Why ETFs are Dangerous for Private Investors and the Global Financial System

by Andrew Alexander
The views represented in this article are solely those of the author and may not be construed as in any way representative of the views or policies of Oxford Royale Summer Schools.


Imperfect knowledge is the cornerstone of the capitalist system.

You should also read…

Entrepreneurs operating untested models and willing to gamble everything on a hunch drive productivity and living standards, while the rise and fall of nations is a consequence of the decisions they make on resource allocation – an area in which the right mixture at the right time is a rare beast, not least given the tendency of governments and corporations everywhere to drive with their eyes fixed on the rear view mirror. Imperfect knowledge, however, is a different thing to ignorance. Whereas imperfect knowledge straddles Donald Rumsfeld’s categories of known-unknowns and unknown-unknowns, calling us to make judgements based on our understanding of probability rather than the illusion of certainty, ignorance amounts to a refusal to gather the information required to make a judgement on what should reasonably be seen as a known-known, a category in which it is perfectly possible to have complete knowledge.

Widespread misunderstanding of financial products burns investors and leads to crashes — like 2008

Image shows a house with a large red and white 'foreclosure' sign in the foreground.
The dangers of collateralised housing debt products should have been evident to speculators.

This has a particular relevance to investment. Let’s take the case of the collateralised housing debt products which did so much to accelerate the global crash in 2007/8. Any investment offering a return above zero comes with a degree of risk attached. In this case, most investors would have understood that the health of the portfolio of housing loans underlying the investment would have an impact on its value and their returns. This was a known-unknown, a form of imperfect knowledge. Speculators buying housing CDOs did not know the shape of the housing loan market at expiry but had decided, with this risk in plain view, that this represented, on the balance of probabilities, a sound investment. It was not these decisions, based on imperfect knowledge, which was to cause such pain. Instead, it was the element of the investment process which was driven by ignorance. It was perfectly possible to see how CDOs were put together and to model out the pricing dynamics of each tranche of debt when applied to individual houses. It was perfectly possible to see that logically taking a large number of bad risks and packaging them together did not render the subsequent creation a better risk and therefore deserving of a higher price. Investors chose not to understand the mechanics of the product they were buying, preferring to take a view on the dynamics of the market it operated in. When they did so, they exposed themselves to the disaster which eventually befell them.
Wilful ignorance of the mechanics of financial products is one of the great frustrations of our age. The British consumer has been the most enthusiastic proponent of this approach to money management in the world. Whether being sold endowment mortgages, interest rate swaps, credit card insurance or anything else besides – the British consumer’s investment approach is a combination of a failure of curiosity when it comes to the mechanics of what it is buying (a failure which grows greater in proportion to the financial risk assumed) and then a steadfast belief that it is the government’s job to retrospectively reallocate responsibility for basic due diligence from the buyer to the vending financial institution. Seller Beware.

The ETF Boom       

All of which makes me distinctly nervous about the explosion in the Exchange Traded Fund market in recent years. If wilful ignorance of product mechanics is a key factor in investors misallocating funds and failing to understand prices, and if this process of mis-allocation and mis-pricing is in itself a threat to the global economy when it occurs at scale and on a global basis (both reasonable inferences from the past seven years), then any reasonable survey of the global financial marketplace would point to the ETF market as potential source of significant disruption and turmoil in the years ahead.

Image shows the Sao Paulo Stock Exchange. Two men sit watching the huge wall of numbers.
Some sources are expecting the US ETF industry to grow to $15.5tn by 2023.

Before exploring why this should be the case, it is worth taking a survey of the incredible growth of the ETF market in a relatively short space of time, as well as looking at the rationale for their creation and subsequent popularity. In this context, it would be useful to define in broad terms what constitutes an ETF – although there are significant differentiations between types of ETF. Broadly speaking, an ETF is a security tradable on an open exchange which seeks to track, as an index fund would, the price movements in an index, commodity market, or group of assets. Theoretically, this model combines the exposure to a broad asset class which an index fund would offer with the wider array of trading options which an exchanged based security would give, such as the option to short sell. The first ETF was launched in 1993, the S&P 500 tracking SPDR, which the Wall Street Journal estimates still accounts for around 16% of all ETF assets on its own. From one product in 1993, the ETF market grew to over 1,400 by the start of 2012 and continues to expand. A market with only $79 billion of assets under management in 2000 has now expanded to $2.4 trillion, a figure expected to rise to $3.6 trillion by 2016.

Image shows two bars of gold, reflected by mirrors, so that at first glance it looks as if there are eight bars or more.
The old maxim comes to mind: if something looks too good to be true, it usually is.

It is not hard to see why. Sold packaged as a cheap way to track complicated markets, and one with instant buy-in and exit points, unlike the fund industry, financial consumers have taken ETFs to be the best way of gaining market exposure free from hassle. Take gold, for instance, whereas previously a person wishing to speculate that the gold price would increase would have needed to buy and store physical gold (downsides: transaction costs, storage costs, security risks, illiquidity) or purchase gold stocks as a proxy (downsides: management risk, political risk, inability to increase production, ties to futures contracts delivering a lower rate than spot price), they are now able to trade the spot price thanks to ETFs. It is a simple argument, but it hides some worrisome truths. ETFs, as should be clear to anybody who does the basic due diligence of seeing how they are constructed, neither track the underlying market accurately nor do they offer the security of holding a stock or a physical commodity. In fact, they require significant financial sophistication to understand and their existence creates an unnecessary danger for the world economy.

