Bonds At 5%: Is The 10-Year Treasury Finally A Buy?

The Fed Has Left The Bond Market

It is crucial that we begin this discussion by discussing what changed. For most of the years after 2008, the Federal Reserve was the largest buyer in the Treasury market. It held overnight rates near zero from late 2008 through 2015, and again from 2020 into 2022. It also bought trillions of dollars of bonds through “quantitative easing.” The purpose was explicit: push long-term yields below where a free market would set them.

It worked. From 2009 through 2021, the 10-year yield averaged about 2.3%. From 1990 through 2007, the average was closer to 5.8%.¹ The 2.3% era was not normal. It was the product of a central bank deliberately suppressing the price of money.

Today, the setup is the reverse.

• On September 16, the Fed raised its target range by a quarter point to 3.75%-4.00%. That was its first hike since 2023.
• The median official expects another before year-end.²
• The Fed’s “reserve management” bill purchases were cut to zero in August.³
• The balance sheet remains large at about $6.76 trillion, but is no longer growing. Notably, its share of GDP continues to fall.

Such is the backdrop for every bond investor right now. The buyer who set the price for 15 years has stepped back. It’s now private investors setting yields based on what they need to be paid. That’s the first reason 5% bonds deserve a fresh look.

How A Bond Yield Gets Set

A bond is a loan. You hand the government $1,000; it pays you a fixed coupon and returns your $1,000 at maturity. The yield is what lenders demand for tying up money for that long. As I wrote in a recent piece on what the data actually says about rising interest rates, demand has two main inputs: the growth the economy offers elsewhere and the inflation that erodes the dollars being repaid. On top of that sits the “term premium,” the extra pay for accepting the risk of holding a ten-year commitment.

Price and yield move in opposite directions because the coupon is fixed. If you own a bond paying 4% and new bonds pay 5%, nobody pays full price for yours. That’s the entire mechanism behind the 2022 losses. Rates went up, and prices went down, but nothing about the government’s ability to repay changed.

Is A 5% Treasury Yield Really “High”?

It feels high because a generation of investors is anchored to a 2% world. History says otherwise. Since 1953, the 10-year yield has averaged roughly 5.6%, and it has typically run a bit below the economy’s nominal growth rate.¹ To wit:

“If real growth is 2.5% and inflation is 3.5%, then a 6% nominal yield breaks even before any term premium. The investor isn’t going to lend at 2% in a 6% nominal economy because that’s a guaranteed loss of purchasing power and a worse return than the broader economy offers.”

Plug in today’s numbers. Real GDP grew about 2.2% over the past year, and headline CPI ran 3.4% in August.⁴ That puts nominal growth somewhere in the mid-5% range. A 10-year yield of 5.24% is sitting right where the economy says it should. A 5% yield is simply the price of money in an economy growing 5% in nominal terms.

Here’s where I’d add one caution to the popular version of this story. Wage growth is NOT what’s pushing yields up. Average hourly earnings rose 3.1% over the past year, slightly below the inflation rate.⁵ Workers are treading water in real terms. And when I broke down what is really driving normal interest rates, looking at the move in the 10-year from 2023 through mid-2026, roughly 97 of the 117 basis points came from real yields and term premium, with long-run inflation expectations moving only about 20.^6. In other words, the market is demanding more pay for duration in a world of $40 trillion in federal debt. That matters, and we’ll come back to it.

Look at the last row. In 2021, stocks offered more earnings yield than Treasuries offered in interest. That was the whole case for “TINA,” There Is No Alternative.

Today, a Treasury pays more than twice the stock market’s cyclically adjusted earnings yield. Even on Wall Street’s more generous forward estimates, the S&P 500 trades at 19.2 times earnings. That implies an earnings yield of about 5.2%, essentially identical to the 10-year, so an investor weighing bonds at 5% against stocks isn’t giving up any current yield to own the safer asset.⁷ My colleague Michael Lebowitz laid out this shift in From TINA To TIGA: Diversification Pays Again. There Is a Good Alternative.

