Rate Regimes and Asset Returns

Level, change, and which one reprices

The level of rates enters a valuation twice: through the discount rate applied to the cash flows, and through the growth that the economy will produce, which feeds the cash flows themselves. A high-rate world with strong nominal growth can support high multiples, and a low-rate world with stagnant growth can produce low ones — which is why the simple statement “rates up, stocks down” is a description of a transition rather than of a level. What matters is whether the level is embedded in expectations and whether growth justifies it. The change is the operative variable, and it works through the multiple rather than through the earnings. If the earnings estimate is unchanged and the required return rises by two points, the present value of a distant stream falls by roughly the duration times the change — which is the same arithmetic as the bond lesson, applied to an equity. The practical consequence is that the response to a rate move is proportional to how far out the cash flows sit: a utility with a stable dividend has a duration of perhaps fifteen years, a mature industrial twenty, and a company whose profits are expected in a decade forty or more. A single rate move therefore compresses a whole market unevenly, and the pattern of the compression is the market’s own statement about which cash flows it considers distant. The third distinction is real against nominal. Equities are claims on real output, so the rate that matters for them is the real required return; inflation that is fully passed through to prices leaves real earnings intact and only raises nominal earnings. A rise in nominal yields driven by inflation expectations can therefore leave equities relatively unharmed if the pricing power is there, while a rise in real yields — the compensation for capital itself — compresses multiples whichever way inflation is moving. That is why the useful question about a rate move is not how large it is but whether it is real or nominal, and it is the same decomposition the fiscal and expectations lessons needed. Two points of required return, three horizons — A utility, cash flows about 15 years out: a de-rating of roughly 15 × 2% ≈ 30% on the multiple · A mature industrial, about 20 years: roughly 40%, partly offset by pricing power ← · A growth company, cash flows 40 years out: the multiple can fall by more than half · The offsetting force: the earnings estimate, which is why the two must be separated Duration arithmetic on an equity is an approximation and it assumes the cash flows are fixed. A company that can raise prices with inflation, or whose revenue grows faster in a stronger economy, has a partial offset that no duration number captures — which is why the de-rating is a first approximation and the earnings revision is the second force.

What the historical record actually shows

Three empirical regularities are worth holding. First, equity returns are negatively related to rate changes and only weakly related to rate levels, and the relationship is stronger for the real rate than the nominal one. Second, the relationship runs through the multiple rather than through earnings, which means that in a year of rising rates the earnings can grow and the shares still fall, which is what makes the experience so disorienting. Third, the effect is not uniform across the market: the compression lands on the longest-duration valuations, so a rising-rate year tends to be one in which cheap, cash-generative businesses outperform expensive, growth-optional ones — a value-versus-growth rotation that is a mechanical consequence of the discount rate rather than a change in the relative quality of the businesses. The 2022 case shows all three at once. The policy rate went from near zero to over four percent in twelve months, real yields rose sharply, and a broad equity index fell about 18% in a year in which earnings were essentially flat. What fell most was the part of the market with the longest duration. What held up best was energy and parts of the value complex, which had both near-term cash flows and pricing power against the inflation that was causing the rate move in the first place. The same lens explains the earlier decades. The long fall in rates from the early 1980s to 2021 was a tailwind for multiples and a persistent advantage for long-duration assets, and it was mostly a re-rating rather than a change in earnings growth — which means that some of the returns from that period were a one-time repricing rather than a repeatable return. Knowing which part of a long-run return came from the discount rate rather than from the cash flows is the difference between a strategy and a regime. • Rate changes matter more than rate levels for equity returns. • The effect runs through the multiple, so earnings and prices can move in opposite directions. • The compression is largest where the cash flows are furthest out. • A decades-long fall in rates contributed a one-time re-rating that cannot be repeated. The mirror image is worth stating: a stable or falling real rate with stable growth is the configuration that supports the highest multiples, and it is also the one whose reversal is most damaging. Priced-as-permanent is the expensive mistake in both directions.

