Treasury Bond Yields vs Stock Returns: When Bonds Actually Win

Treasury Bond Yields vs Stock Returns: When Bonds Actually Win

Most people in their 30s with a brokerage account have been told the same story: stocks always beat bonds over the long run, so why would you bother with Treasuries? It’s a compelling narrative, and for most 20-year holding periods, it’s even statistically defensible. But “over the long run” is doing an enormous amount of heavy lifting in that sentence, and for knowledge workers trying to build real wealth across a 20-to-40-year career, the nuances matter enormously.

Here’s the thing most people miss about this topic.

Related: index fund investing guide

The reality is that there are specific, identifiable conditions under which Treasury bonds not only compete with stocks but genuinely beat them — sometimes by a significant margin. Understanding those conditions isn’t just academic trivia. It can mean the difference between a portfolio that survives a brutal decade and one that forces you to make desperate decisions at exactly the wrong moment.

The Baseline: What the Historical Data Actually Shows

Let’s start with what the bulls are right about. Over rolling 20-year periods since 1928, U.S. large-cap equities have outperformed long-term Treasury bonds roughly 70-75% of the time (Siegel, 2014). That’s a real edge, and dismissing it would be intellectually dishonest. The equity risk premium — the extra return investors demand for holding volatile assets — has historically averaged somewhere between 4% and 6% annually depending on how you measure it.

But that 25-30% of the time when bonds win? Those aren’t random noise. They cluster around specific macroeconomic regimes, and several of those regimes look disturbingly recognizable right now. The 1930s, the 1940s, the 2000s — these weren’t flukes. They were predictable outcomes given the starting conditions. A knowledge worker who understood those conditions in January 2000 and shifted meaningfully toward Treasuries would have watched the S&P 500 lose roughly 45% over the next two and a half years while their bond portfolio compounded quietly in the background.

Condition One: Starting Valuations Are Stretched

The single most powerful predictor of whether bonds will beat stocks over the next decade isn’t the Fed funds rate or GDP growth forecasts. It’s the starting valuation of the equity market. Specifically, the cyclically adjusted price-to-earnings ratio — the CAPE or Shiller P/E — has a remarkably robust inverse relationship with subsequent 10-year stock returns (Campbell & Shiller, 1998).

Here’s the intuition: when you pay 35 times cyclically adjusted earnings for a stock index, you’re essentially locking in a forward earnings yield of roughly 2.9%. A 10-year Treasury yielding 4.5% doesn’t need to do anything heroic to beat that. It just needs to exist. The equity investor, by contrast, needs either significant earnings growth, multiple expansion, or both — just to keep pace.

This is not a theoretical exercise. When CAPE ratios have historically sat above 30, subsequent 10-year real returns on equities have averaged around 0-2% annually (Shiller, 2015). Meanwhile, a 10-year Treasury purchased at a 4% nominal yield and held to maturity delivers its promised return with mathematical certainty, assuming no default — which, for U.S. government debt, has never occurred.

The practical implication: if you’re entering a market with a CAPE above 30, you should be asking yourself very seriously whether the equity risk premium is actually being offered at that price, or whether you’re paying so much for the risk that the premium has evaporated.

Condition Two: Entering a Deflationary or Disinflationary Environment

Bonds are interest rate instruments, and they perform best when rates fall. Rates fall most reliably when inflation decelerates or turns negative. This isn’t complicated, but it has enormous practical consequences for portfolio construction.

During the Great Depression, long-term Treasuries delivered positive real returns while stocks lost roughly 80% from peak to trough. During Japan’s “lost decade” — which turned into two lost decades — Japanese government bonds dramatically outperformed equities. During the 2008 financial crisis, 30-year Treasuries returned over 25% as deflation fears dominated and investors fled to safety.

The mechanism is straightforward: falling inflation reduces the discount rate applied to future cash flows, which increases bond prices. Simultaneously, deflationary environments typically crush corporate earnings, compress equity multiples, and trigger credit events that hammer stocks. The two effects compound in bonds’ favor and work against equities simultaneously.

