Portfolio LabKnowledgeSign inCreate account

ETF basics: costs, replication and distributions

Basics7 min read · updated 9 September 2026 · 6 sources

Two ETFs on the same index are similar, but not identical. What matters most when choosing is the costs (expense ratio), the type of replication and how the fund handles its income, and much less whether an ETF was launched a few years earlier. More important than the choice between two good ETFs, however, is the index, i.e. which markets you buy in the first place.

What an ETF is

An ETF (exchange-traded fund) is an investment fund that is traded on a stock exchange and tracks an index. The fund's assets are held separately and do not belong to the fund company. If the company goes bust, they remain with the investors. Large traders (authorized participants) can at any time exchange new fund units for the securities they contain, or vice versa. If the market price deviates from the value of the securities, this exchange becomes profitable and the deviation disappears again.

Two routes to an ETF unit: the exchange and creation

When selling and in a redemption, the arrows run the other way. The securities are kept separate from the assets of the fund company and are held by an independent custodian.

According to Ben-David et al. (2017), by the end of 2016 ETFs had reached more than 10% of the market value of stocks traded on US exchanges and accounted for over 30% of trading volume. Their main advantage over active funds is cost: French (2008) estimates that investors in the US spent 0.67% of the aggregate market value every year from 1980 to 2006 searching for better returns. A passive investor would have kept that money. More in the article Active or passive.

Costs: TER and tracking difference

The expense ratio (in Europe usually called the TER, total expense ratio) is the ongoing annual cost taken directly from the fund's assets. On $10,000 with an expense ratio of 0.20%, that is $20 a year. Over a long period, the difference adds up:

What actually reaches the investor is shown by the tracking difference: the difference in return compared with the index over one year. It can be better than the expense ratio, for example thanks to securities lending income or tax advantages, or worse because of trading costs when the index changes (see tracking error).

“tracking error in index fund performance is unavoidable because of market frictions”
Frino and Gallagher (2001)

Frino and Gallagher (2001) studied index funds on the S&P 500: the index is a paper portfolio without costs, while the fund has to trade, reinvest dividends and follow index changes. On top of that come the spread and order fees when buying, which for small regular purchases are often more than a year's expense ratio.

Replication: physical or synthetic

TypeHow it worksPoints to note
physical, fullbuys all index constituentstransparent; more expensive for very broad indices
physical, optimized samplingbuys a representative selectionsomewhat higher tracking error
synthetic (swap)holds a basket of securities and swaps its return for that of the indexcounterparty risk (limited to 10% under the EU's UCITS rules)

Physical ETFs often lend out part of their securities against collateral and a fee. This lowers costs but brings a small additional risk. Synthetic ETFs bear the risk that the swap counterparty defaults; under the EU's UCITS rules, this is limited to 10% of the fund's assets, and in practice it is usually collateralized on an ongoing basis. For most investors, the type of replication matters less than the index and costs.

Accumulating or distributing?

  • Accumulating: income is reinvested within the fund. Convenient for building wealth; such share classes are mainly found among European (UCITS) ETFs.
  • Distributing: income is paid into your account, usually quarterly or annually. Useful if you live off the income; otherwise you can reinvest the payouts yourself.
  • For returns, the difference is small. In backtests, distributions must be reinvested, otherwise the distributing fund looks worse (see CAGR, total return).

For advanced readers: how ETFs affect markets

Because ETFs are tied to their securities through arbitrage, they pass waves of buying and selling on to the stocks they hold. Ben-David et al. (2018) find that stocks with a high ETF ownership share are more volatile, and not because of new information about the companies. For a long-term investor this means: even broadly diversified ETFs move together in periods of stress, and market prices can temporarily deviate from the value of the securities when arbitrage stalls, for example with bond ETFs in crises.

Sharpe (1991) describes the logic of passive investing with simple arithmetic: before costs, all active investors together earn exactly the market return, because together they hold the market. After costs, the average actively managed dollar must therefore lag behind the passive one. Studies of fund returns must include funds that have been liquidated. Malkiel (1995) shows how much survivorship bias otherwise flatters the results.

Checklist for choosing an ETF

  1. Index first: which countries, company sizes and sectors do I want? An MSCI World contains no emerging markets, whereas a FTSE All-World does (see world portfolio).
  2. Costs: compare the expense ratio and, where available, the tracking difference over several years.
  3. Size and age: a few hundred million dollars in assets and several years of history reduce the risk of closure.
  4. Tradability: tight bid-ask spreads at your broker, and no fees on purchases if you buy regularly.

How Portfolio Lab calculates it

Portfolio Lab calculates with total return prices: distributions are reinvested, and the expense ratio is already reflected in the fund's price. You enter additional costs on top of the expense ratio (for example for custody or advice) in the Backtest as a percentage per year; order fees, spread and fees for regular purchases can be specified as at a broker. Under Securities you can find ETFs by name or ISIN, with expense ratio, fund size, replication, distribution policy and domicile; the screener filters and ranks them by return and risk. More in the backtest methodology.

Frequently asked questions

How important is the expense ratio really?

For the same index, it is one of the few differences you can reliably plan for. A few hundredths of a percentage point, on the other hand, hardly matter; then tracking difference and tradability count.

Are synthetic ETFs riskier?

Under the EU's UCITS rules, the counterparty risk is limited by law, and in practice it is usually collateralized. For most investors, the difference from physical replication is small.

What happens to my money if the fund company goes bust?

The fund's assets are held separately from those of the company. They remain with the investors; the fund is wound up or transferred to another company.

What is the difference between the expense ratio and tracking difference?

The expense ratio is the reported ongoing cost. The tracking difference shows how far the fund actually lagged behind the index, including securities lending income and trading costs.

Literature

  1. Ben-David, Itzhak; Franzoni, Francesco; Moussawi, Rabih (2017): Exchange-Traded Funds. Annual Review of Financial Economics 9 (1), pp. 169–189. doi:10.1146/annurev-financial-110716-032538
  2. Ben-David, Itzhak; Franzoni, Francesco; Moussawi, Rabih (2018): Do ETFs Increase Volatility?. The Journal of Finance 73 (6), pp. 2471–2535. doi:10.1111/jofi.12727
  3. French, Kenneth R. (2008): Presidential Address: The Cost of Active Investing. The Journal of Finance 63 (4), pp. 1537–1573. doi:10.1111/j.1540-6261.2008.01368.x
  4. Frino, Alex; Gallagher, David R. (2001): Tracking S&P 500 Index Funds. The Journal of Portfolio Management 28 (1), pp. 44–55. doi:10.3905/jpm.2001.319822
  5. Malkiel, Burton G. (1995): Returns from Investing in Equity Mutual Funds 1971 to 1991. The Journal of Finance 50 (2), pp. 549–572. doi:10.1111/j.1540-6261.1995.tb04795.x
  6. Sharpe, William F. (1991): The Arithmetic of Active Management. Financial Analysts Journal 47 (1), pp. 7–9. doi:10.2469/faj.v47.n1.7

Full bibliography

This article explains a method; it is not investment or tax advice. Worked examples are simplified and rounded. All terms in the glossary.