What is an ETF tracking difference and why does it matter?

Tom Bailey explains the main reasons why an ETF will have a tracking difference.

Most exchange-traded funds (ETFs) have one seemingly simple task: track the index they are supposed to. The investor buys an ETF hoping to gain the returns of the underlying index, be it the FTSE 100, S&P 500 or something more exotic.

However, ETFs will almost never provide the exact same return as the index they are tracking. As a result, investors in an ETF will receive a slightly lower return than the underlying index. This is known as the ‘tracking difference’ and is an important concept for any ETF investor to understand.

• The cheapest way to buy high-flying FTSE 250 index

• Want to invest ethically? ii’s ACE 40 list of ethical investments can help

• Your chance to win £1,000: take part in the Great British Retirement Survey

The tracking difference of ETFs varies. Generally, an investor wants as small a tracking difference as possible. Here are the main reasons why an ETF will have a tracking difference.

Tracking difference is not the same as tracking error. Tracking error is a metric that assess volatility of performance between the ETF and the index it is aiming to replicate.

Fees and charges

The fee an ETF charges is usually expressed as either its ‘ongoing charge’ or the ‘total expense ratio’ and can be found on the ETF’s factsheet.

The fee the investor pays will create a difference between the return the investor receives and the return the underlying index suggests they would receive. Usually, these numbers are rather small, as ETFs are famous for having low fees, with some charging as little as 0.04%. Nevertheless, it is still a fee. If an ETF tracking the FTSE 100 charges 0.1%, your returns will be 0.1% lower than that of the index.

For more niche ETFs, these fees can reach as high as 0.8%. The tracking difference this creates can add up over the years, so it is something investors should keep an eye on.

Transaction costs

Another reason for tracking difference is the cost of rebalancing the portfolio. Indices are often rebalanced at set intervals. The FTSE 100, for example, is rebalanced every quarter. Companies that have fallen in market capitalisation value are removed from the index, while companies that have grown large enough are added. The index can also change due to companies delisting from the stock market or being taken over by another firm.

When these changes happen, the ETF provider has to buy or sell the relevant shares in the portfolio to reflect the index. This is called ‘rebalancing’ and is vital to ensure that the ETF’s portfolio is still tracking the index. However, this incurs costs and so creates a difference between the portfolio’s return and the index’s return.

• Tom Bailey: active funds are useful – without being in your portfolio
• Active versus passive: a brief introduction

An ETF tracking the FTSE 100, or another similar large index, is likely to have fairly low transaction costs. However, ETFs tracking more niche indices may have to carry out a more extensive rebalancing, meaning a bigger tracking difference. For example, if an ETF follows an alternatively weighted index, they may need to buy and sell more stocks to rebalance their portfolio.

A good example is an equally weighted ETF. If each company in the portfolio is meant to be weighed equally, say at 1% each, as share prices move up or down, the weightings will change. As a result, the ETF will need to buy or sell portions of their holdings to get the portfolio’s weighting back to where it is meant to be.

Another example is a ‘capped’ index ETF. Some thematic ETFs have an upper limit on what any single company’s weighting can be. As a result, they will need to periodically sell some of a stock to get below this threshold.

The type of companies in an index and ETF can also increase transaction costs. Generally, smaller companies are less liquid, meaning they are more costly to sell. Therefore, an index tracking a basket of less liquid stocks will end up having a higher tracking difference due to the increased transaction costs.

Related to this is a potential time drag. When an index rebalances, it is just data being changed. But for the ETF to copy this, the portfolio manager will have to buy and sell shares in the portfolio. In the time it takes to do this, the price of the shares in question may have gone up or down. This will again create a difference between the returns of the index and the ETF.

Cash drag

ETFs sometimes have uninvested cash in a portfolio. This is usually because of dividends. If stocks in the portfolio pay dividends, ETFs don’t automatically hand those over to investors. Instead, these dividends are paid out at periodic intervals, such as every quarter. The time between these dividends being paid and being distributed to investors can create a difference in performance.

Alternatively, the dividend payments can be reinvested into the ETF (known as accumulated distribution). However, this also creates a tracking difference because reinvesting them means creating more transaction costs.

Stock lending

One way ETF providers generate income is by lending out the stocks in their portfolio. Stocks are usually lent to brokers or other market participants to be used for ‘going short’. Shorting a stock entails lending a stock and selling it, in anticipation of the stock price falling and buying it back at a lower price and returning it to whoever it was borrowed from. Borrowing a stock costs money – so for ETF providers, lending some of their stocks is an easy way to make money.

The income generated can be used to lower the fees the ETF charges to investors. Therefore, stock lending can sometimes result in the ETF having a lower tracking difference. It should be noted that only physically replicated ETFs can do this and not all of them do.