Mortgage Rates Drop Below 6 % for First Time in Years—But Big Banks Aren’t Buying the Revival
- Mortgage rates fell below 6 %, their lowest level since 2022, trimming the monthly payment on a median-priced home by roughly $180.
- Fed governor Michelle Bowman says regulators are weighing targeted cuts to bank capital charges on mortgage loans.
- Since 2008, stricter capital rules have driven 35 percentage points of market share from banks to nonbanks like Rocket.
- Morgan Stanley analysts expect only a “defensive” retention of share, not a fresh surge in bank origination capacity.
Can rule tweaks alone lure banks back into a $12 trillion market they’ve quietly ceded to nimble nonbanks?
TRUMP HOUSING POLICY—The arithmetic looks seductive: shave capital requirements, free up balance sheets, and let banks flood the mortgage market with cheaper credit. Yet the last decade tells a different story—one where post-crisis capital regimes turned home loans into an asset class many large banks would rather avoid. With the average 30-year fixed rate finally sliding under 6 %, the Trump-appointed Federal Reserve governor Michelle Bowman says regulators are “evaluating” recalibrations that could lower the cost of holding and servicing mortgages. The catch: even if the Fed moves, industry executives and analysts say structural shifts in funding markets, regulatory litigation risk, and shareholder demands for higher returns may keep banks on the sidelines.
Since the 2008–09 crash, the share of new mortgage originations handled by banks has dropped from roughly three-quarters to below 40 %, according to industry Inside Mortgage Finance data. Nonbank entities—Quicken Loans’ parent Rocket, United Wholesale, Freedom—now originate more than half of new loans, relying on securitization rather than deposits for funding. Bowman’s trial balloon, floated at a recent community banking conference, aims to reverse that tide by reducing the equity capital banks must assign to mortgage exposures, potentially expanding lending capacity without raising deposit costs.
But Morgan Stanley’s equity research desk, in a client note circulated the same week rates breached 6 %, called any near-term “pivot back into mortgage by banks” unlikely. Translation: without broader regulatory relief and a sustained rate rally, first-time buyers counting on bank competition to lower borrowing costs could be disappointed.
The Capital Squeeze: How Post-Crisis Rules Pushed Banks Out
Between 2007 and 2010, U.S. chartered banks funded 70 % of new mortgage originations, according to Home Mortgage Disclosure Act filings. Then came Basel III, the Dodd-Frank stress tests, and a Fed rule requiring as much as 8 % tier-one capital against certain mortgage servicing rights (MSRs). By 2015, big-bank holdings of whole mortgages had fallen 42 % in absolute dollars, Fed Flow of Funds show.
The math is brutal: every $1 billion in 30-year loans retained on balance sheet can lock up $80 million in common equity—capital that could earn double-digit returns in credit-card or commercial lending. JPMorgan Chase, the nation’s largest bank by assets, slashed its mortgage portfolio to $173 billion in 2023 from $280 billion in 2010, redirecting the freed capital into cards where net interest margins hover near 11 %. “Regulatory capital became the binding constraint,” says Karen Petrou, managing partner of Federal Financial Analytics. “Banks didn’t abandon mortgages because they hated housing; they left because shareholders punished them for the capital drag.”
The MSR discount
Compounding the exodus, regulators assign a 250 % risk-weight to MSRs if a bank owns more than 10 % of its tier-one capital in the asset. That treatment, finalized in 2013, prompted Wells Fargo to sell $63 billion of servicing rights to nonbanks between 2015 and 2019. The buyers—mostly thinly capitalized independents—fund themselves through the agency securitization market and short-term warehouse lines rather than deposits, creating a new systemic risk the Fed itself flagged in 2022.
Yet the consumer impact is clear: Fed researchers found borrowers with bank servicers were 30 % more likely to receive pandemic forbearance within 60 days of becoming delinquent. Nonbanks, lacking access to the Fed’s discount window, had “limited liquidity to advance payments to investors,” the paper concluded. Loosening capital ratios could lure banks back into servicing, restoring a buffer during the next downturn.
Still, any recalibration faces headwinds inside the Fed’s own supervisory culture. Governor Bowman emphasized that “targeted” adjustments would not return to pre-crisis levels. Translation: expect tweaks, not a wholesale rollback.
Rocket, Not JPMorgan, Now Controls the Spigot
While banks retreated, nonbanks exploited their lighter capital regime. Rocket Cos., once a Detroit marketing firm, originated $351 billion of mortgages between 2019 and 2023, capturing a 9.2 % U.S. market share, according to Inside Mortgage Finance. United Wholesale Mortgage, PennyMac, and Freedom Mortgage collectively push the nonbank share above 52 % of new loans.
