WhitmanTrading

How to Buy an ETF

To buy an exchange-traded fund, choose the index first, then compare expense ratios among funds tracking it. Check the spread during normal hours, because it is a second fee. Place a limit order rather than a market order, and avoid the opening minutes when spreads are widest.

An exchange-traded fund is a fund with a share’s trading behaviour attached. Most of what is specific to buying one comes from that second half rather than the first.

Before you start

The fund’s total expense ratio from its own factsheet. The complete figure, not the headline. It is the annual charge and the number most comparisons stop at.

The spread on it during normal trading hours. The gap between bid and offer. For anyone buying regularly this is a recurring cost and it is invisible on every factsheet.

A limit price decided before the order is placed. The price you are willing to pay. Deciding it in advance is what stops the order from accepting whatever the book happens to hold.

The steps

1. Choose the index before the fund

A candlestick chart with a broad index path marked.
The index inside it is still the decision that matters. Illustrative chart - not real market data.

What you own is decided here. Two funds on the same index hold the same companies in the same proportions, so this choice does far more than the next one.

2. Understand what the wrapper changes

A candlestick series representing a fund that trades continuously.
A fund that trades like a share, all day, at a quoted price. Illustrative chart - not real market data.

It trades continuously at a quoted price rather than once daily at a calculated one. That is convenient and it introduces a spread, which a traditional fund does not have.

3. Compare expense ratios among funds on that index

The first half of the price series representing an annual charge.
The expense ratio is the one number you control. Illustrative chart - not real market data.

Sort by it and take the cheapest that passes the remaining checks. On this site’s arithmetic, 75 basis points removes 20.2% of a thirty-year pot.

4. Check the spread before committing

A candlestick chart with a volume histogram beneath it.
And the spread is a second fee the factsheet does not show. Illustrative chart - not real market data.

Look at the bid and offer during normal hours. A wide spread on a fund you buy monthly costs more over a year than a small difference in expense ratio.

5. Know that the price can drift from the holdings

The second half of the price series showing price diverging from value.
The price can drift from what the holdings are worth. Illustrative chart - not real market data.

The quoted price is set by supply and demand and usually sits very close to the value of the holdings. Usually is not always, and the gaps are largest when markets are disorderly.

6. Use a limit order

Price bars with a limit price drawn as a horizontal line.
Use a limit order, because a market order accepts any price. Illustrative chart - not real market data.

Name the price you will pay. A market order takes whatever is available, which in a thin moment can be materially worse than the price on screen.

7. Avoid the opening minutes

A candlestick chart with session boundaries marked.
Avoid the first minutes, when the spread is widest. Illustrative chart - not real market data.

Spreads are widest at the open while underlying prices are still settling. Waiting costs nothing and removes the worst pricing of the day.

8. Check the fund is large enough to survive

A section of the price series representing fund scale.
A small fund can close, and closing is a taxable event. Illustrative chart - not real market data.

Very small funds get closed. A closure forces a sale on the provider’s timetable rather than yours, which can create a tax bill you did not choose.

How to tell it worked

Review your first 12 months of purchases.

Count how many used a limit order. 12 out of 12 if you buy monthly. A market order is the one avoidable way to pay materially more than the screen price, and it takes no extra effort to avoid.

A candlestick chart annotated with the round-trip cost.
Switching between them costs 2% of a bar each time. Illustrative chart - not real market data.

Add up the spread paid across those 12 purchases and compare it to the annual expense ratio. For a regular buyer of a widely held fund the two are often comparable, and almost nobody measures the first one.

Then check your fund’s 12-month return against the index it tracks. A persistent gap wider than the expense ratio means the tracking is poor and the cheap headline fee was not the real cost.

Why the spread is the fee nobody quotes

A decades-long candlestick view with no activity marked.
The holding period is what the whole thing was for. Illustrative chart - not real market data.

A factsheet reports the expense ratio and never reports the spread. One is charged annually against the fund’s assets; the other is charged on every transaction and appears nowhere in the fund’s documentation because it belongs to the market rather than to the provider.

For a buy-and-hold investor the expense ratio dominates. Over thirty years, on this site’s arithmetic, 5 basis points costs 1.5% of the pot and 150 costs 36.5%. A one-off spread is negligible against that.

For a monthly buyer the arithmetic shifts. Twelve spreads a year, every year, is a recurring cost that compounds in the same direction as the fee — which is why step four exists and why it is skipped almost universally.

The original data

Of the 24,971 unique videos in research/search-study-corpus.jsonl, 90 have an instruction-shaped title mentioning ETFs, at a median of 15,114 views across 76 channels, with a maximum of 1,503,238. Index funds appear in 26 at a much higher median of 88,014. The counts come from site/rank_howto.py.

A falling stretch of the price series cut short at the decision bar.
It is trading below its holdings. Buy the discount? Illustrative chart - not real market data.

90 ETF videos at 15,114 against 26 index-fund videos at 88,014. Three and a half times the supply for a sixth of the audience per video. The technically precise term loses badly to the plain one, which is the same pattern this corpus shows almost everywhere.

The answer to the question on that chart is that a discount to holdings is usually a warning rather than an opportunity. Persistent gaps appear when the underlying market is disorderly or the holdings are hard to price — which means the discount is telling you the fund’s own valuation is uncertain, not that you are being offered something cheaply.

When it fails

A sideways, range-bound candlestick series.
In a flat decade the fee is the only thing that moved. Illustrative chart - not real market data.

In a flat decade the charges are the only certain event. The fund returns the index minus its costs, so when the index goes nowhere the expense ratio and the accumulated spreads are the entire measurable outcome. It is the strongest argument for steps three and four, and it only becomes obvious in retrospect — which is exactly when it is too late to have chosen differently.

The second failure is a market order. It accepts any price the book offers.

A third is buying at the open. The spread is at its widest and the underlying is unsettled.

A fourth is comparing expense ratios while ignoring spreads. For a regular buyer they are comparable costs.

A fifth is a very small fund. Closure forces a sale on somebody else’s timetable.

And a sixth is switching for a few basis points. A round trip costs 2% of a median bar’s range, which usually exceeds the saving.

ETF investing covers the structure and how creation and redemption keep the price honest. Index funds is the traditional wrapper and where the fee comparison started. And tracking error is the measurement behind a persistent gap to the index.

What I actually do

The habit worth building is checking the spread before the fee. Two funds tracking the same index can have near-identical expense ratios and very different spreads, and if you are buying monthly the spread is charged every time while the expense ratio is charged once a year. For a regular buyer the second fee is often the larger one.

— Michael Whitman

This page is educational, not financial advice. Test every idea on your own charts before risking money.