Mortgage rates appear to move on their own, and the most common explanation — "the Fed raised rates" — is wrong often enough to be actively misleading. Rates on 30-year mortgages regularly fall on days the Fed hikes, and rise on days it does nothing.
The actual chain has four links, and each one explains a different part of what you see.
Layer one: the bond market, not the Fed
The Federal Reserve sets the federal funds rate, which governs overnight lending between banks. It directly influences short-term rates — credit cards, home equity lines, adjustable-rate loans after adjustment — and only indirectly influences a 30-year fixed mortgage.
Long-term mortgage rates track long-term bond yields, particularly the 10-year Treasury. That yield reflects what investors collectively expect about inflation and growth over the coming decade.
This is why rates can move opposite to Fed decisions. Bond markets price expectations in advance; by the time a hike is announced, it is usually already in the price. What moves yields on the day is whether the accompanying commentary was more or less hawkish than expected.
Why the 10-year, for a 30-year loan
It seems arbitrary until you notice that 30-year mortgages rarely last 30 years. People sell, refinance, or pay off early, and historically the average life of a mortgage has been closer to a decade.
Investors buying mortgage-backed securities therefore price them against a benchmark with similar effective duration, which is the 10-year Treasury rather than the 30-year bond.
Layer two: the spread
Mortgage rates sit above Treasury yields because a mortgage carries risks a Treasury does not. The gap is the spread.
The largest component is prepayment risk. If rates fall, borrowers refinance and the investor gets their money back early, precisely when reinvesting it means accepting a lower yield. If rates rise, borrowers keep their cheap loans and the investor is stuck holding a below-market asset. That asymmetry has a price.
There is also credit risk on loans not fully guaranteed, and general liquidity conditions in the mortgage-backed securities market.
Historically the spread has often been near 1.7 percentage points, but it is not a constant. It widens sharply during financial stress and when the Federal Reserve steps back from buying mortgage-backed securities. A widening spread can push mortgage rates up even while Treasury yields are falling — which looks inexplicable if you are only watching one of the two.
Layer three: your own pricing adjustments
The rate you are quoted is the market rate adjusted for the specific risk you present. These are loan-level pricing adjustments, published by the loan purchasers and applied mechanically.
Credit score is the biggest lever, with material pricing differences across score bands. Loan-to-value ratio matters, and interacts with the score. Property type carries adjustments — a condo or a multi-unit property prices worse than a single-family home. Occupancy matters a great deal: an investment property carries a substantial add-on relative to a primary residence. Cash-out refinances price above rate-and-term refinances.
Two people can walk into the same lender on the same morning and receive quotes more than a full point apart, entirely from these adjustments.
Layer four: the lender itself
Finally, each lender adds its own margin, and that margin varies with how much volume they want, their cost structure, and their pipeline capacity. A lender who is overwhelmed will quote higher to slow incoming applications.
This is the layer that makes shopping worthwhile. The first three layers are largely outside your control on any given day; this one is a genuine spread between lenders on identical borrowers, and it is why several quotes on the same day frequently differ meaningfully.
What this means for timing
Rates move daily and can move meaningfully within a single day when economic data is released. A rate lock fixes your rate for a defined period, typically 30 to 60 days, and that is the tool for managing the risk between application and closing.
Trying to time a bottom is not a strategy anyone executes reliably, because the inputs are inflation expectations and bond market sentiment — the two things professional forecasters are worst at predicting.
The controllable levers are the ones in layers three and four: improve the credit score before applying, increase the down payment if it crosses an LTV pricing band, and collect quotes from several lenders on the same day so you are comparing like with like. Those routinely produce more improvement than waiting for the market to cooperate.