When an index stops being an index

What is a stock market index? The CFA Institute defines it as follows:

"A security market index represents a given security market, market segment, or asset class, usually constructed as a portfolio of marketable securities known as constituent securities."

Two words stand out in that definition: represents and marketable securities.

An index represents a market, which means it can be used as a benchmark. In an investment mandate, that benchmark may reflect a market, sector, or asset class. It can also be used to assess investment manager performance, or beta. More broadly, it often plays a central role in portfolio construction, especially in long-term strategic asset allocation.

The reason an index can serve as a representative benchmark is that its constituents are marketable. In other words, investors can buy and sell them. The definition does not specify whether those securities must be public or private, which is likely a historical reflection of how markets developed. Public markets were simply much less developed when many of these definitions were first formulated.

Because an index is meant to represent a market, it naturally becomes attractive to investors who do not want, or do not need, a manager to select individual securities. That has helped fuel the rise of passive index funds. As computing power has improved, these funds have become more efficient to run, which has helped reduce fees.

With exchange-traded funds (ETFs), passive investors can buy or sell their holdings throughout the trading day, rather than dealing only once a day. It is no surprise that ETFs and passive index funds have grown rapidly over the past few years.

ETFs have also enabled many new forms of index investing, including factor strategies and bond indices. But I have started to wonder whether some indices are now being created primarily to support ETF launches. In other words, are indices increasingly serving a distribution purpose rather than acting as representative market benchmarks?

One example is the rise of personalised, or customised, indices. These have become popular among some discretionary managers and private banks. Want a bespoke version of an index for a particular market, sector, or asset class? It can be built. But the key motivation is often to make the index investable for end clients.

To do that, the index needs rules. Those rules allow it to be calculated regularly and provide a framework for how the strategy is run. That may be necessary, but the investment implications of those rules may not be obvious on day one. The creators of the S&P 500, for example, could not have anticipated that it would one day be dominated by mega-cap technology and AI companies. That kind of outcome can create disappointment later on, and may be less likely if the client had instead appointed a dedicated active manager with discretion.

Recently, I listened to a pitch from an index provider on an ETF linked to autocalls. The ETF’s underlying is an index, call it index 1. That index tracks the performance of a large number of autocall strategies, traded via total return swaps between the ETF and investment banks. Those autocall strategies are themselves linked to another index, index 2, which applies volatility control and a set of customised construction rules.

That sounds complicated, because it is. A customised index with volatility control is already highly specialised. Yet another index is then built on top of that structure, based on a large number of autocall strategies linked to index 2, simply so index 1 can be packaged in an ETF.

The presenter argued that the autocall strategy is quantitative and rule-based, not discretionary, and that the ETF makes the strategy available to retail investors in a form that can be bought or sold at any time during the trading day. I am not fully convinced. What I see instead are multiple layers of complexity, with hidden costs along the way.

The ETF reportedly reached more than US$1 billion in assets under management within a few months of launch. Perhaps I am missing something. Or are we beginning to see a new phenomenon of index washing?

James Chu, CFA

Head of Investment Solutions