Why Do You Own Bonds In The First Place?

This is the question I think too many investors have lost sight of after 2022. They check the price of their bond fund every week, see red, and conclude that an asset class that has cushioned portfolios through nearly every stock market crash of the last century is somehow “broken” because of one inflation shock. That’s like judging a house by one season’s weather. The price is the weather. The coupon and the maturity are the climate.

You own bonds for three reasons: lower volatility, the return of your principal, and a reliable stream of income. Since 1928, the annualized volatility of a 10-year Treasury has been about 5.5%, compared with roughly 15.5% for stocks.1 A Treasury held to maturity returns its full face value regardless of what happened to its price along the way. And with bonds at 5%, the income finally does real work. As I argued in “Why ditching bonds in your portfolio is the wrong move,” those three jobs don’t disappear because one bad year made headlines.

“But Lance, my bond fund is still underwater from 2022.”

I understand the frustration. Those losses came from a 1.5% starting yield, and the chart above shows how rarely bonds fail at their actual job. Notice that the failures cluster in a specific kind of year. The five years when stocks lost while bonds also fell were 1941, 1969, 1987, 2018, and 2022. Most were years of rising rates or rising inflation. Bonds protect you against a growth shock. They don’t protect you against an inflation shock, because inflation is the one thing that hurts both. That’s the trade-off you accept, and it’s why the starting yield matters so much. A bond bought at 1.5% had no cushion when inflation hit. A bond bought at 5% has a lot of it.

Let’s do some basic math. At a 5.24% yield, a 10-year Treasury has a duration of about 7.7 years. That means yields could rise roughly 0.68 percentage points over a year before the price loss wiped out the income. In late 2021, with the yield near 1.5%, that cushion was about 0.16 points.⁸ The same asset, purchased at a different price, carries a completely different risk.

Are Treasuries Still The “Risk-Free” Asset?

“The U.S. has $40 trillion in debt and lost its last AAA rating. How can anyone call Treasuries risk-free?”

It’s a reasonable question, and the facts behind it are real. Moody’s cut the U.S. to Aa1 in May 2025, joining S&P and Fitch at AA-level ratings. Net interest costs are projected to reach about $970 billion in 2025, roughly triple the 2017 to 2021 average.⁹

However, “risk-free” has always meant one specific thing: credit risk. The United States borrows in the currency it issues, and its debt is the collateral that banks, money market funds, foreign central banks, and the plumbing of the global financial system all run on every single day. Default isn’t the risk. The probability of the Treasury failing to pay a coupon or return principal is about as close to zero as finance gets. Such is why the downgrades barely moved the market on the day they happened. A rating agency changing its letter grade didn’t change the government’s ability to print the dollars it owes. The fiscal path still matters, though. Investors feel it through a higher term premium and bigger price swings rather than missed payments.

Let me be very clear on this point. When it is stated that Treasury bonds are a “risk-free” asset, that doesn’t mean Treasuries carry no risk at all. They carry two, and both showed up in 2022. The first is inflation risk, the chance that the dollars you get back buy less than you expected. The second is duration risk: price swings that come with long maturities when rates move. Right now, the Fed’s own messaging makes clear which one worries it most.

“The plain fact is that inflation is too high, and has been for too long.” – Kevin Warsh, Chair, Federal Reserve, September 2026

So the answer is yes and no. Treasuries remain the closest thing to a risk-free asset for return of principal. They are not risk-free for purchasing power, and they never were. The 1970s proved that long before 2022 did.

What Should Bond Investors Expect Over The Next Decade?

This is where the data gets compelling. Bonds have rarely been this unloved, and with good reason. Through September, the 10-year Treasury’s total return over the trailing five years ran about -2.4% a year. The low point came in May 2025 at roughly -3.7% a year, the worst five-year stretch in data going back to 1871.¹⁵ It’s natural to assume an asset class with that record is broken. History says the opposite.

