Do You Need Bonds in Your Portfolio? Why I Hold 100% Stocks in 2026

I'm in my thirties, I've been investing for years, and I have never owned a single bond. Not one. Not a government bond, not a bond fund, not even the sliver of them hiding inside a target-date fund. My portfolio is one all-world stock ETF, bought automatically every month, and that's the whole thing.

So when someone asks me, do you need bonds in your portfolio, I answer as a guy who owns exactly zero of them and sleeps fine.

For years that felt like a confession. Every investing book on my shelf, every retirement calculator, and a solid half of Reddit repeat the same line: you need bonds, and you need more of them with every birthday. Holding none felt like breaking a rule written by Benjamin Graham and John Bogle personally.

Here's what I actually think, argued honestly. The "age in bonds" rule is out of date for a lot of people, the newest research quietly agrees, and yet the case for zero bonds breaks in one place nobody puts on the thumbnail. You get both halves.

The "age in bonds" rule is older than you think

The rule fits on a napkin: take 100, subtract your age, and that's your percentage in stocks. The rest goes in bonds. Thirty years old? Seventy percent stocks. Sixty? Forty. It's tidy, it's memorable, and it has a pedigree. Graham liked a version of it, and Bogle said he used it himself later in life. Later versions crept up to "110 minus your age," then "120," as people lived longer and yields fell.

You don't have to go looking for this rule. It's the default. Roughly $4.8 trillion sits in target-date funds right now, up more than 20% in a single year, and they're built on exactly this idea. You pick a fund with your retirement year on the label, and it shifts you out of stocks and into bonds as you age, whether you chose that glide path or not. I got into how that default actually works in target-date funds and the 401(k) default, because it's worth knowing what you've been quietly opted into.

And I want to be fair to the rule before I start poking it. Bonds have earned their reputation in real crashes. In 2008, while the S&P 500 fell about 37%, ten-year Treasuries returned +20%. That's the dream scenario: the boring asset zigs while the exciting one zags, and the mix hurts less. Some bonds have genuinely bought you that calm. The rule isn't stupid, it's just pointed at the wrong variable.

The research that flipped the script on the 100% stock portfolio

For most of my investing life, the case against bonds was just the return math. Stocks win over the long haul, and it isn't close.

Take Damodaran's dataset, which runs back to 1928. A hundred dollars put into US stocks with dividends reinvested grew to roughly $1.16 million by the end of 2025. The same hundred in ten-year Treasuries? About $7,750. In cash, around $2,600. After inflation that's roughly 6.8% a year for stocks, 1.5% for bonds, and basically nothing for cash. Over a lifetime, that gap isn't a rounding error. It's the difference between financial freedom and working an extra decade.

Data: Aswath Damodaran, NYU Stern, historical returns 1928-2025.

Bonds cost you that growth in exchange for a smoother ride. That was the whole tradeoff, and for decades it was the settled debate. Then a working paper went further than "stocks win on average."

In "Beyond the Status Quo," Aizhan Anarkulova, Scott Cederburg and Michael O'Doherty ran something more ambitious. Instead of the usual US-only history, they built a block bootstrap across 38 developed markets from 1890 to 2019 and asked which mix gave people the best shot at a comfortable retirement over a whole lifetime. Their answer is why the paper's been downloaded to death: an all-equity portfolio, roughly one-third domestic stocks and two-thirds international, no meaningful bond allocation, at every age. Not just while you're young. The whole way through.

The numbers they hang on it are loud. A target-date investor, they estimate, would need about 63% more savings to match the retirement outcome of the all-equity strategy. The chance of running out of money before you die came out at 8.2% for the all-equity mix versus 16.9% for a target-date fund, less than half. Average wealth at retirement, roughly 30% higher.

Data: Cederburg, Anarkulova & O'Doherty, 'Beyond the Status Quo' (SSRN 4590406). Simulated; a contested working paper.

If that holds, the age-in-bonds rule isn't just suboptimal. It's expensive enough to cost you years of your working life.

Now the honesty part. A lot of blogs would spike the football here. This is a working paper, not settled science, and it has serious critics. Cliff Asness at AQR called it, and I'm quoting, "not financial analysis, it is finger painting." His point is fair: stocks carry higher expected returns, so of course a backward-looking simulation that rewards whatever rose most ends up shouting "buy more stocks." Others note the 1890 to 2019 history includes markets an ordinary person could never actually have bought, and a 33% home weight is still a big bet in a small country. Morningstar keeps its 2026 safe withdrawal rate near 3.7% precisely because it assumes real humans can't stomach a 100% stock ride through retirement.