The real rate is the one that binds

A nominal policy rate is not the number that reprices assets. What matters is the **real** rate — the nominal rate minus expected inflation, or equivalently the yield on inflation-protected debt — because that is the return you earn in purchasing power and, roughly, the rate that sits in a discount formula alongside real cash flows. Two economies can share a 4% policy rate and have very different real rates: one with 1% expected inflation and one with 5%. The first is tight; the second is loose. The headline rate alone cannot tell them apart. This is why the relationship between rates and equity valuations is unstable in the short run and clearer over longer spans. In the years when inflation expectations were anchored, a change in the nominal rate was mostly a change in the real rate, and long-duration assets fell. When inflation expectations move with the nominal rate, the real rate barely changes and valuation pressure is weaker than the headline suggests. The 2022 repricing worked because real rates rose sharply from deeply negative levels; the same nominal move from a different inflation path would not have done the same damage. For a portfolio, that gives a simple ordering of what to watch. Expected inflation first, the real rate next, and the nominal policy rate last, as a *signal* about the first two. When a central bank is at 3.75–4.00% and inflation expectations are near target, the real rate is meaningfully positive — a regime that historically favours cash flows today over cash flows far away. “Rates went up and stocks fell” is not a law. It is a description of the cases where the real rate rose, and it is silent about the cases where it did not.

The duration of a business

The read above separated a rate’s level from its change. The other half of the framework is on the asset side, and it is the same idea in a different market: every business has a **duration**, the weighted average timing of the cash it will produce, and that duration predicts which equities a discount-rate change lands on hardest. The analogy is direct. A company whose value rests almost entirely on cash flows far in the future — a business investing heavily now for a payoff years away — is like a long zero-coupon bond: a change in the discount rate changes the present value of the whole thing at once, with no nearer cash flows to cushion it. A company producing stable cash today with modest reinvestment needs is like a short-dated bond, and the same rate change moves it much less. The labels growth and value are a rough proxy for this, and they are a worse one than they appear, because a mature company with a collapsing core and a growth company with a dominant, cash-generating franchise can sit on the wrong side of their labels. The practical way to rank holdings is therefore by duration-like measures rather than by label. A low ratio of current cash flow to value implies a long duration, so a high multiple is a duration statement as well as a valuation statement. The payout ratio says how much of the return is near-term. And the sensitivity of the multiple to a change in the risk-free rate, measured from history, is the direct estimate — noisier than the others and the most relevant. The framework earns its place because it says *where* a rate move lands, not merely whether it lands. The repricing of 2022 is the clearest recent evidence: the longest-duration equities fell the most, and that ordering explained the dispersion across the market better than the growth-versus-value classification did, because it is the mechanism rather than the label. Two cautions belong beside it. Duration is not a fixed property of a business: a company’s cash-flow timing changes with its investment opportunities and its competitive position, so a sensitivity estimated from one period carries into another only loosely. And the exposure has two separate forms — the *level* of rates affects the multiple permanently by changing the discount applied to the terminal value, while the *change* produces the repricing event — which is why a portfolio can be positioned differently with respect to each. The practical use is a stress test rather than a forecast. Rank the book by a duration proxy, and ask what the portfolio does if the discount rate changes by a percentage point, with the long-duration names moving worst. That is a more informative test than a market-wide beta, because it names the mechanism. • A business’s duration is the weighted timing of its cash flows, and it sets rate sensitivity. • A low current cash flow relative to value implies a long duration, hence a high multiple effect. • The 2022 dispersion was ordered better by duration than by the growth-value label. • Stress the book by duration, because that names the mechanism rather than the statistic. The level and the change are separate exposures: a permanently higher rate level compresses a long-duration multiple every year, while the change produces the one-off repricing. A portfolio can be positioned differently with respect to each, and usually is, without the holder intending it.

Sources

• The equity discount rate, multiples and interest ratesStandard equity-valuation practice; MR13 on duration, and the Fundamentals subject on the cost of capital and the discount rate
• Rate changes, inflation and long-horizon equity returnsEmpirical literature on the stock-bond correlation and inflation regimes
• Long-duration equities and the de-rating of 2022Contemporaneous market evidence; standard duration-based explanations