The Flaws in the ETF Model

This is an article which is largely addressed towards assumptions, lazily made, which will over time occasion significant financial losses for investors who hold them. The compound effect of a huge number of individual investors, encouraged by a monetary environment which penalises caution, making the same errors in risk-weight over a sustained period is a highly destabilising one. We have already seen the effect which algorithm-driven ETFs can have on individual markets – by one estimation, 68% of the trades contributing to the 999 point Dow ‘flash crash’ were driven by ETFs vainly trying to track the market on its way down and in doing so, sustaining the momentum of the downturn. Investors at present are vastly underestimating the complexity and potential for underperformance from ETFs. There are two main reasons I would suggest for this, and both turn upon the extent to which variations in what underlies an ETF are wilfully ignored or misunderstood by investors.

1. Synthetic ETFs do not give exposure to the asset investors believe they do

Image shows the prices of shares on a stock exchange board.
Swaps and derivatives give investors less security than physical replication.

Investors are used to the trust/fund model of financial diversification. Traditionally, if you were to buy, say, a Japanese equity fund, then the underlying assets of that fund would be….Japanese equities. If the fund needed to liquidate your holding, and your exit was not balanced by a new purchase, it would sell shares at the market value. Broadly, you could track the Nikkei and know roughly how your basket of shares could be doing. Some ETFs use this model, known as physical replication – holding the asset which they claim to be tracking. Many do not. One of the advantages of the ETF model over the funds model is lower management charges, yet the constant updating of, say, a basket of Japanese equities requires a significant architecture and constant trading if it is to achieve full replication (some aim for partial replication by selecting a few stocks from an index to trade – this may result in very different returns, a problem elsewhere also, see below). In response to this problem, the ETF industry has developed an alternative approach: synthetic ETFs use swaps and derivatives to target a performance which tracks the market on a daily basis. There are obvious problems with this model. One, the retail investor often does not understand what the product they have bought is, instead they assume it to be the result of physical replication. Two, unlike a commodity or a stock, the securities used by ETFs are hard to price. This means that realising the expected market value in the event of a large sell off may be impossible – the investor is exchanging cash for a promise, rather than tangible ownership of an asset, and the value of promises dips quicker than the value of assets in a panicked market, creating a liquidity risk. Finally, because the market is more complex, there is a greater opportunity for misallocation of funds – as has allegedly been the case with the relatively junior UBS trader who ran up a $2bn loss through derivatives trades for ETFs. Given the above, it would appear to be self-evident that synthetic ETFs should trade at a discount to physical replication ETFs once risk is factored into pricing, and I believe that they would had investors made the effort to make themselves sufficiently appraised of what exactly they were buying.

2. Failure to track markets

Image shows a one dollar bill on fire.
The potential dangers of ETFs have been seriously underestimated.

Leveraged ETFs attempt to amplify the performance of the market which they track by a specific amount, for instance double (as is the case with the popular ProShares Ultra S&P 500 ETF which promises a 2% change in value for a 1% movement). Importantly, these ETFs reset every day. This makes them potentially toxic for long-term investors who buy them without being prepared to manage them actively (and a number of US banks have already been fined for mis-selling these products). It is not difficult to construct a scenario in which an investor could make the correct call on the direction of a market, but make a significant loss as a result of the use of leverage.
Consider this scenario: an investor buys a leveraged gold ETF on Monday morning anticipating that the market will go up. He is an occasional investor and does not close out his position every evening. The following week’s trading is a volatile one, but, checking the spot price on Friday, the investor finds it is flat. Checking the price of his ETF, he finds that he is, in fact, sitting on a loss of xx%.How has this happened? Say gold opens at $100 and then declines $10 a day on days one to three. It then improves $15 a day on days four and five. This leaves the spot price at $110, a 10% increase over the week. However, because the ETF resets every night, the decline of 10% on day one leaves 90 units, the following day’s 10% decline leaves 81 units, the next day’s 10% loss leaves 72.9. A 15% uptick the following day leaves the fund on 83.8, and the next day’s 15% increase leaves it on 96.4. Whereas the price of gold has not moved, holding a leveraged ETF has resulted in a 3.6% loss for the week. This is important for stability reasons – price discovery in financial markets tends very seldom to be linear, rather, prices rise or fall in a volatile fashion. In very volatile markets, the use of leverage is likely to exacerbate losses to existing holdings, adding to an existing sense of panic or crisis. Again, the failure of the investing public to understand the mechanics of these products leaves them over-exposed.
The ETF, therefore, occupies a strange place in the modern financial market. While these assets allow investors to diversify, they seldom realise what they are diversifying into, and the extent to which they will also diversify their returns. This has two important impacts on our financial outlook. The first is that in a financial culture which abhors individual accountability, there will at some point, when returns fail to pass muster, be a mass backlash over ETFs culminating in a demand for compensation payments and the diversion of bank funds from useful investments in the real economy. The second is that the investment bias towards ETFs will –through leverage and the discounted value of promises at times of crisis – amplify the next financial crisis the world faces. In other words, these are complex, potentially dangerous financial products which require a sophisticated financial understanding to trade profitably – it should be of great concern that they are perceived as a fairly vanilla method of portfolio management by the investing public.

Your email will not be shared and you can unsubscribe whenever you want with a simple click.

Image credits: banner; foreclosure; stock exchange; gold; stocks; dollar