. Stop if management commentary shows no follow-through on fixed-fee adoption.
• Stay neutral on C despite the strongest public cost-control posture; Citi is more likely to gain optics than hard dollars in the next quarter, so treat any move as sentiment-driven and fade if the stock rallies solely on this story.
• Watch for spillover into other vendor categories: if the banks formalize AI-based benchmarking in legal, expect consulting and accounting procurement to follow within 6-12 months. That is the higher-conviction structural short thesis for service providers, but it is not yet actionable from the current ticker set.