September 21, 2026

SusanIsaacs

The 5.9% Mortgage Rate Pipedream

, , , , , , , , , , 

Mortgage rate forecasting fails because rates depend on unpredictable global bond market swings and investor fear rather than moving in lockstep with the Federal Reserve.

For months, I’ve been telling buyers that building a real estate decision around the expectation of a mortgage rate below six percent was a risky bet. Not because I was certain about where rates were going. Nobody is. But forecasts aren’t promises, and too many housing decisions were being made as though 5.9% had already been penciled onto the calendar.

There was good reason for buyers to have that number in their heads. In September 2025, Fannie Mae forecast that the average 30-year fixed mortgage rate would fall to 5.9% in the fourth quarter of 2026. That wasn’t some internet mortgage guru making a prediction on TikTok. It came from Fannie Mae’s Economic and Strategic Research Group.

A year later, 5.9% is looking considerably more like a pipedream. Fannie Mae’s September 2026 forecast now puts the average 30-year rate at 6.8% in the fourth quarter, followed by 6.7% throughout 2027. Mortgage News Daily’s daily index stood at 7.20% on September 18. Those aren’t directly comparable measurements, since one is a quarterly forecast and the other is a daily rate index based on lender rate sheets, but they illustrate just how far the market has moved from the sub-6% expectations buyers heard a year ago.

We’ve Seen This 5.9% Before

This isn’t Fannie Mae’s first trip to 5.9%. In February 2024, Fannie forecast that the 30-year fixed mortgage rate would decline to 5.9% by the end of 2024. Just one month later, it raised that forecast to 6.4%. Stronger employment and inflation data had changed expectations about how aggressively the Federal Reserve would cut rates, which pushed longer-term interest rates higher. The economic assumptions changed, so the mortgage forecast changed with them.

Then 5.9% came back. In September 2025, Fannie forecast it again, this time as the average rate for the fourth quarter of 2026. By July 2026, that fourth-quarter forecast had risen to 6.4%. The September forecast now has it at 6.8%.

Here’s that progression in one place:

ForecastExpected 30-year mortgage rate
February 20245.9% by end of 2024
March 20246.4% by end of 2024
September 20255.9% average, Q4 2026
July 20266.4% average, Q4 2026
September 20266.8% average, Q4 2026
MND, Sept. 18, 20267.20% daily rate

This Isn’t About Bad Forecasting

It would be easy to look at that table and conclude that the forecasters don’t know what they’re doing. I don’t think that’s the useful conclusion.

Economic forecasts are built from assumptions about inflation, employment, economic growth, Federal Reserve policy, Treasury yields and a long list of variables that can change. Fannie says so itself. Its forecasts are based on assumptions, are subject to change without notice, and changes in the information underlying those assumptions can produce materially different results.

March 2024 gives us a nearly perfect example. Fannie didn’t wake up one morning and decide 5.9% had been a bad guess. Employment and inflation came in stronger than expected, financial markets reduced their expectations for aggressive Fed cuts, longer-term rates moved higher, and Fannie changed its mortgage forecast accordingly.

The same problem confronts anyone trying to forecast mortgage rates today. The economic sands keep shifting. Inflation changes. Employment changes. Markets revise their expectations for the Fed. Treasury yields respond, mortgage-backed securities respond, and lenders price mortgages accordingly. A mortgage forecast is a snapshot of where economists think those moving pieces are headed based on the information they have at the time.

Somewhere Along the Way, A Forecast Becomes A Promise

The economists may understand all of those qualifications. Consumers don’t necessarily hear them.

“Fannie Mae forecasts 5.9%” is easy to understand. The assumptions underneath the forecast aren’t nearly as catchy, and somewhere between the research department, the headline, the mortgage industry, the real estate industry and the consumer, a conditional forecast can begin sounding suspiciously like a scheduled event.

That can affect real decisions. A buyer waits another six months because rates are supposed to come down. A homeowner decides not to sell because the next mortgage should be cheaper next year. Someone stretches for a payment assuming refinancing will rescue the budget later. Another buyer renews a lease and waits for the promised land of 5-something rates.

None of those decisions is automatically wrong. The problem is making one of them because somebody’s economic model produced a particular mortgage rate for a particular quarter a year or two into the future.

The problem wasn’t the forecast. It was treating the forecast like an appointment.

What 90 Basis Points Actually Costs

The difference between Fannie’s September 2025 forecast of 5.9% and its current 6.8% fourth-quarter forecast is 0.9 percentage points, or 90 basis points. On a mortgage, that isn’t a rounding error.

