How Bond Yields and Stock Prices Interact

A Treasury yield is the base layer of the return investors require from stocks, so when yields rise and nothing else changes, the present value of companies' future profits falls. But the news that moves yields can also move expected earnings and the extra return investors demand for risk, which is why stocks sometimes fall as yields climb and sometimes rise alongside them.

Key takeaways

In the constant-growth model, the required return is the Treasury yield plus an equity risk premium; with a 4% premium and 3% growth, a yield rise from 2% to 4% cuts the model's fair P/E by 40%.
Stocks whose value sits in distant cash flows behave like long-term bonds: in an illustrative example, a long-duration stream loses 67% from a 2-point yield rise that costs a short-duration stream 33%.
Why yields rise matters: in 2022 the whole rise came from real (inflation-adjusted) yields, as the 10-year TIPS yield rose 2.62 points while expected inflation slipped, and the S&P 500 price index fell 19.4%.
In 2,241 rolling one-year windows ending October 2017 to October 2026, daily stock returns and yield changes were positively correlated about 66% of the time, with three lasting changes of sign.
The "Fed model", which compares the market's earnings yield with the 10-year yield, is widely quoted but criticized for mixing a real figure with a nominal one.

The discount-rate link

A share is a claim on cash a company will pay out over many years. Investors price it today by discounting those amounts at a required return: a risk-free rate, typically the 10-year Treasury yield, plus an equity risk premium, the extra annual return expected for holding stocks instead of Treasuries. The simplest version is the constant-growth, or Gordon growth, model:

P   = D1 / (r - g)
r   = y + ERP
P/E = payout / (r - g)

P      = share price today
D1     = dividend expected over the next year
r      = required return on the stock
g      = long-run annual growth rate of dividends (must be below r)
y      = risk-free yield, e.g. the 10-year Treasury yield
ERP    = equity risk premium
payout = share of earnings paid out (dividends / earnings)

Neither the premium nor the growth rate can be observed. NYU Stern professor Aswath Damodaran backs out the premium implied by S&P 500 prices; his estimate for October 1, 2026 was 4.20%, against a 5.29% Treasury rate. The Federal Reserve's May 2026 Financial Stability Report uses what it calls a crude equity premium measure, the forward earnings-to-price ratio minus the real 10-year Treasury yield, which was near a 20-year low as of April 2026.

Illustrative example: the same company at three yields

Assume an equity risk premium of 4%, a payout ratio of 50% and dividend growth of 3% a year, all held fixed. Only the Treasury yield changes.

10-year yield	Required return (r)	r − g	Fair P/E (0.5 ÷ (r − g))	Change from 2% yield
2%	6%	3 points	16.7	—
3%	7%	4 points	12.5	−25%
4%	8%	5 points	10.0	−40%

The effect depends on the gap between r and g: with 4% growth, the same yield rise takes the fair P/E from 25.0 to 12.5, a 50% fall.

Real markets do not hold the premium and growth fixed. During 2022 the 10-year Treasury yield rose from 1.52% to 3.88% (year-end values), and the S&P 500 price index fell 19.4%, far less than the model implies for a 2.36-point rise. Over the same year, Damodaran's year-end implied premium series moved from 4.24% to 5.94%, and the earnings figure it uses rose from 206.38 to 219.49 index points: the premium and the cash flows both moved.

Real yields versus nominal yields

A nominal Treasury yield is mainly a real yield plus compensation for expected inflation. The 10-year Treasury Inflation-Protected Securities (TIPS) yield measures the real part, and the gap between the two, published as the 10-year breakeven inflation rate, is a market-based gauge of expected inflation over the next decade.

Illustrative example: an inflation-only rise

Keep the earlier assumptions, but suppose the yield goes from 2% to 4% only because expected inflation rose 2 points, and nominal dividend growth rises from 3% to 5% with it. The required return goes from 6% to 8%, r − g stays at 3 points, and the fair P/E stays at 16.7. The 40% fall in the first example assumes growth does not rise with the yield, which describes a higher real yield, or inflation that profits fail to match. Whether profits keep pace with inflation depends on companies' pricing power.

