Best ETFs to Buy and Hold for 20 Years

A 20-year ETF is judged on durability and cost, not on this year's return. Three traits do most of the work. First, broad diversification: a fund that holds hundreds or thousands of companies cannot be wrecked by any single one failing, and 20 years is long enough for once-dominant companies to stumble or disappear. Second, very low fees, because (as the math below shows) a small annual percentage compounds into a large dollar drag over two decades. Third, your own discipline to keep contributing and stay invested. The funds that fit are deliberately boring: VOO, VTI, and VT are broad, cheap, and easy to hold.

A broad index fund also handles its own housekeeping: it holds whatever the market holds, so new leaders show up inside it without you switching funds. That self-updating quality, and the case for never selling and passing the holding on, is covered in depth in our hold-forever guide. What a concrete 20-year horizon adds on top of all that is arithmetic: the compounding math, the fee drag, and a finite end date. The rest of this guide is about those three. For the growth-leaning version of the same long-horizon logic, see our best ETFs for long-term growth guide.

The reason a 20-year horizon is worth treating as its own case is the compounding. Twenty years is long enough that the curve bends steeply: most of the end balance comes from growth on growth, not from what you put in. The table below is a hypothetical illustration only, not a prediction or guarantee. It assumes a $10,000 starting amount plus $300 contributed every month, compounded monthly at three fixed annual rates, with all figures rounded.

Two things stand out. First, the gap between rates widens dramatically by year 20: the same contributions reach roughly $172,000 at an illustrative 6% but around $301,000 at 10%, so small differences in annual return compound into large end gaps. Second, of the roughly $226,000 in the 8% column, only about $82,000 is money you actually contributed; the rest is compounding. That is the whole argument for a long horizon. Real returns are never this smooth: markets fall hard in some years, and a real 20-year path zig-zags through crashes rather than rising in a straight line. These numbers show the shape of compounding, not a path you should expect. Walnut is not an investment adviser.

Fees matter more over 20 years than over any short window, because the fee is charged against your balance every single year and quietly steals the future growth that money would have produced. The broad core funds charge around 0.03% a year (VOO and VTI), while plenty of actively managed funds charge 0.50% to 1% or more. That gap looks like a rounding error in year one and turns into real money by year 20. The table below is again a hypothetical illustration: a single $100,000 balance growing at a fixed 8% gross for 20 years, charged three different annual fees.

The 0.03% fund ends near $463,000; the 0.50% fund near $424,000, and the 1.00% fund near $386,000. The same gross return, the same 20 years, and roughly $39,000 to $77,000 evaporates into fees, the kind of difference that buys years of retirement. This is why long-term holders obsess over expense ratios: it is the one cost you fully control, and over two decades it compounds against you exactly as powerfully as returns compound for you. Picking a broad fund near 0.03% rather than a pricey active one is, over a 20-year hold, one of the highest- certainty decisions available. Fees change, so confirm the current figure on each issuer's site.

Three funds cover most 20-year buy-and-hold portfolios, in widening circles. VOO holds the S&P 500, the roughly 500 largest US companies, at around 0.03%. VTI holds the total US market, roughly 4,000 stocks adding the mid- and small-cap tail VOO leaves out, at the same cost. They overlap almost completely at the top, so most people hold one, not both. Either is a sensible single US core for decades.

One circle wider, VT (Vanguard Total World Stock) holds roughly 9,500 stocks across the US plus developed and emerging international markets at around 0.07%. It is the simplest single-fund global core because it never asks you to manage the US-versus-international split yourself. If you would rather hold US and international separately, VTI plus VXUS (total international) is the two-fund version of the same idea. Those who want a higher-growth tilt sometimes add VUG (large-cap growth), SCHG, or QQQ (the Nasdaq-100) on top of a broad core, accepting bigger swings in exchange for a growth lean. For the absolute simplest starting point, see our best ETFs for beginners guide, and for the never-sell framing specifically, our best ETFs to buy and hold forever guide.