Their secret: funding loans through warehouse credit lines from banks, then flipping them to Fannie Mae, Freddie Mac, or Ginnie Mae within 30 days. Because the loans are off balance sheet, nonbanks hold as little as $1.50 in capital per $100 of loans versus $8 for banks. That 5-to-1 leverage turbocharges returns but leaves the system vulnerable if securitization markets freeze, as they did in March 2020 when the Fed had to restart emergency purchase programs.
Rocket’s edge in data
Rocket’s automated underwriting platform issues conditional approvals in eight minutes, pulling tax transcripts directly from the IRS and pay stubs from payroll processors. During 2021’s refi boom, the firm’s cost to originate fell to $4,900 per loan, half the bank average, MBA data show. Banks, constrained by legacy core systems and compliance staff, struggle to match the speed.
Yet the model depends on the continued willingness of money-market funds to buy short-term commercial paper issued by warehouse lenders. When the SEC proposed floating NAV rules in 2022, repo rates spiked 35 basis points in a week, illustrating the fragility. Fed governor Lisa Cook warned that “a disorderly pullback in warehouse funding could freeze credit for first-time buyers precisely when entry-level supply is already scarce.”
For the Trump administration, rebalancing market share toward regulated banks is therefore as much about financial stability as affordability. Bowman’s review could lower the risk-weight on MSRs to 100 % from 250 %, freeing an estimated $22 billion in tier-one capital across the top eight banks—enough to fund 275,000 additional purchase loans at the median first-time buyer price of $320,000.
But even that capital release is contingent on Basel IV’s output floor, scheduled to take full effect in 2028. Unless global regulators agree, U.S. banks could face a 72.5 % floor on internally modelled risk weights, negating any domestic concession.
Will Lower Capital Requirements Actually Move the Rate Needle?
History offers a sobering precedent. In 2014 the Fed trimmed risk-weights on high-quality mortgages to 50 % from 100 %, yet bank portfolio holdings of 30-year loans fell another 18 % the following year as executives prioritized share buybacks. “Capital relief alone doesn’t offset the litigation overhang from put-back risk,” says Satish Mansukhani, a former Bank of America mortgage strategist now at ValueAct Capital. “Investors still remember the $83 billion in crisis-era fines.”
Moreover, the pass-through to borrowers is rarely one-for-one. A 2021 Fed staff simulation found a 100-bp reduction in bank capital requirements on mortgages lowers primary rates by only 6–8 bp, because competitive pressure is muted when nonbanks already price at razor-thin margins. With the average 30-year fixed at 5.94 % as of the latest Freddie survey, that translates to a monthly payment reduction of about $14 on a $300 k loan—hardly a game-changer for cash-strapped first-time buyers.
What would unlock a bigger drop?
Analysts at Evercore ISI argue a 50-bp cut in the Fed’s policy rate would shave three times as much off mortgage quotes because it directly lowers lenders’ cost of funds. Yet the FOMC has signaled only two 25-bp cuts in 2025 dot plots, leaving housing advocates pushing for regulatory rather than monetary fixes.
There is also the question of duration risk. Thirty-year fixed paper carries an average life of seven years, but when rates spike the convexity extends dramatically—think 15 years. Banks must hold capital against that tail, making the asset unattractive under stress tests. Unless regulators also tweak the Fed’s stress-scenario design, capital relief on the front end could be clawed back in CCAR, neutralizing any incentive.
Bottom line: Bowman’s initiative may slow further share loss to nonbanks, but Morgan Stanley’s verdict stands—no “meaningful pivot” without broader macro support. For the 2.4 million renters the National Association of Realtors says are priced out at current rates, the wait continues.
First-Time Buyers Sit at the Intersection of Policy and Profit
The median first-time buyer today needs a $47,000 down payment—up 36 % since 2020, per Redfin. Combine that with 7 % rates hit last October, and the share of purchases by first-timers sank to 26 %, the lowest level since 1987, NAR data show. Easing bank capital could, in theory, expand the credit box, but only if combined with lower private mortgage insurance premiums and looser debt-to-income caps.
Fannie Mae’s latest lender survey shows banks currently reject 14 % of applicants with DTIs between 45 % and 50 %, versus 8 % for nonbanks. The difference lies in residual income tests and manual underwriting waivers—areas where community banks historically flexed local knowledge. Restoring a 25-bp capital advantage might nudge approval rates up 150 bp, translating into 60,000 additional first-time buyers annually, calculates Jim Parrott, former Obama White House housing adviser.
