ETFs Explained: What You're Actually Buying — and the 4-Point Checklist for Picking One
There is now something like $10 trillion sitting in US-listed ETFs. For a lot of people, an ETF will be the only investment product they ever need to understand. And yet if you ask an ETF owner what they actually bought, the usual answer is a shrug: "it's basically a stock, but diversified." That's not wrong. It's just missing the parts that cost people money.
I've seen investors spend a week deciding between two S&P 500 funds that are 99.9% identical, then buy a 3x leveraged semiconductor fund on a whim the following Tuesday. The knowledge gap isn't about picking the right ticker. It's about knowing what kind of thing an ETF is in the first place. So let's fix that.
A fund that got listed like a stock
The name spells it out: Exchange-Traded Fund. A fund that trades on an exchange. To see why that mattered, remember what owning a fund used to involve. You bought a mutual fund through paperwork, your order filled once a day at whatever the fund was worth at the close, and selling meant more paperwork. Fine for 1985. Nobody misses it.
What the ETF did was take the same basket of holdings and list it, the way a company lists its shares. Apple listed a business. SPY listed a portfolio. Once a portfolio trades like a stock, everything about it gets easier: you can buy one share of it in any brokerage account, you can see its price all day, and competition among issuers has pushed the annual cost of the big index funds down to almost nothing. Vanguard's S&P 500 ETF charges 0.03% a year. That's three dollars on a $10,000 position. The fund industry of the 1990s would have considered that a typo.
Same underlying idea — a pooled basket of securities — but the ETF version trades like any listed stock.
Somebody is selling you this
One habit will make the entire ETF universe make sense: look at every fund from the issuer's side of the table. To BlackRock, Vanguard, State Street, or any of the smaller shops, an ETF is a product. Food companies put out new flavors every year, not because the world was short a flavor, but because new products get shelf space and attention. Fund companies work the same way. Robotics gets hot, robotics ETFs appear. Then AI ETFs, uranium ETFs, covered-call income ETFs, single-stock funds offering two times the daily move of Tesla. None of this is a scandal. It's merchandising. But it changes the question you should ask when a shiny new ticker crosses your screen, from "is this theme real?" to "who was this product built to be sold to, and is that person me?"
It's also worth knowing how big the winners have become, because the scale explains some strange facts about the market. Index money has grown to the point that Vanguard and BlackRock sit near the top of the shareholder list of nearly every large American company. They didn't pick Apple or NVIDIA. Everyone else kept buying funds that have to hold Apple and NVIDIA. New wrappers can even move the underlying market itself. Bond ETFs took an asset class that used to trade in institutional-size blocks and made it buyable in $100 pieces, and a wall of retail money followed. The spot bitcoin ETFs approved in early 2024 did the same thing for crypto, with the price response to match. When a wrapper makes something newly easy to buy, the buying is not a side effect. It's the event.
The four numbers to check before you buy
Say you've settled on the idea. Broad US large caps, or semiconductors, or dividend payers, whatever it is. Type the theme into a screener and you'll get a half-dozen competing tickers that look interchangeable. They aren't. Four numbers separate them, and every one is free to look up:
Four numbers, all public, separate near-identical tickers. When in doubt, the biggest fund with the lowest fee.
Net assets. How much money the fund holds tells you how many other investors already trusted it, but that's not really why it matters. Small funds trade thin, track their index loosely, and, the part nobody thinks about until it happens to them, they get closed. Issuers liquidate hundreds of underperforming ETFs every year, and when yours is one of them you get cashed out on their schedule, not yours, possibly with a tax bill attached.
Expense ratio. The annual fee, shaved off the fund's value a little each day. You never see a bill, which is exactly why people ignore it. For two funds tracking the same index, the fee is close to the only difference there is; the cheaper one wins by construction, every year, forever. This is the one number where "it's only half a percent" is the wrong instinct. Half a percent, compounded over a working lifetime, is a serious amount of your money.
Trading volume. An ETF has two prices, in a sense: what it trades for, and what its basket is worth. Professional arbitrage keeps the two glued together, but the glue needs volume. On a heavily traded fund the gap is pennies. On a sleepy one, the fund can drift to a premium or discount against its own holdings, and whichever side of that gap you're on, you paid for it.
Distributions. When the stocks inside the fund pay dividends, the fund passes the cash through. Some pay quarterly, some monthly, some once a year. If you're living on the income, the schedule matters. If you're twenty years from needing it, this is the one line on the checklist you can skim.