Before this cycle, the 10-year trailing five-year return had fallen to roughly zero only four times since 1871: 1959, 1970, 1981, and 2018. The first three were each followed by a better decade, and two of them by a much better one.

Notice the pattern inside the pattern. The size of the rebound tracked the yield investors were left holding after the damage was done. In 1981, the losses had pushed the 10-year to 15%, and the following decade was one of the greatest bond bull markets in history. In 1959, the yield was under 5% and kept climbing through the 1960s, so the recovery was modest. And 2018 is the warning. Trailing returns were near zero then, too, but the yield was only 2.9%, and bond investors earned less than 1% a year over the following eight years. Bad trailing returns are only half the signal. What matters is whether they leave you holding a high starting yield.

Today, they do. The losses since 2020 came from yields rising from about 0.5% to more than 5%, which is precisely the setup that preceded the 1970 and 1981 recoveries. That’s also why the starting yield remains the best single guide to what comes next. Since 1881, it has explained about 91% of the variation in the 10-year return over the following decade. For anyone buying bonds at 5% today, starting points between 4.75% and 5.75% produced an average annualized ten-year return of about 5.6%, and the worst outcome in that group was still about 3.2% a year.¹ Past results don’t guarantee future returns, and a bond sold before maturity can return more or less than its starting yield. Such is the advantage of owning a contract rather than a claim on future profits. Stocks don’t offer anything close to that kind of predictability.

Equities work differently, but valuation still matters over long horizons. I’ve written about this before in the best measure of future stock market returns, and the pattern below is why I keep coming back to it.

Think about that carefully before deciding to ditch bonds from your portfolio. Over the last 126 years, whenever stock market valuations traded above 30x, Treasury bonds outperformed stocks 93% of the time.

A caveat matters here. Those 57 starting points with a CAPE above 30 come from just two episodes, 1929 and 1997 through 2002. That’s a thin sample, and two episodes can’t prove a rule. But the full regression of the CAPE against forward stock returns points in the same direction, implying roughly 1% to 2% per year from today’s level.¹ Vanguard’s capital markets model, which uses a broader set of inputs, produces a hypothetical projection of 4.2% to 6.2% a year for U.S. stocks over the next decade.¹⁰ Even the optimistic end of that range sits barely above what a Treasury pays today, with far more volatility attached. Will stocks underperform bonds over the next ten years? Nobody knows for certain. But are investors being paid much to take that bet?

Where The Bond Bulls Could Be Wrong

The strongest case against buying bonds at 5% is that yields keep going up. The bears are right about the deficit math. The government is projected to add another $22 trillion of debt over the next decade. That means a steady flow of new supply that the market must absorb.9 Oil is above $91 a barrel, 16 of 18 Fed officials expect another hike this year, and the term premium has been climbing.² If inflation reaccelerates into something closer to the 1970s, with energy prices feeding into wages and the Fed forced to keep hiking into a slowing economy, a 10-year bought at 5.2% could see yields of 6% or higher first. That outcome is possible.

Here’s the problem with treating it as the base case. Inflation shocks driven by energy have historically destroyed demand within months, which pulls growth and yields back down. Higher borrowing costs also work against demand the longer they last. And the starting-yield math works in your favor even in a bad scenario. This is because the income keeps compounding while you wait. A bond investor who bought at 5% and watched yields rise to 6% is reinvesting coupons at 6%. That’s how bond math repairs itself, slowly, without anyone needing to call the top in rates.

As Bob Farrell’s Rule #9 reminds us, when all the experts and forecasts agree, something else is going to happen. Today, the consensus is that rates are going higher for longer and that bonds are dead money. I’d be careful betting on that consensus at the highest yields in a generation.

