Bonds Yield 5% Again. Should You Rebalance Your 60/40 Now?

Ben Carlson’s October 13, 2026 post on A Wealth of Common Sense makes a simple case. A 60/40 portfolio of global stocks (ACWI) and U.S. bonds (AGG) that was left alone from the end of 2019 had drifted to 76.6% stocks by September 30, 2026. Bond yields are now above 5%, so a 60/40 no longer needs a huge stock return to earn a decent one. His advice: rebalance back to your target weights, because “one day your best performers will be your worst performers.”

We agree with the advice. An investor who chose 60/40 and now holds 77/23 owns a portfolio they never picked, and fixing that is the whole job of a rebalancing policy. Two of the supporting claims are worth testing against data, though, because they shape what investors expect to get from the trade: that rebalancing “will pay off at some point,” and that 5% bond yields are a reason to do it now. We ran both through 98 years of annual U.S. returns.

Drift Changes the Risk You Own

A drifted portfolio behaves like a more aggressive one, because it is one. The table applies four bad calendar years from Aswath Damodaran’s return data (S&P 500 and the 10-year Treasury, total returns) to a 60/40 portfolio and to Carlson’s drifted 76.6/23.4 mix.

The 2022 row matters as much as the 2008 row. When stocks and bonds fall together, the mix makes no difference, and rebalancing buys nothing. Rebalancing protects you only against the kinds of stock declines that bonds have historically offset: recessions and growth scares. The guide on whether bonds still diversify covers why that relationship flips when inflation drives markets.

Over a Century, Never Rebalancing Earned More

Every long backtest we found reaches the same result. Over the full sample, a portfolio left alone earns more than a rebalanced one, because stocks earn more than bonds and the unrebalanced portfolio slowly becomes a stock portfolio.

The Vanguard 2019 row is the one taxable investors should notice. Getting back on track (McNamee, Paradise, and Bruno) charged a 30% income tax and a 20% long-term capital gains rate on every trade, and the never-rebalanced portfolio still earned the most. It also had 14.0% volatility against 11.4% to 11.8% for the rebalanced versions, and the worst Sharpe ratio in the table. Vanguard titled that figure “Any reasonable rebalancing strategy beats not rebalancing at all,” and it means beats on risk-adjusted return. Both Vanguard papers say the purpose of rebalancing is to control risk; the 2019 paper puts it as “not to maximize returns.”

Over Any Single Decade, It Was a Coin Flip

A century is longer than anyone’s holding period, so we repeated the test over every rolling window in the Damodaran data: a 60/40 S&P 500 / 10-year Treasury portfolio, rebalanced each December 31 or never, with no fees or taxes.

The windows overlap, so 89 ten-year windows hold far fewer than 89 independent decades. The pattern by era is clear anyway. The rebalanced portfolio ended richer in windows that contained a long stock slump: those starting in 1928 to 1937, 1965 to 1974, and 1993 to 2008. The drifting portfolio won in long bull markets: windows starting in the 1940s and early 1950s, around 1990, and 2009 to 2016. Across all ten-year windows, the median difference in annual return was 0.01 points, in favor of rebalancing.

Carlson’s line that rebalancing “will pay off at some point” is right in one sense. Over 98 years there were several stretches when it left investors with more money. Whether a given investor’s decade is one of them is a forecast about the stock market, and nobody can make it reliably. What rebalancing did more consistently was cap risk. Over ten years the drifting portfolio had the deeper peak-to-trough decline in 60 of 89 windows and the worse single year in 60. Over twenty years its median worst decline was −20.2% against −13.9%.

Rebalancing did not always cushion a crash. Starting in 1929, the rebalanced portfolio kept buying stocks on the way down to 1932 and fell further on annual data, −40.1% against −34.4% for the one that drifted. A rebalancing rule controls how much stock you own going into a decline. How deep the decline goes is up to the market.

Starting Yield Predicts Bond Returns Over Long Horizons

The second half of Carlson’s case is about bonds. A bond fund’s yield is its income, and a fund’s return also includes price changes when rates move. Over one year the price change can swamp the yield. Over longer horizons the two offset: rising rates cut prices but let the fund reinvest at higher yields. Leibowitz, Bova, and Kogelman (2014) showed that for a fund that holds its duration steady, the annualized return converges to the starting yield at a horizon of about twice the duration minus one year. For a fund with a six-year duration, that is about eleven years.