Passive, active, and the stuff in between
Every ETF lives somewhere on one spectrum. At one end, passive funds track an index and promise nothing else: own the S&P 500, get the S&P 500. This is Jack Bogle's index fund wearing new clothes, and his old argument still carries the flows: most managers who claim they'll beat the market don't, especially after their fees, so stop paying for the attempt. The money has agreed with him for twenty years running.
At the other end, active ETFs pay a manager to disagree with the index on purpose. Some earn it. Most, over long stretches, don't, and the fee is charged either way.
And in between sits the crowded middle: theme and sector funds. Robotics, clean energy, cybersecurity, semiconductors. Technically these are passive, since each one tracks an index. But somebody built that index by hand, and choosing the theme is itself an active bet. A robotics fund doesn't care what the S&P 500 does; it cares what robotics does. You've made a call, whether it felt like one or not.
The useful rule: match your homework to the type. Buying passive? Your homework is one line, the fee, because the index determines everything else. Buying active? Your homework is the manager, the record, and whether the outperformance survived the costs. Buying a theme? Your homework is the holdings. Open the fund page and read the top ten before your money goes in. Theme indexes are hand-assembled and routinely stuffed with names you'd never associate with the theme, and it's better to find the surprise conglomerate in position three before you own it.
The aisle where the math is against you
Now for the products that are built for traders and end up held by everyone else. Leveraged ETFs promise a multiple of the index's daily move. The catch is the word daily. The multiple resets every day, and in a choppy market that reset grinds the fund down even when the index goes nowhere. Watch the arithmetic:
Volatility decay: +20% then −20% costs a 1× fund 4%, but a 2× fund 16%. The longer and choppier the ride, the worse it gets.
Start both funds at 100. The index gains 20%, then gives back 20%: it lands at 96, down four points, the normal cost of a round trip. The 2x fund gains 40% to 140, then loses 40% of 140, which is 56 points, and lands at 84. Same market, same round trip, four times the damage. Nothing broke. That's just what daily resets do in a sideways market, and markets spend a lot of time sideways. A leveraged fund can be a legitimate short-term trade after a washout, held for days or weeks with an exit in mind. As a buy-and-hold position it's a leaking bucket with a good story.
Inverse ETFs, the ones that go up when the market goes down, get the same decay plus the ongoing cost of the derivatives inside, and they fight one more headwind: the US market rises far more often than it falls, so a standing bet against it starts every year behind. Unless you believe you can call the rare, genuine trend break, and be honest about that, they're best left alone. One more fee worth dodging while we're here: if you buy international funds, check whether you've picked a currency-hedged version. Hedging costs real money every year, and for a long-horizon investor the currency exposure is usually worth keeping rather than paying to remove. It's the same instinct as avoiding repeated currency conversion in general. Every trip through the exchange window leaves something behind.
What actually moves an ETF's price
A last honest point, because it changes how you should hold these things. A single stock has an anchor: earnings, assets, a business someone can value. A theme ETF mostly has flows. Money comes in, the fund must buy its basket. Money leaves, it must sell. Usually that's harmless plumbing. But when a hot theme pours serious inflows into a basket of small companies, the forced buying moves those very stocks up, which improves the fund's chart, which attracts more inflows, which forces more buying. The loop runs on itself, with no valuation brake anywhere in it, until the flows turn around. If you own a theme fund whose chart has gone vertical, some of what you're looking at is the crowd's money arriving, not the theme working. Issuers publish flow data; a glance at whether money is still coming in tells you more about a theme fund's near future than its holdings do. Broad index funds barely feel any of this, since their baskets are far too big for flows to distort. Which is one more quiet argument for keeping the boring stuff at the center of the portfolio.
Where this leaves you
An ETF is a fund in a stock's clothing: cheap, liquid, one click away, and sold to you by someone with a product catalog. Buy the biggest, cheapest fund that matches your actual idea. Know which kind of bet you're making, passive, active, or theme, and do that type's homework. Read the top holdings of anything with a story in its name. And leave the leveraged and inverse shelf to people who trade for a living, most of whom also leave it alone. If you want to see what the boring version really did, our simulator will replay a monthly contribution into an S&P 500 or semiconductor ETF across a decade of real history, and our piece on index funds vs individual stocks takes up when the simple basket is honestly the better buy.