Chris Bennett is chairman of mortgage industry hedge advisory firm Vice Capital Markets, now in its third decade of service to mortgage lenders across the country.
Mortgage industry consolidation ramped up significantly in 2025. HousingWire tracked 62 mergers, acquisitions, exits, investments and joint ventures across originators, servicers, technology platforms and title, appraisal and valuation firms last year, up from 37 in 2024. Several of the largest deals reshaped who holds servicing, including Rocket Companies’ $14.2 billion acquisition of Mr. Cooper and Bayview Asset Management’s take-private of Guild Mortgage. In February 2026, Pennymac agreed to acquire Cenlar’s subservicing business, a pending deal that would add roughly $740 billion in unpaid principal balance and make Pennymac one of the largest subservicers in the country.
Servicing has also changed hands. MBA’s National Delinquency Survey shows independent mortgage banks (IMB) serviced 9% of loans by count in 2011 and 61% in 2025, with depositories falling from 91% to 39% over the same stretch. In agency servicing, the segment most lenders actually sell into, nonbanks held 73% of balances as of June 2026. That portfolio does different work for an IMB than it does for a bank because there is no branch network or checking account to keep the borrower close. The servicing book is the customer list, and marketing to it is how the next loan gets made. Consolidation moves those books toward the operators who work them hardest and changes the math on every loan a lender sells.
The recapture math belongs in best execution
Every mortgage sold into the secondary market carries a servicer with it, and that servicer’s ability to win the borrower’s next loan varies widely. ICE Mortgage Technology’s August 2026 Mortgage Monitor put the industry-wide servicer retention rate for refinances at 31% in the second quarter of 2026, down a percentage point from the first quarter and three from Q4. That average masks substantial variation, as nonbank servicers retained 37% of refinancing borrowers, compared with 14% for banks. The range among individual firms is wider still. In my experience, some of the largest aggregators that actively market to their portfolios recapture 60% to 70% or more of their borrowers, many large depositories sit in the mid-teens and some IMBs recapture only 2% to 3%.
Secondary desks already do a version of this math in another corner of their lives. Many lenders informally dock a slow-to-purchase or stipulation-heavy investor a few basis points before comparing bids. The same logic applies more directly to recapture. A buyer paying two basis points more on today’s bid because it expects to win the borrower back is pricing in a future revenue stream that may once have belonged to the seller. Converting that buyer’s recapture rate into basis points of long-run revenue at risk, then weighing it against the cover price, reorders the bids.
Consolidation makes the assumption riskier
Handicapping recapture used to be straightforward because the buyer taking the loan was the one that would still be servicing it years later. Consolidation removes that certainty. When a high-recapture aggregator acquires a mid-recapture servicer, borrowers sold years earlier under one set of assumptions are suddenly marketed by a sharper operation than the original sale contemplated. A lender can’t predict which servicer will be acquired next, but it can handicap the risk. The working assumption should be that consolidation continues and the likeliest targets are servicers with under-monetized portfolios a sharper operator could buy and start marketing to immediately. That assumption belongs in the execution decision alongside the price.
There are structures that address this head-on. Some buyers will pay less for the servicing asset in exchange for an agreement not to market to the borrower and, in some cases, to refer the borrower’s inbound inquiries back to the originator. The trade is upfront dollars for downstream relationship protection. It doesn’t suit every shop, but for a lender that would rather not service in-house or hand over its borrower to a high-recapture operation, it is worth pricing out.
Non-solicitation agreements, which bar a servicing buyer from marketing to the transferred borrower, are the contractual backbone of that arrangement and a standard feature of many MSR sale contracts. Their enforcement is now being tested in court. Rocket Companies sued United Wholesale Mortgage (UWM) in May 2026 over three bulk mortgage servicing rights (MSR) pools covering roughly 182,000 loans and $65 billion in unpaid principal balance. UWM had sold the servicing on those loans to Mr. Cooper in 2024. Rocket, which closed its acquisition of Mr. Cooper in October 2025, alleges UWM then used refinance incentives and broker outreach to solicit the same borrowers, and is seeking nearly $100 million. UWM disputes the allegations. The case is pending, but whatever the outcome, a non-solicitation clause offers protection only as far as someone is willing to enforce it. A lender that trades price for one should watch prepayment speeds on the pools it sells because unusually fast payoffs are what alerted Rocket to the alleged breach.
The same logic applies to primary market pricing
The MBA’s first-quarter 2026 Quarterly Mortgage Bankers Performance Report shows independent mortgage banks earned a pretax production profit of $727 per loan, on average revenue of $12,626, against all-in production costs of $11,898. That 16-basis-point margin is less than half the industry’s long-term average of 39 basis points, which dates to 2008.
A common practice is to pull a stack of competitor rate sheets, apply a single, fixed basis point margin across the book and hand the loan officer whichever price wins. A flat margin treats every loan the same, regardless of the underlying economics. A $200,000 loan priced at a 250-basis-point margin generates $5,000 in revenue against an average manufacturing cost of $11,898, a shortage of nearly $7,000. A $800,000 loan at the same margin generates $20,000 in revenue, more than covering costs, but a competitor willing to sharpen its pricing on larger balances will win that deal instead. The engine ends up filling the pipeline with loans that can’t cover their own costs while ceding the loans that could.
Basis points drive rate sheets, loan sales and commissions, so that is the unit most shops use to measure margin. What shows up on the profit and loss statement are dollars. Pricing by loan size, segment and borrower economics protects the margin that a flat basis-point rule gives away.
In the two leanest years on record, the clearest divider I saw between the lenders that hit their return-on-equity targets and the ones that scraped by wasn’t volume. It was whether they priced deliberately to the primary market or simply mirrored the competition. A shop that outsources that policy decision to a stack of competitor rate sheets has handed away its margin along with it.
Primary market pricing and secondary market execution are usually managed by different desks and reviewed on different timelines. Both are bets on what a loan is worth over the years. A lender that prices thoughtfully to the borrower but sells the resulting relationship to whoever bids two basis points higher today is protecting one side of its business while giving away the other. Managing both decisions on the same long horizon is what keeps your margins healthy and strong in a market this thin.
(Views expressed in this article do not necessarily reflect policies of the Mortgage Bankers Association, nor do they connote an MBA endorsement of a specific company, product or service. MBA NewsLink welcomes submissions from member firms. Inquiries can be sent to Editor Michael Tucker or Editorial Manager Anneliese Mahoney.)