We checked this on the 10-year Treasury, comparing each December 10-year yield from the Federal Reserve’s H.15 series with the annualized return over the following years.

Extending the yields back to 1928 with Robert Shiller’s long-rate series gives the same answer: a ten-year correlation of 0.95, within a point in 66 of 89 windows. The misses are informative. Buyers at the end of 1970 and 1971, at yields near 6%, earned about 3.4% a year over the next decade as inflation pushed rates up. Buyers at the end of 1976, at 6.87%, earned 9.97% as rates fell in the 1980s.

On October 13, 2026 the 10-year Treasury yielded 5.21%, according to Summitward’s daily Treasury feed. History says that is a reasonable central estimate for a 10-year Treasury held for a decade, with a typical miss of about a point. It says much less about the next year. A fund with a 4.5% yield and a five-year duration loses money in a year when rates rise more than about 0.9 points; the stress test in the bond diversification guide lets you try other cases.

Higher Starting Yields Made Rebalancing Cheaper

If bond yields predict bond returns, they should also predict how much rebalancing costs, since the cost of rebalancing is the gap between what stocks and bonds earn. The rolling windows bear that out.

When windows began with yields of 5% or more, rebalancing back to 60/40 cost nothing at the median and came out ahead more often than not. That fits Carlson’s chart. Treat it as suggestive. The high-yield windows come almost entirely from one rate cycle, the rise to 1981 and the long fall after it, and the low-yield windows overlap with the postwar boom and the post-2009 bull market. Low yields also failed as a signal to load up on stocks: the 1928 to 1937 starts had yields below 4%, and rebalancing won every one of them.

The 7% Chart Is an Identity

Carlson’s chart, from his firm Ritholtz Wealth Management, shows the stock return a 60/40 needs to earn 7% at bond yields from 1% to 6%. It is one line of algebra:

Every bar checks out: 11.0% at a 1% bond yield, 8.3% at 5%, 7.7% at 6%. The chart’s “We are here” box covers both the 4% and 5% bars. At the 10-year Treasury’s 5.21%, stocks need 8.2%. Three things the chart leaves to the reader:

• 7% is a nominal target. With the market’s 10-year breakeven inflation rate at 2.35%, 7% nominal is about 4.5% real. Using the 2.88% 10-year TIPS yield for bonds, stocks need about 5.7% real, a little above the roughly 5% real we use as a planning assumption.
• The bond input is a yield. As the table above shows, it is a good estimate of a decade’s bond return and a poor estimate of next year’s.
• A weighted average of expected returns differs from the compound return you earn. Volatility lowers compound returns, and rebalancing between imperfectly correlated assets adds a little back. The rebalancing bonus guide works through both effects.

The chart’s point stands: a 5% bond leaves less work for stocks than a 1% bond did. How likely a 60/40 is to earn 7% is a separate question the chart does not try to answer.

Rebalancing Back to Target Differs From Changing the Target

Carlson writes that “it made sense to over-rebalance your portfolio in the first part of this decade because bond yields were so low,” and that now it makes sense to go back. Those are two changes to the target allocation, made because of market conditions. A rebalancing policy, by contrast, trades back to a fixed target whatever yields are doing.

Both approaches can be reasonable, but they need different evidence. Shifting from 60/40 to 80/20 because bonds yield 1% is a bet that stocks will beat bonds by more than usual, and it raises the losses you have to sit through. In the 2020s that bet paid off. The same signal in the late 1920s and 1930s would have hurt, as the last table shows. If you did move to 80/20 on purpose in 2020, ask whether your goals, horizon, or ability to take losses have changed since. If they have not, going back to 60/40 because yields rose is a second tactical move. The guides on locking in TIPS yields and tactical equity risk for retirees cover the line between reacting to prices and forecasting them.

Higher yields do not settle whether bonds will hedge. Whether bonds rise when stocks fall depends on what is moving markets. AQR’s model (Brixton et al., 2023) ties the stock-bond correlation to how volatile growth is relative to inflation, and the Bank for International Settlements found the U.S. correlation turned positive in mid-2021. A 5% yield gives bonds more room to rally in a recession. It does not stop them falling with stocks in an inflation scare.

Treasuries Crossed 5% for the First Time Since 2007

Carlson writes that high-quality bonds now yield more than 5% “for the first time in almost 20 years.” That holds for the 10-year Treasury. In FRED’s daily series, its last close above 5% before this year was 5.04% on July 19, 2007. The October 2023 peak closed at 4.98%. This year it first closed at 5.00% on September 15 and reached 5.31% on October 5.