What Bonds At 5% Mean For Your Portfolio

Let’s take a moment to visualize this carefully. The real payoff from bonds at 5% shows up in portfolio outcomes rather than in any single yield. Suppose an investor needs a portfolio to earn 6% a year to stay on track with long-term goals. For this example, put half the money in Treasuries yielding 5% and the other half in dividend-paying stocks yielding 2%. This is a hypothetical example, not a recommendation, but the arithmetic is worth walking through.

The Treasury half contributes 2.5 percentage points in interest per year. The stock half contributes another 1.0 point in dividends. That’s 3.5 points of the 6% target, or roughly 58%, delivered as cash before the stock market moves at all. The remaining 2.5 points have to come from price gains on the stock half. Since 1928, the S&P 500’s price alone, excluding dividends, has gained 6.4% annually.¹⁴

Therefore, with only 50% of the portfolio exposed to market risk, market volatility risk is greatly reduced. For example, during a 10% market decline, the portfolio would be down 5% in price. However, the portfolio would still pick up the 3% in income and dividends. Such would leave the portfolio with only a 2% decline during a 10% drawdown. That math allows investors to survive the long-term investing horizon. Today, with Treasuries at 5%, a 6% portfolio target asks stocks for less than their long-run average price gain.

Now run the same portfolio in late 2021. The Treasury half yielded about 1.5%, so interest contributed only 0.75 points, and total income covered just 1.75 points of the 6%. Stocks had to rise roughly 8.5% a year to close the gap. Over rolling ten-year periods since 1928, the S&P 500’s price has compounded at 8.5% or better only about 45% of the time. It has compounded at 5% or better about 64% of the time.¹⁴ The target is the same, but the odds are very different. Higher bond yields add income and lower the bar the stock market has to clear.

It also changes what a bad decade looks like. Recall the valuation data above, which suggests stocks may struggle to deliver much from today’s prices. If stock prices had simply gone sideways for 10 years, the 2021 version of this portfolio would have earned about 1.75% per year, below inflation. The version built with bonds at 5% would still earn about 3.5% a year from income alone. That’s hardly a great outcome, but it’s a survivable one, and getting through bad decades is the entire job of the bond side of a portfolio.

It is crucial to know that none of the numbers above is guaranteed.

1. Dividends can be cut,
2. A bond sold before maturity can be worth more or less than its face value, and
3. Fees, taxes, and inflation all reduce what an investor keeps.

What is most important is how you build the bond side of your portfolio.

• A ladder of individual Treasuries lets you lean on the starting-yield math directly.
• Shorter maturities reduce duration risk, and
• Inflation-protected Treasuries, with real yields near 3%, hedge the risk that nominal bonds can’t.¹¹

The right mix of bonds in your portfolio depends on your time horizon, income needs, and how much volatility you can tolerate. That’s exactly why it’s worth working through with an advisor rather than a headline.

Bottom line: At 5%, the 10-year Treasury offers a return that history says has been far more predictable than what a stock market priced for perfection is likely to deliver.

The last time bonds paid this much, the market was just starting a decade in which stocks went nowhere. Trade accordingly.

Sources and notes:

 