This is the one place a 20-year plan genuinely diverges from a forever hold: a finite horizon has an end date, and that makes the sequence of returns near the end matter in a way it never does for an open-ended holding. Picture two investors who both average the same return over 20 years. The one who happens to hit a 30% crash in year 19, just as they plan to draw the money, can end up far worse off than the one who hit the same crash in year 3 and had 17 years to recover. The average return is identical; the timing of the bad year is what differs. A true buy-and-hold-forever holder can shrug at a late crash because there is no deadline; a 20-year holder may not have that luxury.

That is why a finite horizon can argue for gliding a portion of the portfolio toward bonds (for example a total-bond fund like BND or AGG) in the final years, say as year 15 to 20 approaches, to dampen how hard an end-of-horizon crash hits. The trade-off is that bonds usually grow more slowly, so gliding too early gives up compounding you wanted; the point is to reduce the risk concentrated at the finish line, not to abandon stocks. How much, and when, depends entirely on when you actually need the money, which is a personal decision a website cannot make for you. The contrast is the key idea: a forever hold need not glide at all, because it has no finish line. For that never-sell framing, see our hold-forever guide. Walnut is not an investment adviser.

A 20-year frame is long for a reason: it has historically been long enough for the broad market to recover from the crashes that happen inside it. Most rolling 20-year periods for the broad US market have ended higher than they started, even windows that contained severe downturns along the way. This is a description of the past, not a promise about your particular 20 years. The honest caveat is real: a 20-year window that happens to end in a deep slump can finish disappointingly, and there is no rule that says it must end well.

The practical takeaway is not that 20 years guarantees a good outcome; it is that the length of the horizon has historically given downturns time to pass before the finish line, which is exactly why the end-of-horizon glide above matters most in the final stretch rather than throughout. Investors who captured two decades of compounding were usually not the smartest stock-pickers; they were the ones who kept contributing and stayed invested through the drops in between. Past performance does not predict future results, and broad index funds can still fall sharply in any year.

Costs and holding counts are approximate as of early 2026; verify the current figures on each issuer's site. The three equity funds share the same pattern: broad, cheap, and self-updating. The bond row (a total-bond fund like BND or AGG) is the optional ballast some investors glide into as the end of the horizon nears, not a core most people start with on day one of a 20-year plan.

For a fixed 20-year plan, an AI assistant is most useful for the arithmetic this guide is built on: seeing what your current holdings would do under different illustrative growth rates, checking the fee drag on the funds you actually own, and figuring out roughly when an end-of-horizon glide toward bonds might come into view. The useful questions are concrete: at an illustrative 6%, 8%, or 10%, what range could my current holdings reach in 20 years; how much am I paying in fees across my funds each year; how concentrated am I in a few mega-caps; and how has my portfolio actually tracked the S&P 500 over time rather than just last week. Seeing real numbers tends to quiet the impulse to react to headlines.

That is where Walnut fits. It connects your existing brokerage and lets you ask, in plain language through Claude, ChatGPT, or a built-in assistant, what you hold, what it costs you in fees, how the pieces overlap, and how each position has done against the market, so a 20-year plan (including when to start thinking about bonds as year 15 to 20 nears) is something you can actually see and revisit. It is read-only by default, and you approve any trade. Walnut is not an investment adviser; it helps you understand and act on your own portfolio rather than telling you what to buy or when to sell.

The best ETFs to buy and hold for 20 years are the broad, ultra-low-cost index funds: VOO or VTI for a US core at around 0.03%, or VT for the whole world in one ticker. But the fund choice is the easy part of a 20-year plan; the math is the point. Over two decades, compounding does most of the heavy lifting (in the 8% illustration above, only about a third of the end balance was money you contributed), a high fee quietly drains tens of thousands of dollars, and the finite end date means a late crash can hurt more than an early one, which is why some investors glide toward bonds like BND as the finish line nears. Keep contributing, keep costs near 0.03%, and mind the last few years; those are the levers a concrete 20-year horizon actually gives you.

From a connected account you can dig into any of these as an ETF, look at an individual stock one of them holds, or compare it against the full set in our best ETF in every category guide. Holdings, weights, and fees change over time; treat the specifics here as a starting point and confirm on each provider's site before deciding.