US Treasury historical returns from 1992 to mid-2026 ranged from about 2.7% a year for T-bills to 4.9% for 10-year Treasuries. But after inflation, cash and short-dated government bonds barely broke even, while a 60/40 stock-bond portfolio turned $10,000 into roughly $180,000.
Most investors know that stocks beat bonds over the long run. What they rarely see is the actual scoreboard for the “boring” end of the market - the difference between holding cash, rolling 2-year notes, or riding 10-year Treasuries for a third of a century. So we built it. Using total-return index data from Bloomberg for the period from the first quarter of 1992 through the second quarter of 2026 (34.25 years), we compared six portfolios on the metrics that matter: annualized return, return after inflation, volatility, drawdowns, and best and worst years.
The single most common mistake in this kind of analysis is to treat a bond’s yield as its return. It isn’t. A bond’s total return is the coupon income plus the price change as interest rates move plus the effect of reinvesting coupons:
Total Return = Coupon Income + Price Change + Reinvested Coupons
When rates fall, a bond’s price rises and total return beats the starting yield; when rates rise, total return can be far lower - even negative - despite a positive yield. To capture that correctly, we used published total-return indices rather than raw yields:
Two honest caveats. First, the data is quarterly, so the drawdown figures below are a lower bound — they miss the deepest intra-quarter dips (the S&P’s 2007–09 slide, for instance, was worse month-to-month than the quarterly series shows). Annualized returns and calendar-year figures are unaffected. Second, index ladders are a clean stand-in for a real portfolio, but they carry no fund fees or taxes.
Here is the full scoreboard. Every figure is annualized over the 34.25-year window; “real” return is after U.S. inflation, which ran 2.59% a year over the period. Portfolios are rebalanced quarterly.
| Portfolio (1992–2026) | Return (nominal) | Return (real) | Volatility | Max drawdown* | Worst year | $10,000 grew to |
|---|---|---|---|---|---|---|
| 100% T-bills (cash) | 2.69% | 0.10% | 1.1% | −0.03% | +0.04% | $24,806 |
| 100% 2-year Treasuries | 3.32% | 0.71% | 2.1% | −5.1% | −3.8% | $30,595 |
| 100% 10-year Treasuries | 4.92% | 2.28% | 7.1% | −20.9% | −14.9% | $51,831 |
| 50/50 (stocks / U.S. Agg) | 8.18% | 5.45% | 8.1% | −22.1% | −17.5% | $147,667 |
| 60/40 (stocks / U.S. Agg) | 8.80% | 6.06% | 9.6% | −27.4% | −21.6% | $179,889 |
| 100% S&P 500 | 11.02% | 8.22% | 15.8% | −45.8% | −37.0% | $358,788 |
Source: Bondfish analysis of Bloomberg total-return index data; U.S. CPI for inflation. *Max drawdown from quarterly data — a conservative (lower-bound) estimate of the true peak-to-trough loss.
Growth of $10,000 by portfolio, 1992–2026. Source: Bondfish analysis of Bloomberg total-return index data.
Three patterns jump out, and each is worth unpacking: the near-zero real return on cash, the reward for taking duration risk, and the remarkable efficiency of a simple stock-bond blend.
Over 34 years, T-bills earned 2.69% a year - and inflation took almost all of it. The real return on cash was just 0.10% annually, which is to say roughly nothing. Ten thousand dollars in T-bills grew to about $24,800 in nominal terms, but in purchasing power it essentially stood still.
What cash did deliver was near-perfect stability: a maximum drawdown of −0.03% and not a single negative calendar year. That is the trade. Cash is the one asset here that never scared you and never grew your wealth. For an emergency fund or money you need next year, that is exactly the point. For a multi-decade portfolio, it is a slow leak.
Cash didn’t lose you money. It just quietly declined to make you any.
Stepping out of cash and into government bonds is where the premium - the extra return investors demand for locking up money longer - shows up in the data.
The numbers track that theory almost exactly:
| Holding | Nominal return | Real return | Volatility | Worst year |
|---|---|---|---|---|
| T-bills | 2.69% | 0.10% | 1.1% | +0.04% |
| 2-year Treasuries | 3.32% | 0.71% | 2.1% | −3.8% |
| 10-year Treasuries | 4.92% | 2.28% | 7.1% | −14.9% |
Source: Bondfish analysis of Bloomberg total-return index data.