1. RIA Advisors calculations from Robert Shiller’s monthly U.S. stock market and long-term interest rate dataset (GitHub mirror, datasets/s-and-p-500), 1871 through September 2023. Ten-year Treasury total returns were modeled as those of a constant-maturity par bond repricing monthly. Shiller’s rate and price series are monthly averages, so calendar-year figures run one to four points off published returns (for example, 2022 Treasuries at -15.0% here versus roughly -17.8% in NYU Stern’s Damodaran dataset), with signs matching in every year checked. Forward-return studies use monthly starts from January 1881 through September 2013. Pre-1953 rates are long-term government yields rather than a constant 10-year maturity.
2. Federal Reserve FOMC statement and Summary of Economic Projections, September 16, 2026, as reported by Advisor Perspectives and Chase. Chair Warsh quote via Chase, “Fed Raises Rates in September, Officials Signal One More Hike in 2026.”
3. Wolf Street, “Fed Cuts Reserve Management Purchases (RMPs) to Zero, Starting August 14,” August 13, 2026. Balance sheet at $6.76 trillion, 20.7% of Q2 GDP. The pause was extended into October per reports on September 16, 2026.
4. Bureau of Economic Analysis, GDP third estimate for Q2 2026, released September 30, 2026. Bureau of Labor Statistics, CPI for August 2026, released September 11, 2026: headline 3.4% and core 2.4% year over year. Nominal growth estimate is an RIA approximation combining year-over-year real GDP and CPI.
5. Bureau of Labor Statistics, Employment Situation for August 2026: average hourly earnings up 3.1% year over year.
6. Lance Roberts, “Normal Interest Rates: What The Debt Panic Gets Wrong,” RealInvestmentAdvice.com, August 21, 2026.
7. FactSet Earnings Insight, September 25, 2026: forward 12-month P/E of 19.2. Shiller CAPE near 41.5 per Michael Lebowitz, “From TINA To TIGA,” September 30, 2026. December 2021 CAPE and yield from the Shiller dataset. 10-year yield of 5.24% on October 1, 2026, per Trading Economics, cross-checked against a 5.26% close on September 29. WTI crude at $91.49 per Charles Schwab market update, October 1, 2026.
8. RIA Advisors’ calculation of modified duration for a 10-year par Treasury at 5.24% (7.7 years) and 1.5% (9.3 years). Break-even yield rise is the yield divided by duration, a one-year approximation.
9. Peter G. Peterson Foundation, “Is the U.S. Headed for More Credit Downgrades?” 2026. Moody’s downgrade to Aa1, May 16, 2025.
10. Vanguard, “Setting Realistic Expectations,” Vanguard Capital Markets Model forecasts as of June 30, 2026. Forecasts are hypothetical and not guarantees.
11. 10-year TIPS real yield near 3.0% per Michael Lebowitz, “From TINA To TIGA,” September 30, 2026.
12. 10-year Treasury yields (constant maturity) and the effective fed funds rate from Federal Reserve data via Massive Market Data, January 2000 through September 29, 2026, plotted as monthly averages with the final point at the September 29 close. Regime averages computed from daily data: 4.60% (January 2000 to mid-December 2008), 2.27% (mid-December 2008 to mid-March 2022), and 4.04% (mid-March 2022 to September 29, 2026).
13. Kim-Wright 10-year term premium of 1.02% on September 25, 2026, via FRED (THREEFYTP10). Market 10-year breakeven inflation of 2.29% for August 2026 via Massive Market Data, with the Cleveland Fed model estimate at 2.57% for September. Real rate shown as the remainder. Term premium estimates vary by model, and the NY Fed ACM model read about 0.8% in May 2026.
14. RIA Advisors’ calculations. S&P 500 price-only compound growth of about 6.4% a year from January 1928 to September 2026, and rolling ten-year price returns since 1928, from the Shiller dataset (monthly average prices). The 50/50 portfolio is a hypothetical illustration, not a recommendation or a record of actual results. It assumes a 2% dividend yield, Treasuries held to maturity at the stated yields, and no fees, taxes, dividend changes or rebalancing.
15. RIA Advisors’ calculations of trailing five-year and forward ten-year annualized total returns for a modeled constant-maturity 10-year Treasury, using Shiller long-term rates from 1871 through September 2023 and Federal Reserve 10-year yields via Massive Market Data from October 2023 through September 2026 (monthly averages, which agree with Shiller’s series to within 0.01 point over 2015 to 2023). “Near-zero” episodes are the lowest trailing five-year return within each stretch of returns below 1.5% per year. The April 2018 forward figure covers 8.4 years. Because the series uses monthly average yields, returns will differ modestly from index data.
    

A message from Advisor Perspectives and VettaFi: Discover something new! Click here to register for our upcoming webcasts.

---

© Real Investment Advice

More Energy Topics >