Exploring different ETF strategies: full replication, sampling, and swap-based

Only if you are Star Wars fan you might get why I chose this cover image. If you are not, let me spoil it for you: clones! Or in this case replication (strategies). In this article I'll walk you through the main ETF replication strategies. But first let's do a quick recap on ETFs.

What are ETFs?

An ETF, or Exchange-Traded Fund, is a type of investment fund that holds a collection of assets such as stocks, bonds, or commodities. They normally do this by copying (or replicating) an index, that is a list of companies that match specific criteria. Do you want to invest only in American companies? There's an ETF for that. Do you want to invest only in healthcare-related companies in Europe? You got it.

ETFs have revolutionized the investment world. From its humble beginnings in the early 1990s, the ETF market has exploded, boasting over 7,000 different ETFs and managing trillions of dollars in assets worldwide.

You can also read my other article on ETFs.

Why ETFs are popular

ETFs offer numerous benefits: lower costs compared to mutual funds, tax efficiency, high liquidity, and easy access to a diverse array of assets.

Fun Fact: the largest ETF in the world, the SPDR S&P 500 ETF (SPY), has over $500 billion in assets under management, making it larger than many mutual funds combined.

When it comes to ETFs, the strategy behind how they replicate their underlying index is crucial. The index they replicate is essentially a benchmark — a collection of securities (like stocks or bonds) that the ETF is designed to track. The index serves as a model portfolio that the ETF aims to mirror in performance, composition, and sometimes in sector weighting.

What are the main indexes?

There are indexes for every flavor possible, with new ones constantly added. Some common types that are replicated by ETFs include:

Stock market indices

• S&P 500: this is one of the most well-known indices, comprising 500 of the largest publicly traded companies in the U.S. An ETF like the SPDR S&P 500 ETF (ticker: SPY) tracks this index.
• Dow Jones Industrial Average (DJIA): the DJIA includes 30 large U.S. companies and is often tracked by ETFs like the SPDR Dow Jones Industrial Average ETF (DIA).
• NASDAQ-100: this index includes 100 of the largest non-financial companies listed on the NASDAQ stock exchange, tracked by ETFs like the Invesco QQQ ETF (QQQ).

International indices:

• MSCI EAFE: represents large and mid-cap companies in developed markets outside of the U.S. and Canada. ETFs like the iShares MSCI EAFE ETF (EFA) track this index.
• FTSE 100: comprises the 100 largest companies on the London Stock Exchange, tracked by ETFs like the iShares FTSE 100 ETF (ISF).

Sector indices:

• MSCI World Healthcare Index: focuses on global healthcare companies, tracked by ETFs like the iShares Global Healthcare ETF (IXJ).
• S&P Global Clean Energy Index: tracks companies involved in clean energy production, tracked by ETFs like the iShares Global Clean Energy ETF (ICLN).

Bond indices:

• Bloomberg Barclays U.S. Aggregate Bond Index: this is a broad index covering U.S. investment-grade bonds, tracked by ETFs like the iShares Core U.S. Aggregate Bond ETF (AGG).

Commodity indices:

• S&P GSCI: a broad-based commodity index covering physical commodities, tracked by ETFs like the iShares S&P GSCI Commodity-Indexed Trust (GSG).

When it comes to replication strategies for ETFs, there are three main options: Full Replication, Sampling, and Swap-Based. Understanding these strategies will help you make informed investment choices and hopefully maximize your returns.

Tracking error

Before we jump into the replication strategies though, it's good to get the concept of tracking error as no matter the replication strategy, the tracking error or tracking error volatility (TEV) is a key metric to look at when comparing funds. This metric is the extent to which the return of a fund differs from its benchmark over a period of time. In case of passive funds such as an ETF tracking the S&P 500 index, we are expecting the tracking error to be as close as possible to zero. If on the other hand the fund is actively managed, we do expect the management of the fund to deviate slightly, hopefully in ways that will increase the returns compared to the fund. Even in this case though, anything above 7% would mean that the fund is doing something very different from the benchmark that I wanted to mirror.

When checking out this metric, make sure to first read the fund composition and goal on the official fund prospect. Sometimes it can happen that the benchmark used for reference on a platform doesn't really fit what is stated (as goal) on the fund prospect. Imagine if on a fund prospect you read this:

This Fund seeks to track the performance of the Index, a widely recognised benchmark of U.S. stock market performance that is comprised of the stocks of large U.S. companies.

But then on the platform/report of your choice, they would measure the tracking error volatility using an Emerging market index as benchmark. This would definitely be not correct. While this error is not common for big and mature funds, it can happen for those smaller, maybe even actively managed funds.