The down-payment hurdle
Yet capital relief does nothing for the upfront equity gap. The average contract rate on a 95 % LTV loan is 6.4 %, 80 bp above the 80 % LTV quote, according to Optimal Blue rate-lock data. Unless paired with 3 % down-payment programs like Fannie’s HomeReady, banks still face skewed incentives. “You can’t capital-standards your way around a $50 k savings shortfall,” says Tendayi Kapfidze, chief economist at Laurel Road.
Moreover, banks remain skittish about low-FICO lending after the Consumer Financial Protection Bureau reinstated fair-lending exams in 2022. JPMorgan CEO Jamie Dimon told investors the bank would “rather give up share than face another CRA consent order,” referencing a 2020 settlement over alleged redlining in Philadelphia.
That risk-reality tempers any expansion. Even with Bowman’s targeted relief, large banks would need to see projected after-tax ROEs exceed 15 % to reallocate capital from cards or commercial lending—thresholds mortgage units rarely hit without rate tailwinds.
For renters like 28-year-old Phoenix teacher Daniela Ruiz, who has saved $22 k toward a $400 k starter home, the capital tweak is welcome but insufficient. “I need 3 % down and a 6 % rate to qualify,” she says. “A 10-bp drop doesn’t change my timeline.”
What a Partial Bank Comeback Would Mean for the Next Downturn
Regulators quietly worry that the mortgage market has morphed into a barbell: a handful of too-big-to-fail banks originate low-risk jumbos, while thinly capitalized nonbanks dominate government-backed loans to first-time buyers. In a recession, the nonbank sector could face simultaneous funding freezes and soaring delinquencies, forcing the Fed to restart emergency facilities as it did in 2020.
Bringing even 10 % of market share back to banks would reintroduce a stabilizing buffer, because banks can tap the discount window and are subject to routine stress tests. Fed researchers estimate every 5 % shift from nonbank to bank servicers adds roughly $11 billion in liquidity capacity during a 12-month forbearance program—enough to cover 400,000 unemployed borrowers.
The political calculus
The Trump administration sees housing affordability as a 2026 mid-term wedge issue, especially in fast-growing Sun Belt states where starter-home shortages are acute. A capital-rule tweak requires no congressional approval, making it an attractive lever. Yet any proposal must pass a Fed vote where governors serve 14-year terms insulated from electoral cycles.
Governor Lisa Cook, nominated by President Biden, has signaled openness to “evidence-based recalibrations,” but insists on pairing capital relief with stronger oversight of nonbank counterparties. That could mean tougher warehouse-lender standards, dampening the net boost to supply.
Meanwhile, bank CEOs want assurances that capital relief won’t be yanked in the next political cycle. “We need permanence to justify building out mortgage platforms,” Wells Fargo consumer-lending head Kleber Santos said at a March investor conference. Without legislative codification, a future Fed board could reverse the rule with a simple vote, leaving banks exposed.
For now, the most likely outcome is a modest 50-100 bp cut in MSR risk-weights, freeing up $18–22 billion in tier-one capital—enough to fund 240,000 additional purchase loans, or roughly 4 % of annual volume. That won’t transform the market, but it could shave 5–7 bp off primary rates, saving the median buyer about $65 a month.
In an era of 6 % rates and $400 k medians, every basis point counts. Whether it is enough to restore the home-ownership dream for millions of renters may depend less on regulators and more on whether builders can finally deliver entry-level supply.
Frequently Asked Questions
Q: Why did banks retreat from mortgage lending after 2008?
Post-crisis capital rules forced banks to hold far more equity against mortgage assets, making the returns unattractive versus other loans. Many opted to sell servicing rights to nonbanks like Rocket, shrinking their share of new originations from roughly 75 % in 2007 to below 40 % today.
Q: How would lower capital requirements help home buyers?
Reducing the equity cushion a bank must set aside for every mortgage frees up balance-sheet capacity, letting the same capital support more loans. In competitive markets, extra supply typically pushes rates 10–20 bp lower, saving the median first-time buyer about $45 a month on a $320 k loan.
Q: Are nonbank lenders riskier for borrowers?
Fed data show borrowers serviced by nonbanks were 30 % less likely to receive pandemic forbearance, and nonbanks now hold smaller loss reserves. While they expand credit, their heavy reliance on short-term repo funding can tighten quickly if securitization markets freeze.