Investment-grade corporate bonds are a different story. On October 19, 2023, the ICE BofA U.S. Corporate Index yielded 6.44%, above its 5.94% on October 14, 2026, and its AAA-rated subset yielded 5.64%, close to today’s 5.73%. FRED: ICE BofA corporate effective yield. Corporate yields also include pay for credit risk, so a 5.9% corporate yield and a 5.2% Treasury yield are not the same opportunity.

Closing a 17-Point Drift Without a Large Tax Bill

Take a $1 million portfolio at 76.6/23.4. Holding prices fixed, getting back to 60/40 means either selling $166,000 of stock and buying bonds, or adding $276,667 of new money, all of it to bonds, without selling anything. Few households have that much new cash on hand, so contributions alone will not fix a drift this large quickly. Work through the cheaper options first:

1. Trade inside tax-advantaged accounts. Selling stock and buying bonds in a 401(k) or IRA triggers no tax. In Vanguard’s 2019 simulation of a household split evenly between taxable and tax-advantaged accounts, rebalancing in the tax-advantaged side first added 44 basis points a year after tax without raising volatility. The household portfolio guide explains why the allocation that matters is the total across accounts.
2. Send new money and dividends to bonds. This closes part of the gap at no tax cost, though dividends and interest in a taxable account are still taxable when paid.
3. Trade back to the edge of your band. Vanguard’s 2024 research on target-date funds trades only when a 60/40 drifts 2 points and then only back to 1.75 points from target. In its simulations that cut transaction costs to about a quarter of monthly rebalancing’s, worth 15 to 22 basis points a year for an accumulating investor. A DIY investor pays taxes rather than trading costs, but the logic carries over: a smaller trade back to the band edge realizes fewer gains than a full reset.
4. In taxable accounts, sell the highest-basis lots. Selling specific lots (see tax-efficient selling) and pairing gains with harvested losses keeps the bill down. If the whole $166,000 came from taxable shares with a 40% gain, the realized gain would be $66,400. At a 15% long-term rate plus the 3.8% net investment income tax, that is about $12,500 of federal tax before state tax. Much of that tax would have come due eventually, so the cost is mostly the lost deferral. The exception is low-basis stock you expect to hold until death, when heirs receive a stepped-up basis.

Try It: Check Your Own Drift

The checker starts with the $1 million example split across a 401(k) and a taxable account. Because the stock trim is set to happen inside the 401(k), it restores 60/40 with no tax. Change the balances, cost basis, and targets to match your accounts.

When Not to Go All the Way Back

• You own an all-in-one fund. Target-date funds and Vanguard’s LifeStrategy funds rebalance internally, so the LifeStrategy 60/40 fund Carlson cites (VSMGX) holds its 60/40 mix. Check the stock and bond total across your other accounts instead.
• Your target has changed for a reason you can state. A longer horizon, a larger pension, or a higher savings rate can justify more stock. Write the new target down first, then rebalance to it.
• The only overweight is low-basis stock in a taxable account. If the breach is small and selling means a large gain, especially later in life, partial rebalancing with new cash and dividends may be the better trade-off.
• The bonds you would buy do not match the job. Money for inflation-adjusted spending may belong in TIPS rather than a nominal aggregate fund; see asset-liability matching.

How We Ran the Numbers

Stocks are the S&P 500 and bonds are a 10-year constant-maturity Treasury, both annual total returns from Damodaran, 1928-2025. Portfolios start at 60/40 and either rebalance on December 31 or never trade. There are no fees, taxes, or cash flows, and annual data understates declines that happen inside a year. Starting yields are December values of the H.15 10-year yield (FRED GS10) from 1953, extended to 1927 with Shiller’s long-rate series. The ACWI/AGG drift figures are Carlson’s, from his firm’s chart; our S&P 500 / 10-year Treasury version drifted to 78.2% stocks by the end of 2025.

Frequently Asked Questions

Should I rebalance now that bonds yield 5%?

If your portfolio has drifted outside the band in your rebalancing policy, yes. The yield is a minor reason. The main one is that a 77/23 portfolio loses more in a stock crash than the 60/40 you chose. In our 1928-2025 test, ten-year windows that started at yields of 5% or more saw rebalancing cost nothing at the median, though nearly all of them come from one rate cycle.

