Short Rates Vs Long Rates: Why Central Banks Don’t Set Mortgage Rates

If you follow bond markets for long enough, you’ll eventually come across the apparent paradox of central banks cutting rates, yet long-term bond yields barely budging, or even rising. Market commentators will say things like, ‘the market is refusing to let long-term yields come down’, and point out that it’s the long end that matters for mortgages and corporate borrowing. That can feel backwards. If the bank rate is falling, shouldn’t borrowing costs fall too?

It turns out that whilst the central bank has tight influence over the short end of the yield curve, it only has indirect influence over the long end. Short rates are mostly about what the central bank is doing right now and what markets think it will do over the next year or two. Long rates are about what investors think the whole economy will look like over the next decade, plus an extra layer of compensation for risk and uncertainty. When those long-run beliefs shift, long yields can ‘disconnect’ from the bank rate, and that disconnect is exactly why mortgages and long-term corporate debt can stay expensive even as the central bank cuts rates.

What People Mean by ‘Short Rates’

‘Short rates’ refers to interest rates on very short maturities. Think overnight rates (SONIA in the UK, SOFR in the US), and money-market rates out to a few months. In practice, many people extend ‘short’ to include government yields out to around two years, because that part of the curve is still dominated by expectations of near-term central bank policy.

This end of the curve is essentially a forecast of the next handful of policy meetings. If markets expect the central bank to cut over the next six to twelve months, short-dated yields usually decrease quickly, often before the first cut actually happens. Conversely, if inflation surprises on the upside and markets price in rate rises, short yields jump. That’s why the short end often looks like a live scoreboard of ‘what happens next’.

What People Mean by ‘Long Rates’

‘Long rates’ are yields on longer maturities such as 5, 10, or 30 years. These are set in markets where the buyer is taking on exposure to a long stream of future cashflows, and therefore to long-run inflation outcomes, long-run growth, and the uncertainty around both. Longer dated bonds also carry a lot of interest-rate risk (duration). That is, when rates move, prices move a lot.

This matters because many economically important borrowing rates are tied to intermediate and long maturities. Fixed-rate mortgages, long-dated corporate bonds, and many asset-backed markets reference longer-term government yields and swap rates (plus a spread). So whilst the bank rate might influence the cost of floating-rate borrowing and savings accounts fairly directly, fixed-rate borrowing is often anchored by the long end.

The Mental Model

A useful way to think about a long-term yield is:

Long Yield ≈ Average Expected Future Short Rates + Term Premium

This isn’t a perfect identity in every nuance, but it’s a powerful organising framework.

The first component, ‘average expected future short rates’, is exactly what it sounds like. A 10-year yield reflects, roughly, the market’s view of where short rates will sit over the next decade, averaged over that horizon. If markets think that the central bank will cut today but then return rates to a higher level later, the average over ten years might not change much. In that case, the 10-year yield does not fall much either.

The second component, the ‘term premium’, is the extra yield investors demand for holding long-duration bonds instead of rolling short-term instruments. You can think of it as compensation for uncertainty and risk over long horizons: inflation surprises, growth regime changes, fiscal uncertainty, and the simple fact that long-dated bond prices can be volatile when rate expectations shift.

Why Long Yields Can Stay High Even As the Central Bank Cuts

Once you separate long yields into those two pieces, the ‘paradox’ stops being a paradox.

First, markets may not believe the cuts will last. The central bank can cut a few times, but if investors think inflation will remain sticky, or that growth will rebound, they may expect the bank rate to stabilise at a relatively high level over the medium term. In other words, the near-term path changes, but the long-run average does not. Long yields then remain elevated because they are pricing the whole decade, not the next three meetings.

Second, the term premium can rise and offset everything else. Even if the average expected path of short rates is drifting down, the extra compensation demanded for holding longer-dated bonds can rise at the same time. This can happen when investors are more nervous about inflation volatility, when uncertainty rises, when bond-market volatility increases, or when the supply-demand balance in government bonds shifts against prices.

One sentence that captures this ideas is: the central bank may cut the bank rate because it thinks current macro-economic conditions warrant it, whilst markets simultaneously demand a higher term premium because they dislike long-horizon risks. Those two forces push in opposite directions, and the long end reflects the net result.

The Yield Curve: What Each Segment ‘Cares About’

It’s helpful to map this onto the yield curve.

The front end (roughly 0–2 years) is dominated by expected central bank policy. This is why short-dated yields often behave like a high-frequency macro forecast.

The middle of the curve (roughly 2–5 years) is a transition zone. It still reacts strongly to policy expectations, but it also starts to incorporate more uncertainty about the business cycle and inflation persistence.

The long end (5–30 years) increasingly reflects the market’s view of long-run inflation risk, long-run real growth, and the term premium. It is also the segment most exposed to shifts in bond supply, central bank balance sheet policy, and structural demand for duration.

This is why you can see rate cuts pulling down the front end whilst the long end barely moves. Indeed, markets may agree about the next few months, whilst disagreeing about what the next ten years will look like.

Why The Long End Matters For Real-World Borrowing Costs

When people say the long end sets borrowing costs, they’re pointing to the fact that many loans are effectively priced as:

Borrowing Rate ≈ Reference Market Yield + Spread

For fixed-rate mortgages, the reference is usually an intermediate-to-long maturity benchmark (and often swap rates and mortgage-backed pricing rather than the bank rate). Lenders want to lock in a rate that covers their own long funding cost and hedging cost. If those market yields stay high, mortgage rates stay high.

For corporate bonds, the reference is often the government yield at a similar maturity, plus a credit spread. Even if the central bank cuts, a company issuing a 10-year bond cares about the 10-year government yield and the market’s required spread for its credit risk.

For car loans and student loans, the link to the long end can be looser because pricing also depends on lender funding models, securitisation markets, credit underwriting, and competition. But the same general truth holds: the cost of term funding in markets matters, and that is not pinned one-for-one by the bank rate.

So when long yields ‘refuse to come down’, it is entirely plausible for headline borrowing costs to remain stubbornly high, even if short rates are falling.

A Practical Example: Two Different ‘Rate Cutting’ Worlds

Imagine the central bank cuts because inflation is falling and growth is slowing. Markets believe inflation is beaten and recession risk is real. In that scenario, expected future short rates will fall and the term premium may fall too. In which case, the whole yield curve tends to shift down. In that world, mortgages and corporate yields typically follow the bank rate.

Now imagine a different scenario where the central bank cuts, but markets see it as ‘fine-tuning’ rather than a full easing cycle. Inflation is lower than last year but still uncertain, fiscal borrowing is heavy, and investors demand more compensation to hold longer-dated bonds. Expected future short rates fall a bit, but the term premium rises. The front end drops and the long end barely moves.

Both scenarios involve cuts but they just have very different long-end mechanics.

The Takeaway

Short rates are the central bank’s territory. Long rates are the market’s verdict on the long-run path of policy, inflation, growth, and the price of uncertainty. If you keep that distinction in mind, you stop expecting the long end to be dictated by the bank rate. Sometimes it will. Sometimes it won’t. And when it doesn’t, that’s not irrationality. It’s the bond market telling you that the story beyond the next few meetings is different from the story in the next few years.