How to Choose an Index Fund: Index First, Fee Second
Choosing an index fund means deciding which index you want to own first, then picking the fund that tracks it most cheaply and most closely. The index decides what you own; the fee, compounded over the whole holding period, decides how much of its return you keep.
This page is about choosing between index funds. For what an index fund is, and why its low cost matters, see What is an index fund?. For the steps of buying one as an exchange-traded fund, see How to invest in ETFs.
How it works
Choosing an index fund is four checks, in this order. Which index. How that index weights its members. What the fee costs over the whole time you will hold. And how closely and cheaply the fund actually tracks.
The order is the point. The fee is the easy number to compare, so it usually gets compared first. The index is the one that decides what you own, so it has to come first.
Check 1: which index
One decision matters more than the fee and gets far less attention: which index. A fund tracking a domestic large-company index and one tracking a global all-cap index are different investments with the same structure, and the choice between them affects the outcome far more than five basis points either way.
The second-order questions follow from that. How many companies, weighted how, in which countries, and whether small companies are included at all. A cheap fund tracking the wrong index for your purpose is a worse outcome than an expensive one tracking the right one — which is the opposite of how the choice is usually presented.
And the index itself is a set of rules somebody wrote. Inclusion criteria, rebalancing dates and weighting caps are all decisions made by a committee at an index provider — so “passive” describes your behavior rather than the product’s.
Check 2: how the index weights
Almost every major index weights by market capitalization. The largest company gets the largest holding, which means the fund automatically holds more of whatever has already risen.
Five hundred holdings is not five hundred equal bets. A capitalization-weighted fund can have a third of its value in ten companies, so the diversification implied by the name overstates what is there. Read the top ten holdings and their combined weight before you buy.
Check 3: what the fee costs over the whole holding
A basis point is one hundredth of a percentage point. At five of them, compounding the fee alone across thirty years removes 1.5% of the pot.
At one and a half percent it removes 36.5%. Same period, same arithmetic — more than a third of the final amount, before any question of whether the manager added value.
That is why the annual figure is misleading. One and a half percent sounds small against a market that might return seven; over thirty years the two compound against each other and the fee takes a share of every year’s growth, not just of the original sum.
The intermediate figures fill in the picture. Twenty basis points removes 5.8% of a thirty-year pot and seventy-five removes 20.2%. The relationship is close to linear at these levels, which makes the comparison between two funds straightforward: multiply the fee difference by roughly twenty-four to get the share of the final pot at stake.
Check 4: tracking error and the spread
Tracking error is the gap between fund and index. It comes from fees, from cash held for redemptions, and from sampling where a fund holds a subset rather than everything — and it is worth checking alongside the headline fee.
Volume matters for the exchange-traded versions. A thinly traded tracker has a wide spread, which is a cost that never appears in the published fee.
In practice
The fund you choose is built for decades. Its advantages — low cost, no manager risk, no selection error — all accumulate slowly and are invisible over a year.
Trading it frequently defeats the point. Beginners tend to count only the annual fee; every switch pays a spread as well. On a $10,000 holding a fee of five basis points is $5 a year, so a single round trip through a 0.1% spread costs twice the whole year’s fee.
So choosing well once beats switching often. A fund picked on the four checks above rarely needs replacing, and each replacement pays the spread again.
A stop sits awkwardly on a decades-long holding. The whole proposition is continuing to hold through declines, and an automatic exit contradicts it.
No choice of fund protects you in a fall. Holding everything means holding everything on the way down, and a gap affects the whole index at once.
What a good choice does not fix
It does not make the fund diversified by holding count. Weighting decides that.
It does not make it free. It is cheap, and the fee still compounds.
It does not make it safe. It falls with the market, in full.
And it does not turn it into a strategy. It is a low-cost way to own the market you picked.
How this connects
The cost line has a formal name: the expense ratio. It sits beside tracking error on every factsheet, and providers such as Vanguard and BlackRock compete mostly on it.
The best-known example is SPY, which tracks the largest US companies, while funds following the Nasdaq or the FTSE apply the same method to different lists, which is why the choice of index matters.
Portfolios are usually built from a few of them. A lazy portfolio holds domestic shares, international diversification and bonds, and a target date fund does the same inside one product with an asset allocation that shifts over time.
The philosophy has a following. Bogleheads argue for passive investing because active management must beat market beta after costs, and most fund managers do not deliver alpha net of fees.
Factor investing sits in between, tracking rules-based indexes tilted toward value or size, often sold as smart beta at a higher cost.
Contributions usually arrive through dollar cost averaging, from a workplace retirement plan, a Roth IRA or a stocks and shares ISA, and dividend reinvestment keeps total return compounding.
The main behavioral risk is market timing. Buy and hold is what the structure rewards, while selling in a decline turns concentration risk into a realized loss.
A robo advisor automates all of this, including tax loss harvesting in taxable accounts.
When it fails
It delivers the index, including when the index does nothing. A flat decade produces a flat decade minus the fee, and there is no mechanism in the product to improve on that.
The second failure is a concentrated index sold as broad. A fund holding thousands of companies with a third of its value in ten is a concentrated position with a reassuring name. Check 2 catches it.
A third is buying several that overlap. Three broad funds frequently hold the same largest companies, which adds cost without adding diversification.
A fourth is comparing funds on fee alone. Tracking error and spread both matter and neither is in the headline number, and the index matters more than all three.
And a fifth is trading it. The advantage is entirely in holding cheaply for a long time.
The original data
Of the 24,971 trading and investing videos measured for this site, 132 have “index funds” in the title, from 87 channels, at a median of 69,951 views.
The question: Down 20% and the plan said hold. Sell?
Those figures contain no return assumption at all — they are the fee compounding against itself, so the real cost on a growing pot is larger still. The comparison worth running before choosing between two funds is the difference in fee multiplied by roughly twenty-four, which converts an annual percentage nobody can feel into the share of the final amount it represents. It is the highest-value minute available in the whole subject, and it only has to be spent once.
Related
What is an index fund? explains the vehicle you are choosing between. How to invest in ETFs is the buying process when the fund is exchange-traded. Mutual funds is the actively managed alternative. And passive income covers what these holdings are usually bought for.
I look at what a fee costs over the whole holding period rather than in a single year. The difference between five basis points and one and a half percent is more than a third of the final pot, and it comes from a decision made once and rarely revisited.
— Michael Whitman
This page is educational, not financial advice. Test every idea on your own charts before risking money.