Full Replication strategy

Full replication (or Physical replication) is the most straightforward ETF strategy. It involves the ETF holding every single security in the index it tracks, in the exact proportion as the index. If the index has 500 stocks, so does the ETF, mirroring the index's performance as closely as possible.

Advantages:

• High accuracy: full replication ensures the ETF tracks the index very closely, minimizing tracking error.
• Simplicity: this strategy is easy to understand since the ETF's performance directly reflects the index.

Disadvantages:

• Higher costs: buying and holding all the securities in the index can be costly, especially for indexes with many components.
• Feasibility: full replication is less practical for indexes with a large number of constituents or those including illiquid securities.

Example
A popular example of an ETF using full replication is the SPDR S&P 500 ETF (SPY). Launched in 1993, SPY was the first ETF in the U.S. and remains one of the most popular. By holding all 500 stocks of the S&P 500 index, SPY provides investors with broad exposure to the U.S. stock market.

The image above shows the top holding of the SPDR SPY fund. As you might notice, even with a full replication strategy there might be small differences between the fund and the actual index. This is due to a few factors such as:

• Rebalancing lag: the ETF may not immediately adjust its holdings to reflect changes in the index, leading to small discrepancies in weights due to logistical reasons.
• Management fees: the cost of running the fund (expense ratio) is deducted from the fund's assets. Over time, this can slightly reduce the value of the fund relative to the index, affecting the weights of individual holdings.
• Dividend reinvestment: the S&P 500 index typically assumes that dividends are reinvested immediately in the index while SPY collects dividends and pays them out to shareholders quarterly. During the period between collecting and distributing dividends, the ETF might hold those funds as cash or invest them temporarily, leading to small weight differences.

Sampling strategy

With Sampling, the ETF holds a representative sample of securities from the index, rather than all of them. This subset of securities is selected to mimic the performance and risk characteristics of the entire index. It's more commonly used for commodities ETFs.

Advantages:

• Lower Costs: by holding fewer securities, sampling reduces transaction and management costs.
• Practicality: ideal for large and complex indexes, where full replication would be impractical.

Disadvantages:

• Tracking Error: sampling can lead to a higher tracking error compared to full replication.
• Selection risk: if the sample is not representative, the ETF may underperform the index.

Swap-Based strategy

A Swap ETF is a type of ETF that uses financial contracts called "swaps" to track the performance of an index or asset, rather than directly holding the underlying stocks or bonds. Sometimes they are also called Synthetic ETFs.

In a traditional ETF, the fund directly owns the stocks or bonds that make up the index it tracks (like the S&P 500), but in a Swap ETF the fund doesn’t actually own the underlying assets. Instead, it enters into a contract with another party (usually a bank) to "swap" returns. The ETF gets the performance of the index from the bank, in exchange for paying the bank a fee or a fixed return.

Advantages:

• Low Tracking Error: swap agreements can precisely match the index return, reducing tracking error.
• Cost-effective: ideal for accessing hard-to-reach markets or complex indexes without the cost of holding the underlying securities.
• Tax efficiency: In some cases, swap ETFs may have tax advantages, as they can reduce capital gains taxes compared to traditional ETFs.

Disadvantages:

• Counterparty risk: the ETF is exposed to the risk that the counterparty may default on the swap agreement.
• Complexity: Swap ETFs are more complex than traditional ETFs, which can make it harder to understand what’s happening behind the scenes.
• Regulatory and transparency issues: swaps can be complex and less transparent, raising potential regulatory concerns.

Example

The red bar below shows the performance of the last 3 years for the Invesco S&P 500 UCITS ETF (IE00B3YCGJ38) that uses a swap-based strategy to provide exposure to the S&P 500 index. For comparison, the blue bar is the iShares Core S&P 500 UCITS ETF (Acc) (IE00B5BMR087), an ETF that tracks the same S&P 500 index however using a full replication strategy.

As you can see the difference in returns is minimal (and this can be due to a mix of factors such as fees, transaction costs, tracking error, etc.).

Final Thoughts

Recent years have seen a rise in smart beta ETFs, which use alternative strategies to weight the components of an index. Additionally, regulatory changes have influenced the use of swap-based strategies, particularly in Europe. Advancements in technology, such as AI and machine learning, are likely to influence how ETFs are managed and traded. These technologies can optimize portfolio management, reduce costs, and enhance performance. As investors become more educated and demand more sophisticated products, ETF providers are likely to develop new and innovative strategies.

When choosing an ETF strategy, consider your investment goals and risk tolerance. Full replication offers high accuracy but at a higher cost, sampling strikes a balance between cost and tracking error, and swap-based strategies provide precise index tracking with some counterparty risk.

Happy investing!