Take an $800,000 home with 20% down, leaving a $640,000 mortgage. On a 30-year fixed-rate loan, principal and interest at 5.9% would be approximately $3,796 per month. At 6.8%, it would be about $4,172. That’s roughly $376 more every month, or about $4,512 a year.

At 7.20%, the Mortgage News Daily rate on September 18, principal and interest on the same loan would be roughly $4,345 per month. That’s about $549 more each month than the payment at 5.9%, or roughly $6,590 a year.

Those calculations are rounded and don’t include property taxes, homeowners insurance, mortgage insurance if applicable, condo or HOA fees, or closing costs. Hyperwrite independently checked the 5.9% and 6.8% amortization calculations and flagged the need to specify the 30-year term and distinguish 90 basis points from a “0.9% increase.” Both corrections are incorporated here.

And Now The Forecast Has Changed Again

The September shift didn’t occur in a vacuum. On September 16, the Federal Reserve raised its target range by 25 basis points to 3.75% to 4.00%. Mortgage rates don’t move mechanically with the federal funds rate, and the Mortgage Bankers Association noted that longer-term rates had already incorporated expectations for the Fed move.

MBA Chief Economist Mike Fratantoni said housing and mortgage activity had “slowed abruptly” as mortgage rates rose over the preceding several weeks. MBA expects two additional Fed increases over the next year and expects mortgage rates to remain near current levels over its forecast horizon. That doesn’t tell us exactly where mortgage rates will be next spring, next summer or at the end of 2027. If the last few years have demonstrated anything, it’s why we shouldn’t pretend that it does.

The Pre-Election Dip

Many of us thought there might be at least a slight drop in rates just before the midterms. We expected Trump to push for it, and he probably will. But Warsh has made it clear he’s not Trump’s puppet. Since the September FOMC meeting, that hope has dimmed considerably.

Trump has continued calling for substantially lower interest rates, while Warsh’s first rate increase as Fed chair went in exactly the opposite direction. The September vote was unanimous, raising the federal funds target range by a quarter point to 3.75% to 4.00%, and the Fed’s new projections showed 16 of 18 participants expecting at least one more increase before the end of 2026. The Fed was publicly grounding its decision in persistent inflation and the economic outlook.

Actually, the market started taking some of that hope away before the Fed ever met. As investors increasingly expected a September rate hike, mortgage markets had already begun pricing in that expectation. Mortgage News Daily’s average 30-year rate climbed from 7.12% on September 11 to 7.22% on September 15, the day the FOMC meeting began. MBA Chief Economist Mike Fratantoni later said longer-term rates, including mortgage rates, had already “baked in” expectations for the September hike and future increases.

Then Warsh gave markets something else to price. The September hike itself produced little immediate reaction because traders already expected it. During Warsh’s press conference, however, bond markets reacted to his description of persistent inflation and the prospect of further tightening. The question was no longer whether September’s expected hike would happen. It was how many more might follow.

That showed up quickly in fed funds futures. Instead of looking toward the Fed’s final meeting before the midterms for a little relief, traders began assigning meaningful odds to another quarter-point hike on October 28. The pre-election dip hasn’t disappeared as a possibility, but the market has moved a long way from the scenario many buyers were hoping for.

So What Is A Buyer Supposed To Do With A Forecast?

Use it for what it is. A forecast can tell us what professional economists think is plausible under a particular set of assumptions. It can help explain where risks are building and how expectations are changing. What it can’t do is tell an individual buyer that a 5.9% mortgage will be waiting on a particular date.

That distinction becomes especially important when the housing decision itself may depend on the forecast. Waiting can work. Rates could fall. Prices could fall. Inventory could improve. A buyer’s financial position could improve, too. But the opposite can happen, and several of those things can move in different directions at the same time.

The more useful question isn’t, “When will mortgage rates get back to 5.9%?” It’s whether a purchase works at the price, financing terms and market conditions actually available when the buyer is ready to make it. If rates eventually fall enough to make refinancing worthwhile, that’s an opportunity. It shouldn’t be the assumption required to make the original purchase affordable.

Fannie Mae’s economists will keep forecasting mortgage rates because that’s part of what economic forecasting is supposed to do. Those forecasts will change when the world changes, just as they have before. Buyers can follow activity without building their lives around forecasts.

Sources

Latest articles

Leave a Comment