Exchange-Traded Fund

An ETF operates in two distinct markets. In the secondary market, investors buy and sell existing shares from one another on an exchange, so the transaction affects no one but the two counterparties and the fund's asset base is untouched. In the primary market, only authorised participants deal directly with the fund, exchanging a defined basket of securities (or cash) for a creation unit of ETF shares, or delivering shares back to be redeemed.

This two-tier structure drives the arbitrage. If exchange demand pushes the share price above the value of the underlying basket, an authorised participant can buy the basket, deliver it for new shares and sell them, capturing the spread and expanding supply until the gap closes. A discount reverses the trade. The mechanism is a profit incentive, not a guarantee: it works only while the underlying basket can be traded and priced at reasonable cost.

Total cost of ETF ownership goes well beyond the stated expense ratio. Round-trip trading costs include the bid-ask spread and any brokerage commission, and these are borne only by investors who trade, unlike a fund's ongoing fee, which accrues to all holders over the whole holding period. Short-term traders are therefore dominated by spread costs, while long-term holders are dominated by the expense ratio.

The premium or discount is computed as the ETF market price minus net asset value per share, divided by net asset value, and it can be a real cost if you buy at a premium and sell at a discount. Separately, tracking difference is the realised gap between fund return and index return over a period, driven by fees, cash drag, sampling, dividend reinvestment timing, securities lending revenue and withholding tax; tracking error is the volatility of that gap. Candidates should keep the two terms distinct.

In practice, ETFs serve as core beta exposure in a core-satellite structure, as tactical instruments for tilting toward a sector, factor, country or duration bucket, and as tools for cash equitisation so that inflows are not left uninvested. They also support transition management, letting a plan hold market exposure while a new manager is appointed, and they can be used for liquidity management within a fixed-income sleeve where individual bonds trade thinly.

Structure matters to the risk assessment. Physically replicating funds hold the securities; synthetic products rely on swaps and therefore introduce counterparty exposure, and exchange-traded notes are unsecured debt of an issuer. Securities lending inside a fund adds revenue but also collateral and borrower risk. Fund closure, holdings-based mismatch between label and actual exposure, and concentrated or illiquid underlying markets round out the risk set an analyst should screen.

You should be able to trace the arbitrage step by step in both directions and state the conditions under which it weakens: illiquid or closed underlying markets, stale reference prices, wide creation costs, or restrictions on authorised participant activity. Expect to explain why fixed-income, emerging-market and international equity ETFs display wider and more persistent deviations than large-cap domestic equity funds, and why time-zone differences alone can produce apparent premiums.

You should also be able to compute a premium or discount from a price and a NAV, decompose total cost of ownership between one-off trading costs and recurring fees for a stated holding period, distinguish tracking difference from tracking error, and identify the tax advantage of in-kind redemption in reducing realised capital gains distributions relative to a comparable open-end fund.