Each step out the curve added return and added risk in lockstep. The 10-year ladder earned 2.2 percentage points more per year than cash and was the only Treasury option to deliver a meaningful real return (2.28%). But it paid for that with a -20.9% drawdown and a -14.9% worst year - the brutal 2022 bond sell-off, when the fastest rate-hiking cycle in decades hammered long-duration bonds.
That volatility is duration at work, and it explains why longer bonds swing so much harder than short ones. Duration measures how sensitive a bond’s price is to interest-rate moves, and as a rule of thumb the price change is the negative of duration times the yield change:
Price Change ≈ −Duration × ΔYield
So a portfolio with a duration of one year loses roughly 1% if rates rise one percentage point, while a 10-year-duration portfolio loses about 10%. If you are fuzzy on this, our explainer on the difference between a bond’s duration and its maturity is the place to start. The lesson from 34 years of data: the term premium is real and worth harvesting, but only if you can stomach the interim losses.
Put the safe assets next to the S&P 500 and the compounding gap is stark. Stocks returned 11.02% a year and turned $10,000 into roughly $359,000 - about 7× what 10-year Treasuries produced and 14× what cash produced.
But that return came with equity-sized pain: a -45.8% drawdown (and the true month-to-month low was deeper still) and a −37% worst year in 2008. This is the fundamental tension every investor navigates. Bonds are not there to beat stocks; they are there to be the ballast that lets you stay invested when stocks fall apart. Which is exactly what the blended portfolios show.
The classic 60/40 portfolio — 60% stocks, 40% bonds, rebalanced — is the workhorse of retail investing, and the data shows why it has endured. Over our window it earned 8.80% a year, capturing about 80% of the S&P 500’s return with roughly 60% of its volatility and a much shallower drawdown (−27% vs −46%). The more conservative 50/50 gave up a little return (8.18%) for even calmer ride quality.
Our 60/40 figure lands almost exactly on the long-run average Vanguard cites for the strategy - about 8.8% a year since 1926 - which is reassuring corroboration across two very different time windows. The mechanism is diversification: bonds cushion the equity drawdowns, rebalancing quietly sells high and buys low, and the blend’s year-to-year consistency compounds into a competitive long-run result.
One nuance worth flagging for today’s investors: the stock-bond diversification that powered the 60/40 has been less reliable since 2020, as both BlackRock and Vanguard have noted, because higher and more volatile inflation has at times pushed stocks and bonds down together. The 34-year record is excellent; it is not a guarantee for every future decade.
Nominal returns flatter cash and short bonds. Adjust for the 2.59%-a-year bite of inflation and the ranking sharpens into a clear message about real return on bonds:
Because inflation is what silently erodes fixed coupons, understanding how inflation eats into bond returns is essential context for reading any of these numbers. The headline takeaway is uncomfortable but clean: holding only cash or ultra-short bonds is a decision to tread water in real terms.
Averages hide the moments that make investors sell at the bottom. The worst single years in our data tell you what each strategy actually felt like to hold:
This is the case for owning bonds even when their returns look unexciting: they shorten and shallow the drops that would otherwise tempt you to abandon the plan.
None of this is a forecast - it is a 34-year base rate. Used that way, the data suggests a few durable principles:
US Treasury historical returns from 1992 to 2026 show a clear hierarchy: T-bills preserved purchasing power but grew it by essentially zero, 2-year notes added a whisper, and 10-year Treasuries were the first government bond to deliver a real return - at the price of a −21% drawdown. A simple 60/40 portfolio outperformed every all-bond option after inflation while sparing investors the full −46% gut-punch of owning only stocks. Safety and growth are different jobs; the data says use bonds for the first and a blend for the second.
This article is for general information only and is not investment advice. Past performance does not guarantee future results, and the historical figures here are index-based and exclude fund fees and taxes. Bond investing involves risk, including possible loss of principal. Consider your own circumstances or consult a licensed financial professional before investing.