For knowledge workers trying to read the current environment, the question to ask is not “is inflation high right now?” but rather “is the direction of travel disinflationary, and how fast?” A rapid deceleration from 8% inflation to 3% is a bond-friendly environment even if headline inflation is still positive, because the rate-cutting cycle it implies will mechanically boost bond prices.

Condition Three: The Yield Curve and Real Yield Signals

There’s a specific technical threshold that deserves more attention from non-professional investors: real Treasury yields. The real yield is simply the nominal Treasury yield minus expected inflation (usually proxied by TIPS breakeven rates). When real yields on 10-year Treasuries are meaningfully positive — say, above 2% — Treasuries are offering something genuinely valuable: a guaranteed positive return above inflation.

This matters because stocks’ entire justification for their higher expected return is that they’re riskier than risk-free assets. If the risk-free rate itself is delivering 2-3% real returns, the equity risk premium needs to compensate for substantially more volatility, drawdown risk, and behavioral difficulty. Research on asset allocation across rate regimes suggests that when real yields are high, fixed income allocations deliver better risk-adjusted outcomes even for investors with long time horizons (Asness, Frazzini, & Pedersen, 2012).

Conversely, when real yields are deeply negative (as they were from 2020-2022 when nominal 10-year yields were near 1.5% but inflation was running at 7-8%), bonds are almost guaranteed to lose real purchasing power. That’s not a regime in which Treasuries win. Context is everything.

Duration Risk: The Weapon That Cuts Both Ways

One thing that trips up a lot of smart people when they first explore bond investing is duration. Duration is essentially the sensitivity of a bond’s price to changes in interest rates. A 30-year Treasury has very high duration — its price can swing 20-25% in response to a 1% move in long-term rates. A 2-year Treasury note has low duration and won’t move much.

When bonds win big against stocks, it’s usually long-duration bonds doing the heavy lifting. During the equity bear market of 2000-2002, 20+ year Treasuries returned over 40% cumulatively. During the worst months of 2008, they were the only major asset class with significantly positive returns. But that same duration that creates those spectacular protective returns can destroy you if you’re holding long bonds in a rising-rate environment — as anyone who held TLT (the 20+ year Treasury ETF) through 2022 discovered painfully, watching it lose over 30%.

The practical framework: long-duration Treasuries function as a powerful hedge against deflationary recessions and equity bear markets, but they’re not a set-it-and-forget-it holding. They require a thesis about the interest rate environment, not just a generic preference for “safety.”

The Behavioral Dimension: Why Bonds’ Real Value Is Underpriced

Here’s where I want to push back against the purely quantitative framing. The academic literature on stock returns vs. bond returns typically compares terminal wealth under different allocations, and stocks usually win on that metric over long periods. But terminal wealth comparisons ignore something crucial: what happens to investor behavior during the drawdowns that separate the starting point from the finish line.

The psychological toll of watching a portfolio fall 40-50% — even if you intellectually know it will recover — is severe enough that most investors don’t stay the course. Research on investor behavior consistently shows that individual investors systematically buy high and sell low, underperforming even the assets they hold because of poorly timed decisions during volatile periods (Barber & Odean, 2000). The average equity mutual fund investor has historically earned substantially less than the fund itself, precisely because of panic selling during downturns.

Bonds, particularly Treasuries, help with this problem not because they’re mathematically superior in a vacuum, but because they dampen portfolio volatility enough to keep investors in their seats. A portfolio that loses 20% during a recession is one most people can tolerate. A portfolio that loses 45% is one that produces panic-selling, which permanently locks in losses and guarantees the investor earns far less than the historical equity average.

For knowledge workers specifically — people who have high human capital but demanding jobs that make deep market monitoring difficult — the behavioral argument for a meaningful Treasury allocation is actually quite strong. You want a portfolio you can largely ignore for months at a time without your stomach dropping every time you check your phone.