What the TIPS data showed in 2022 and 2026

In 2022 the rise was entirely real: over the calendar year the 10-year TIPS yield went from −1.04% to 1.58%, up 2.62 points, while the breakeven rate slipped from 2.56% to 2.30%.

2026 shows the limits of reading yields alone. Between February 27 and October 5, 2026, the TIPS yield rose from 1.72% to 2.95% and the nominal yield from 3.97% to 5.31%, yet the S&P 500 price index rose 13.0%: in the model's terms, expected earnings or a lower premium offset the higher discount rate.

Why some stocks react more

Duration measures how sensitive a bond's price is to yields: the further away the cash flows, the bigger the price change. Patricia Dechow, Richard Sloan and Mark Soliman carried the idea over to shares in "Implied Equity Duration: A New Measure of Equity Risk" (Review of Accounting Studies, 2004). In their 2002 working paper they report that the standard predictions for bond duration hold for stocks, that volatility and beta rise with equity duration, and that book-to-market is a crude proxy for it. Duration tends to be low for firms with high return on equity, low growth and low valuations, and high for the reverse.

Illustrative example: two streams worth $100 each

Both streams are valued with the Gordon model at a 6% required return (2% yield plus 4% premium), and each is worth $100 today.

Measure	Short-duration stream	Long-duration stream
Next year's payment, growth	$4, 2% a year	$1, 5% a year
Share of today's value paid in the first 10 years	32%	9%
Value if the yield rises to 3%	$80.00 (−20%)	$50.00 (−50%)
Value if the yield rises to 4%	$66.67 (−33%)	$33.33 (−67%)

The long stream loses two to two and a half times as much because more of its value lies far in the future, where discounting compounds. That is why fast-growing companies are called rate-sensitive and mature, high-payout ones less so, all else equal.

Sector returns in 2022

Sector results in 2022 fit that logic only partly. The figures are calendar-year total returns from the Select Sector SPDR Trust prospectus (filed January 28, 2026) and, for the market, the SPDR S&P 500 ETF Trust prospectus (filed January 26, 2026).

Fund (sector of the S&P 500)	2022 total return	Usual rate story
SPDR S&P 500 ETF Trust (whole index)	−18.17%	Benchmark
Technology Select Sector SPDR (XLK)	−27.71%	Long duration: value tied to distant growth
Real Estate Select Sector SPDR (XLRE)	−26.20%	Income stocks, often called bond proxies
Financial Select Sector SPDR (XLF)	−10.56%	Bank lending margins can widen as rates rise
Utilities Select Sector SPDR (XLU)	+1.44%	Income stocks, often called bond proxies
Energy Select Sector SPDR (XLE)	+64.42%	Driven mainly by energy prices, not rates

Technology fell further than the index, as the duration argument predicts. But the two "bond proxy" sectors finished 27.64 points apart, and energy's gain had little to do with yields: one year cannot isolate rate sensitivity.

Why the sign keeps flipping

The reason yields move decides what happens to stocks. When yields rise on news of stronger growth, higher expected earnings can outweigh the higher discount rate, and the two rise together. When yields rise because real rates climb without a better growth outlook, or on inflation that investors doubt profits will match, the discount-rate effect dominates and stocks tend to fall.

Research ties these regimes to the economy's dominant shocks. John Campbell, Carolin Pflueger and Luis Viceira, in "Macroeconomic Drivers of Bond and Equity Risks" (Journal of Political Economy, 2020), trace the switch in Treasury bonds' exposure to stocks, from positive in 1979–2001 to negative in 2001–2011, to a change in how inflation moved with the output gap. Negative exposure means yields tended to fall when stocks fell. A San Francisco Fed Economic Letter of August 10, 2026 puts it plainly: demand shocks push stock prices and yields up together, while supply shocks lower activity and stock valuations but raise inflation and yields.

Ten years of daily data

The table uses FRED's daily S&P 500 price index, which carries 10 years of history, and the 10-year yield. For each day it computes the correlation between daily log changes in the index and daily yield changes over the previous 252 common trading days, roughly one year. Positive means stocks and yields tended to move together.