Does rebalancing increase returns?

Usually it does not over very long periods. Never rebalancing earned more over the full century in every study we reviewed, including Vanguard’s after-tax test. Over ten-year windows since 1928 it was close to even. Rebalancing reliably lowers risk.

Is a 77/23 portfolio too risky?

It depends on whether you can live with its losses. In 2008 it would have lost about 23% against 14% for 60/40, using S&P 500 and 10-year Treasury returns. If you chose 60/40 because a 14% loss was your limit, 77/23 exceeds it.

Should I sell stocks in my taxable account to rebalance?

Only after using tax-advantaged accounts and new cash. If you still need to sell, sell the highest-basis lots and consider stopping at the edge of your band instead of the exact target.

Do I need to rebalance a target-date or LifeStrategy fund?

No. Those funds rebalance internally. Check the stock and bond mix across all your accounts instead, since other holdings can still drift.

Should I buy long-term bonds to lock in high yields?

Longer bonds lock in today’s yield for longer but lose more if rates keep rising. Match the bond’s duration to when you need the money; the asset-liability matching guide shows how.

Related Guides

• How Often Should You Rebalance Your Portfolio? for frequency, bands, and the 5/25 rule
• Are Bonds Still Good Diversifiers? for when bonds hedge stocks and when they fall together
• The Rebalancing Bonus for the math of compound returns and rebalancing
• What a 60/40 Portfolio Really Owns for what sits inside the bond sleeve
• You Have One Household Portfolio for rebalancing across accounts
• Are Ben Carlson’s Best Inflation Hedges Actually Inflation Hedges?

Key Takeaways

• Carlson’s advice holds: a 60/40 that drifted to 77/23 since 2019 carries more stock risk than its owner chose.
• Over the full 1928-2025 record, never rebalancing earned more (9.45% against 8.34% for S&P 500 / 10-year Treasury) and ended at 99.6% stocks. Over ten-year windows it ended richer about half the time.
• The 10-year Treasury’s starting yield came within a point of its next-decade return in about two-thirds of windows since 1954. Next year’s return is much less predictable.
• Ten-year windows that began at yields of 5% or more saw rebalancing cost nothing at the median, though nearly all come from a single rate cycle.
• Fix drift inside tax-advantaged accounts first, then with new cash, then with high-basis taxable lots, and consider stopping at the band edge.

Sources

1. Carlson, B. (2026, October 13). Is It Time to Rebalance Your Portfolio? A Wealth of Common Sense.
2. Jaconetti, C.M., Kinniry, F.M., & Zilbering, Y. (2010). Best Practices for Portfolio Rebalancing. Vanguard Research, Figure 6.
3. McNamee, J.L., Paradise, T., & Bruno, M.A. (2019). Getting Back on Track: A Guide to Smart Rebalancing. Vanguard, Figures 4 and 6.
4. Zhang, Y., Ahluwalia, H., Daga, A., & Zi, Y. (2024). The Rebalancing Edge: Optimizing Target-Date Fund Rebalancing Through Threshold-Based Strategies. Vanguard Research.
5. Bernstein, W. (1996). The Rebalancing Bonus: Theory and Practice. Efficient Frontier.
6. Leibowitz, M.L., Bova, A., & Kogelman, S. (2014). Long-Term Bond Returns under Duration Targeting. Financial Analysts Journal, 70(1).
7. Brixton, A., Brooks, J., Hecht, P., Ilmanen, A., Maloney, T., & McQuinn, N. (2023). A Changing Stock-Bond Correlation: Drivers and Implications. Journal of Portfolio Management, 49(4).
8. Lombardi, M.J., & Sushko, V. (2023). The Correlation of Equity and Bond Returns. BIS Quarterly Review, December 2023, Box A.
9. Damodaran, A. Historical Returns on Stocks, Bonds and Bills: 1928-2025. NYU Stern.
10. Board of Governors of the Federal Reserve System, H.15, via FRED: GS10 (monthly) and DGS10 (daily). ICE Data Indices via FRED: BAMLC0A0CMEY and BAMLC0A1CAAAEY.
11. Shiller, R.J. U.S. Stock Markets 1871-Present and CAPE Ratio (ie_data.xls, “Rate GS10” column).
12. IRS. Topic No. 559, Net Investment Income Tax and Topic No. 409, Capital Gains and Losses.