When to Tilt: A Practical Framework Without the Complexity Theater

I’m not going to pretend there’s a clean algorithm for when to overweight Treasuries versus stocks. But there are observable conditions that should raise your allocation to bonds above the textbook default:

• CAPE ratio above 30: The equity risk premium is likely compressed. Bonds don’t need to be exciting to beat stocks on a risk-adjusted basis when equity valuations are this stretched.
• Real 10-year Treasury yields above 2%: You’re being paid genuinely to own safety. That’s rare enough to be worth noting and acting on.
• Yield curve inversion: An inverted yield curve (short-term rates above long-term rates) has preceded every U.S. recession since the 1960s. It doesn’t tell you the exact timing, but it tells you the probability distribution has shifted. Recessions are generally bond-friendly and equity-hostile.
• Credit spreads widening rapidly: When corporate bond spreads blow out, it often precedes equity market stress. Long Treasuries in this environment tend to benefit from flight-to-quality flows.
• Inflation decelerating from elevated levels: The direction of the rate cycle matters more than the current level. A central bank pivoting toward cuts is a powerful tailwind for bond prices.

None of these signals is infallible. Multiple signals aligning simultaneously, however, creates a much stronger case for a meaningful Treasury tilt than any single indicator alone.

The Forgotten Power of Holding Bonds to Maturity

There’s one scenario where the entire price volatility discussion becomes irrelevant, and it’s the most straightforward way to think about Treasuries for people who aren’t professional traders: buying and holding to maturity.

A 5-year Treasury note yielding 4.5% purchased today will return exactly 4.5% annually if held to maturity, regardless of what rates do in the interim. The price might fluctuate wildly in years two and three, but the terminal cash flow is fixed. For a knowledge worker who has a specific financial goal — a down payment in five years, a child’s college tuition in seven years, a planned career break in ten years — matching Treasury maturities to spending goals creates a guaranteed real-dollar outcome that no equity investment can match.

This is called liability matching or goal-based investing in the financial planning literature, and it’s genuinely underused by individual investors who think of bonds only in terms of their portfolio return contribution rather than their ability to guarantee specific future cash flows. When your goal is “I need $150,000 in exactly six years,” a 6-year Treasury ladder is not a conservative choice — it’s the only choice that actually guarantees the outcome.

Putting It Together: Bonds as Tools, Not Consolation Prizes

The framing of “stocks beat bonds” versus “bonds beat stocks” is ultimately the wrong way to think about this. Both assets have genuine roles in a thoughtfully constructed portfolio, and the conditions under which each dominates are knowable in advance with reasonable confidence — not perfectly, but well enough to improve on the naive 60/40 default.

Treasury bonds win outright when equity valuations are stretched, when real yields are attractive, when inflation is decelerating, and when the economic cycle is turning down. They win on a behavioral basis whenever they keep an investor from making a catastrophic exit during a crisis. And they win with mathematical certainty when matched to specific future spending goals and held to maturity.

The knowledge worker who understands these conditions and adjusts accordingly isn’t playing defense. They’re playing a smarter version of offense — one that recognizes that avoiding large permanent losses is at least as important to long-run wealth accumulation as capturing the equity premium in years when it’s genuinely on offer.

Last updated: 2026-03-31

In my experience, the biggest mistake people make is

Sound familiar?

References

• Campbell, J. Y., Pflueger, C., & Viceira, L. M. (2025). Bond-Stock Comovements. Working Paper. Link
• Damodaran, A. (2026). Historical Returns on Stocks, Bonds and Bills: 1928-2024. New York University Stern School of Business. Link
• Haddad, V., Moreira, A., & Muir, T. (n.d.). How Quantitative Easing Changed the Bond Market. UCLA Anderson Review. Link
• Kostin, D. (2024). How higher interest rates affect US stocks. Goldman Sachs Insights. Link
• Morgan Stanley Global Investment Committee. (2025). Why Bonds May Keep Beating Stocks. Morgan Stanley Insights. Link

Related Reading

• What Is a REIT and How to Invest in Real Estate
• The Small Cap Value Premium: 97 Years of Data Most Investors Miss
• Roth Conversion Ladder Strategy [2026]