Window ending	Sign	Range of rolling correlation
Oct 17, 2017 – Sep 21, 2022	Positive in every window but one (Nov 24, 2021, at −0.002)	Up to +0.57 (Mar 10, 2020)
Sep 22, 2022 – Mar 27, 2025	Negative throughout	Down to −0.31 (Jul 5, 2024)
Mar 28, 2025 – Apr 8, 2026	Positive throughout	Up to +0.18 (Mar 5, 2026)
Apr 9, 2026 – Oct 8, 2026	Negative throughout	Down to −0.33 (Oct 1, 2026)

About 66% of the 2,241 windows were positive. Every figure sits between −0.33 and +0.57, so even at the extremes yield changes accounted for at most about a third of the variation in daily stock returns.

The window matters too: on the same data a two-year window changes sign three times, and a three-month window 25 times. The San Francisco Fed letter, using monthly data over five-year windows, describes the correlation as mostly positive until the early 2020s and negative since.

The Fed model and its critics

The "Fed model" compares the stock market's earnings yield (earnings divided by price, the inverse of the P/E) with the 10-year Treasury yield: well above it, stocks are called cheap relative to bonds; below it, expensive. By Damodaran's year-end figures, the S&P 500 earnings yield was 4.33% against a 1.51% Treasury rate at the end of 2021 (a 2.82-point gap) and 3.97% against 4.18% at the end of 2025 (−0.21 points).

Clifford Asness set out the objections in "Fight the Fed Model" (Journal of Portfolio Management, 2003). The model compares a real number, earnings that over the long term generally move with inflation, with a nominal bond yield; and as a forecaster of long-term stock returns, he found, it fails where traditional valuation measures do well. Investors did historically pay higher P/E ratios when nominal rates were lower, but he reads that as a habit, possibly a mistaken one, not as proof the model is valid. The Fed's financial-stability measure avoids the first objection by subtracting a real yield, but it still describes pricing, not timing.

Frequently asked questions

Which Treasury yield matters most for stock valuations?

Valuation work usually uses the 10-year yield, because stock cash flows stretch far into the future; Damodaran's premium and the Fed's measure both use a 10-year rate. Short-term rates matter through borrowing costs and cash alternatives, but are not the usual discount-rate benchmark.

How is the breakeven inflation rate calculated?

FRED derives the 10-year breakeven rate from the gap between the regular 10-year Treasury yield and the 10-year TIPS yield, and describes it as a measure of the average inflation market participants expect over the next 10 years. It was 2.33% on October 9, 2026. The gap is not pure expected inflation: Federal Reserve economists Stefania D'Amico, Don Kim and Min Wei found a sizable TIPS liquidity premium in past data, and an inflation risk premium is mixed in too.

Do high-dividend stocks behave like bonds?

Only partly. Dividends can be cut and no principal is repaid, while a Treasury's payments are fixed. The Utilities Select Sector SPDR returned +1.44% in 2022, when bond prices fell sharply, and −7.07% in 2023, per its prospectus.

Do higher interest rates always help bank stocks?

No. Higher rates can widen the net interest margin, what banks earn on assets minus what they pay on funding: the FDIC reported the industry margin at 3.37% in the fourth quarter of 2022, up 82 basis points in a year, the largest annual increase in the history of its Quarterly Banking Profile. The same rise in yields cut the value of bonds banks held, leaving unrealized losses on securities of $620.4 billion at the end of 2022. By the second quarter of 2026 the margin was 3.32%.

Can the equity risk premium be measured precisely?

No, it is inferred from prices and assumptions about future cash flows. On October 1, 2026, Damodaran's estimates ranged from 3.30% using normalized earnings and payout to 5.67% using the 10-year average cash yield, all against the same 5.29% Treasury rate. As the P/E example shows, small changes in the premium move model values a lot.

Checking a move in yields

Split the change: compare DGS10, DFII10 and T10YIE on FRED for the same dates to see whether the move was real or inflation-driven.
Identify the news behind it: a growth surprise and an inflation or policy surprise can move yields the same way with opposite effects on stocks.
Gauge a holding's duration: high price-to-book and P/E ratios on a fund fact sheet roughly signal value resting on distant growth.
Check any stock-bond correlation figure's data frequency, window and period, because the sign can change with each.
Treat earnings-yield comparisons and the Fed's premium measure as descriptions of current pricing, not timing signals.