The Bank of England held Bank Rate at 3.75% on 30 July 2026. The more useful number in that announcement was not the rate. It was the vote: three of the nine members of the Monetary Policy Committee wanted an increase to 4%, and they were outvoted six to three. The Committee does not meet again until 17 September. For anyone with a fix expiring this autumn, the decision that matters is not the one the Bank has already made. It is what you do in the seven weeks before it meets again.
We covered why mortgage pricing rose even as Bank Rate stayed still when the July decision landed. This is the forward half of that question.
The vote is the signal
According to the Bank’s own July 2026 Monetary Policy Summary and Minutes, published on 30 July, the Committee voted by a majority of six to three to maintain Bank Rate at 3.75%. Megan Greene, Catherine L Mann and Huw Pill each preferred an increase of 0.25 percentage points, to 4%.
One meeting in isolation tells you very little. The sequence tells you more. Across the Bank’s published minutes for 2026, the hawkish bloc has grown in a straight line since the spring:
- February: a 5 to 4 majority, with the four dissenters wanting a cut to 3.5%
- March: unanimous, no dissent at all
- April: 8 to 1, with one member voting to raise to 4%
- June: 7 to 2, both dissenters voting to raise
- July: 6 to 3, all three dissenters voting to raise
Five months ago the argument inside the Committee was about how fast to cut. It is now about whether to raise. Nobody has voted for a cut since February.
That is not a forecast, and it should not be read as one. Six members still voted to hold in July, and a split can narrow as easily as it can widen. But it does tell you which direction the risk sits in, and the dissent count has not moved backwards at any meeting since February.
The practical reading for an investor is narrow and worth stating plainly. Pricing your next purchase on the assumption that the next move is downwards is now an assumption that three of nine rate setters actively disagree with, in a year where the number who disagree has risen at every meeting since March.
Why fixed rates moved first
Fixed mortgage rates are not set by Bank Rate. They are set by swap rates, which reflect where the market expects Bank Rate to sit over the life of the fix rather than where it sits today. That is why lenders can reprice upwards in a month when the headline rate does not move at all, and it is why a hold accompanied by three hawkish dissents is not the good news it looks like on a news alert.
It also means the September decision is already partly in the price. Waiting for the Bank to confirm a direction is waiting for information the swap market has already acted on.
The gap that actually decides a deal is not the one between this month’s average rate and last month’s. It is the one between the market average and what a well capitalised buyer can borrow at. A single percentage point of difference on a £400,000 loan is about £4,000 a year on an interest-only basis, for as long as the rate gap holds, which on most buy to let purchases is the difference between a deal that works and one that does not.
Locking a rate six months early
The mechanic that most buyers do not use is the offer window. Many lenders will let you reserve a fixed rate up to six months before you need it. If rates fall before completion, you can usually switch to the cheaper product. If they rise, the reservation holds.
That is an asymmetric position, and asymmetry is the whole point. You are not predicting the September decision. You are removing your exposure to it while keeping the upside if the hawks lose the argument again.
It is not free. Reserving early can mean a product fee committed sooner, and lenders differ on whether and how you can move to a better rate later. The terms of the switch matter more than the headline rate, so confirm the product allows a downwards move before you rely on it.
What this means for property investors
Three things follow. First, if a fix expires between now and spring, start the conversation now rather than after 17 September, because the reservation window is the cheap option and it is only available in advance.
Second, stress test at 4% rather than 3.75%. Three rate setters have already voted for it. A deal that only works on the current rate is a deal that depends on an argument being won inside the Committee, and that is not an underwriting assumption, it is a hope.
Third, remember that the average rate is not your rate. Leverage and loan to value move the number more than the Bank does. The same asset bought at 60% loan to value and at 80% is two different investments, and in a market priced this tightly that difference decides whether a purchase clears its cost of capital. That arithmetic is the substance of whether buy to let works at all in 2026.
If you want stock that has been underwritten against a higher rate rather than the current one, you can see what we currently have available.