Even Bill Bengen, who invented the 4% rule, just raised his own "safe" withdrawal number to 4.7% in his 2025 book. Here's the tell: he did it by widening the equity side, adding international, small-cap and micro-cap stocks, not by adding bonds. If you want what that does to how much you can safely pull from a portfolio, I dug in on the safe withdrawal rate and the 4% rule in 2026.

So: compelling, current, and genuinely contested. I lean toward it being mostly right. I also know "mostly" is doing a lot of work in that sentence.

Why my boring all-world ETF is almost the "optimal" portfolio (with one honest asterisk)

Here's the part that made me laugh when I first read the paper. Its "optimal" portfolio, zero bonds, globally diversified stocks, held for life, is basically a description of what I'd been doing by accident for years. One all-world ETF, autoinvested monthly, no bonds, a big international slice. I hadn't designed the academically optimal portfolio. I was just too lazy to build anything fancier, and the laziness lined up.

Except that's where most people, a lot of finance influencers included, oversimplify it. So let me not.

A cap-weighted all-world fund, Vanguard's VT or the European VWCE, is not one-third domestic and two-thirds international the way the paper models it. As of mid-2026, VT is about 61.9% United States. If you're an American, that fund is nearly 62% domestic, almost the opposite of Cederburg's one-third home tilt. So the lazy line you'll read everywhere, "just buy VT and you own the optimal portfolio," is quietly wrong for a US investor.

The reason it fits me is an accident of geography. I'm European. My actual home country is a rounding error inside that fund, well under 5% of it. So from where I sit, that 62% US allocation isn't domestic at all. It's my international exposure, the biggest chunk of "not home" that I own. For a European holding an all-world fund, the portfolio really is overwhelmingly international, close to what the paper calls optimal. For an American holding the identical fund, it isn't.

Same ETF, opposite story, depending on which passport is in your drawer. If you want the actual mechanics and which funds give you that global spread cheaply, I compared them in the best all-world ETFs. My portfolio matches the spirit of the research, all stocks, globally spread, no bonds, but I got there by an accident of my passport, not because a single fund is magically "the optimal portfolio" for everyone.

The part nobody puts on the thumbnail

If I stopped here, this would be another "bonds are dumb, go 100% stocks, you're welcome" post, and those are everywhere. They're also incomplete, because the all-equity case does not solve two very real problems. It just hides them somewhere you can't see yet.

The first is sequence-of-returns risk, a clumsy name for a brutal idea. Two people can retire with the same million dollars, withdraw the same amount, earn the same average return, and one dies rich while the other goes broke. BlackRock has a clean illustration: same $1M start, same $60k annual withdrawals, same 7% average. The one who hits great returns early cruises to about $1.1M after 35 years. The one who hits a couple of ugly years right at the start runs out of money before the finish. Same average, opposite ending, purely from the order the returns arrived in.

The danger concentrates in the five years either side of your retirement date. A 100% stock portfolio that eats a 40% crash the year you start drawing down can be permanently crippled, because you're selling shares into the hole to pay your bills. This is the one spot where "no bonds" stops being clever and starts being reckless, and it's why I treat sequence of returns risk and the FIRE retirement playbook as the honest counterweight to everything above. During accumulation, a crash is a sale. In the first years of drawdown, the same crash is a wound.

"Sure," you might say, "I'll just hold through it, I've got the stomach." Maybe. I thought so too, and here's the uncomfortable part: I've never actually been tested. I started investing after the last real bloodbath. I have never sat there 100% in stocks and watched half my net worth evaporate the way it did in the 2008 crash, when the S&P fell about 57% peak to trough. My risk tolerance is a theory I've never had to defend in the only exam that counts.

And the data on how people actually behave is not flattering. Morningstar's "Mind the Gap" study tracks what funds returned versus what investors in those funds actually earned. Over the ten years to the end of 2024, the funds returned 8.2% a year. The investors earned 7.0%. That 1.2-point gap is the cost of bad timing, buying high and panic-selling low, and it's worst in the most volatile funds, where it stretches to 1.8 points a year.

Data: Morningstar 'Mind the Gap' 2025.

Read that the right way and bonds stop being a return drag and start being a psychology tool. A slug of bonds or cash won't beat stocks over 30 years. What it might do is stop you selling everything at the bottom of a 50% crash, and one avoided panic can outweigh decades of the equity premium you were chasing.

