Rising U.S. Treasury yields have added fresh strain to the U.S. housing market. The average 30-year mortgage rate, which generally tracks the 10-year U.S. Treasury yield plus a mortgage spread, rose to 7.5% in October 2026, according to Mortgage Bankers Association data, worsening an affordability crunch that remains a top concern among U.S. households.
Last week, President Donald Trump signed an executive order intended to reduce housing costs and expand access to homeownership. Among other measures, the order directs the Federal Housing Finance Agency (FHFA) to cancel mortgage insurance premiums under certain conditions and establishes a task force aimed at lowering homebuilding costs.
While reducing costs is positive for borrowers, we believe several of the order’s changes could ultimately increase mortgage rates if applied too broadly. As we describe below, any changes should be targeted toward new purchases only, in our view, not refinancings. Otherwise, the market could price in more compensation for the increased risk of refinancing (or callability), potentially pushing mortgage rates higher. We believe policymakers can avoid this unintended consequence by limiting these changes to new mortgages, not existing ones.
Beyond the executive order, we believe there are several available tools that policymakers could use to lower mortgage rates in the near term. Doing so could deliver meaningful relief, as a 1-percentage-point drop in the mortgage rate lowers a buyer’s monthly payment by roughly as much as a 10% house-price cut (for more, see our September 2026 commentary, “The U.S. Housing Market Becomes a More Local Story.”)
We outline three policy options, all achievable under existing authorities, that we estimate could collectively lower mortgage rates by about 65-90 basis points (bps), or 0.65-0.9 percentage point.
1) Fannie Mae and Freddie Mac could actively buy more mortgage bonds
Congress created government-sponsored enterprises (GSEs) Fannie Mae and Freddie Mac with a mandate to improve housing affordability and reduce risks to capital markets. The GSEs do this largely by buying mortgages from lenders and packaging them into agency mortgage-backed securities (MBS), a form of mortgage bonds, which they either sell to investors or purchase to hold on their own balance sheets.
Increasing mortgage bond purchases is within the GSEs’ existing authority and aligns with their Congressional mandate – and, importantly, it is what President Trump directed the GSEs to do in January 2026. The goal of increasing GSE purchases is to boost demand for mortgage bonds, which typically raises bond prices and lowers yields, thereby helping to reduce mortgage rates (for more, see our April 2026 commentary, “How Policymakers Can Lower Mortgage Costs and Increase Housing Affordability.”)
Fannie and Freddie have been slow to implement President Trump’s January instruction to buy an additional $200 billion of mortgage bonds, with Freddie Mac reporting that it had sold about $1 billion as of August and Fannie reporting that it had only purchased about $40 billion. If Fannie and Freddie were to simply follow through on this directive, we believe mortgage rates could be substantially lower.
Nor is a $200 billion increase necessarily the limit. Fannie and Freddie could potentially expand their purchases even further, though doing so may require their regulator, the FHFA, to increase the GSEs’ existing $450 billion investment cap. Raising the cap would involve amending the Preferred Share Purchase Agreement (PSPA) from 2008, a step the FHFA and the U.S. Treasury have taken several times, most recently in 2025.
To put it in perspective, the GSEs held about $310 billion of mortgages as of August 2026, according to Haver Analytics data, far below the roughly $1.6 trillion they held before 2008, despite the mortgage market doubling in size in that time. While there are good reasons not to return to pre-financial-crisis levels, Fannie and Freddie still have substantial firepower.
We estimate additional GSE mortgage-bond purchases could lower mortgage rates by 20–30 basis points, depending on the size of the program.
2) Regulators could cut fees for new home loans only, not refinancings
On 30 September, the FHFA released new mortgage pricing based on an alternative credit scoring approach, a change that could reduce mortgage rates for new homebuyers. We think this is a positive step, provided it applies only to new home loans and not to existing ones. Clarifying this distinction could be helpful for the market.
Measures that lower costs only for new-purchase loans may improve affordability without materially affecting mortgage investors. Extending those fee cuts to refinancings, however, would likely increase prepayment risk, prompting mortgage investors to demand more compensation and potentially pushing rates higher for all borrowers.
The clearest illustration of this dynamic is in the contrast between multifamily and single-family lending. Currently, JPMorgan data show that single-family mortgage borrowers pay a premium nearly 70 basis points, or 0.7 percentage point, higher than the premium paid by multifamily borrowers – mainly landlords – because multifamily loans are less refinanceable (or in mortgage-speak, less callable).
The broader lesson is that policies that reduce the callability of mortgages may lower mortgage rates for home purchases, in our view, while measures that make mortgages easier to refinance could potentially backfire and raise mortgage rates. We estimate that cutting fees for new borrowers only could lower mortgage rates by approximately 25-30 bps.
Similarly, some measures in President Trump’s latest executive order would apply the same changes to refinancing as to home purchases, which could widen mortgages spreads if fully implemented without adjustment. We continue to recommend a purchase only approach, which could lower mortgage rates for buyers without disrupting the mortgage market or putting upward pressure on rates and spreads.
3) The Federal Reserve could stop selling mortgage-backed securities
Another potential action is for the Fed to stop allowing its mortgage bond portfolio to “roll off” its balance sheet, which in practical terms is akin to selling mortgages and buying Treasury bonds. If the Fed simply stopped this (i.e., retained its mortgage portfolio at current levels), it could potentially bring down mortgage rates for homeowners – by about 20-30 bps, in our view. (For more, see our September 2025 PIMCO Perspectives, “A Fed Housing Fix That’s Hiding in Plain Sight.”) Ultimately, the Fed makes decisions about the size and composition of its balance sheet in accordance with its dual mandate of price stability and maximum employment.
Conclusion
For many households, the most immediate obstacle to homeownership is the cost of financing a home today. While policymakers have devoted considerable effort to expanding U.S. housing supply – a longer-term objective – policy choices that affect mortgage rates themselves could provide a more near-term path toward affordability.
The debate over mortgage affordability also sits at the center of renewed discussions about the future of the GSEs. In comments last week, Treasury Secretary Scott Bessent indicated that the U.S. may not look to remove the GSEs from their longstanding government conservatorship until 2027 or 2028 and only if it could be done without pushing mortgage rates higher (for more, see our May 2025 commentary, “The Future of the GSEs: Do No Harm.”)
For investors, the flip side of mortgage rates being high is that mortgage bond yields are also high. PIMCO continues to see value in mortgage-backed securities at today’s historically wide spreads and high all in yields. If policymakers succeed at lowering mortgage rates, investors in agency mortgages could potentially benefit from spread compression and price appreciation. If policymakers take no action, investors may continue to compound returns at today's attractive yields.