The twist that keeps me honest the other way, though: bonds are not a guarantee of calm. 2022 is the cautionary tale nobody who loves bonds wants to talk about. Stocks fell about 18%. Normal enough. But the Bloomberg US Aggregate bond index fell about 13%, its worst year since the index began in 1976. The "safe" 60/40 portfolio dropped roughly 17%, its worst since 1937. Both sides sank together.

Data: Morningstar; Bloomberg US Aggregate Bond Index.

The lesson of 2022 isn't "bonds are useless." It's that the safe part of your portfolio has to be genuinely safe for the job. Long bonds got wrecked. Short-term bonds, T-bills and cash held up. A safety sleeve against sequence risk should be the boring short stuff, not something that can drop 13% in the year you need it.

So, do you need bonds in your portfolio? A framework that ignores your birthday

So should you invest in bonds in 2026? Strip out the ideology and it comes down to one question: what job are you hiring the money to do? Age is a lazy proxy for that. This is how I actually think about how much of your portfolio should be in bonds.

If you're in the accumulation phase, ten or more years from touching the money, with a stable income and a real emergency fund, the honest answer is you probably need little or no bonds. Your paycheck is your bond. Every month you're adding new money, so a crash is a discount, not a disaster. Automating contributions matters more here than any bond allocation, because it removes the one button that can actually hurt you, the sell button.

If you're within about five to ten years of drawing the money down, or you already know in your gut that you'd panic-sell in a crash, then yes, start building a bond and cash sleeve. Not because you turned a particular age, but because the job changed. You're about to depend on this money, and sequence risk is now the thing that can end your plan. One to three years of spending held in cash and short bonds means you never have to sell stocks into a crash just to eat.

Keep your emergency fund out of this entirely. It isn't part of your allocation, it's what stops you raiding your investments, and in accumulation a fat cash fund does the same job people usually hand to bonds. And if you want a safe sleeve as rates drift down through 2026, remember 2022 and reach for short bonds, T-bills, CDs or a high-yield savings account, not long duration.

And the question I get whenever I say all this: if not bonds, what about gold? Short version, gold has returned around 5.6% a year nominally over the long run, well under stocks, so I treat it as a small optional hedge, not a core holding. I laid out how much of it, if any, makes sense in how much gold should you hold.

Notice what's missing from all of that. Your birthday. Stocks-versus-bonds allocation by age is the wrong axis entirely. The real ones are your time horizon, your income stability, and your honest read on your own nerve.

What I actually do (and the mistake I'd warn you about)

So after all that, my setup is deeply unglamorous. 100% equities, one accumulating all-world ETF, a fixed amount autoinvested every month whether the market is euphoric or on fire. No bonds. In place of a bond allocation I keep a deliberately oversized cash emergency fund, doing the "don't make me sell stocks at the worst time" job while I accumulate. And I've written the later plan down on paper: about five years before my wife and I pull the FIRE trigger, we start building a bond and cash tent to blunt sequence risk in the danger zone. Not at a birthday. At a fixed distance from the finish line.

The mistake I'd warn you about is the one I'm still exposed to myself. My risk tolerance is theoretical. I said it already, but it's the most important caveat in this whole article, so I'll say it twice. I have never held 100% stocks through a genuine 50% crash. Neither have most of the people confidently telling you online that bonds are for cowards. It's very easy to be brave in a bull market. The exam is graded during the crash, and plenty of us are going to find out we studied the wrong material.

That's exactly why I keep the contributions automatic and the emergency fund fat. I don't fully trust myself in a panic, so I've built a system that never asks me to be a hero. If you're going to run a 100% stock portfolio, do yourself the same favor. Assume you'll be scared, and take away the buttons you might press when you are.

Do you need bonds in your portfolio? Here's where I land

Bonds aren't a moral requirement and they aren't a scam. They're a tool for two specific jobs: smoothing the years right around retirement, and smoothing your own nerves so you don't torch the plan at the bottom. That's it. If you're young-ish, employed, automated and genuinely unshakeable, a 100% stock portfolio is defensible and the newest research has your back. If you're near the finish line, or you know you'd panic, hold some. Match the tool to the job, not to the number of candles on the cake.

For what it's worth, I'll hold zero bonds for another decade, then stop being a purist as my drawdown gets close. Ask me again after I've survived my first real crash fully invested. That's the only answer worth anything, and I haven't earned it yet.

If you'd rather see where you honestly land than guess, that's what I built My Financial Freedom Tracker for. Map your stock and bond split, model a drawdown, and stress-test the five years around your finish line before you bet your retirement on a stomach you've never tested. Disclosure, obviously: it's my product, I run it. But the principle stands with a spreadsheet too. Know the job before you pick the tool.