The Problem with Passive Equities
December 30, 2021
This is the second in a short series of posts which, as the title implies, address four important and inter-related problems for Wealth Management and, by extension, Asset Management and in conclusion offers a Solution.
In the first of this short series, we looked at the problem with Bond Markets and identified some structural issues, not least to do with the measurement and regulation of risk. The notion of a ‘risk-free rate’ was (mis)used in the wake of the Dot-com crash to imply that equities were much riskier than bonds, particularly when, incorrectly, combined with the short term volatility based Value at Risk (VaR) models intended for trading books rather than long-term investors. The net result was a structural push towards low yielding assets, a huge misallocation of capital and ultimately not only a ‘solution’ in terms of the Credit Default Swap market that actually caused a crisis in 2008 but now a situation where many bond portfolios are essentially a bar-bell of cash and high risk junk, with leverage and liquidity the key risks being taken but not being measured.
Passive Equities meanwhile have been widely touted as ‘the answer’ for long term investors on the basis that over time most active investors struggle to beat the index. However, they too suffer from a structural problem and, like bonds, this has also been caused by a mis-representation of risk – in this instance a focus on the benchmark at the expense of (almost) anything else. Without going into the passive versus active argument in too much detail here, we would instead make the point that the real problem today with Passive Equity investing is that the indices themselves have become poor representations of the broad economic exposure that underlines the concept of holding equities as part of a long term investment portfolio.
This problem is largely, although not entirely, a function of index composition and the widespread system of weighting by market capitalisation. As we noted in a recent post (options trading gives options for trading), the five largest stocks in the S&P500 currently make up almost a quarter of the total index weight by market capitalisation, a level not seen since the 1970s.
When you buy the S&P 500 index you are basically buying US Tech
To put it another way, a passive portfolio will have a holding in Apple that will be 10X that of Thermo Fisher, the 24th largest stock in the index. Is it likely to deliver 10x the risk adjusted return? And what about the other 476 stocks? The exposure of a passive investor to the underlying economics of these companies is so small as to be inconsequential. In effect, when you are buying the index, you are basically buying around half a dozen US tech stocks with a very concentrated stock and industry exposure. There is not a problem with this per se, so long as you are aware of the risks being taken to achieve those returns. Unfortunately nobody seems to be considering these as risks.
The reality is then that a passive exposure to the US via the S&P500 is really a concentrated exposure to the US tech giants of Apple, Google (Alphabet), Microsoft, Tesla ,Meta (facebook) and Nvidia. The largest non tech stocks are JP Morgan and Berkshire Hathaway, although between them they are still only slightly larger than Tesla. An exposure to the NASDAQ, while clearly a more conscious decision to focus on Tech, has a similar ‘Top Heavy’ skew and while the widespread claim that without the top 5 stocks the NASDAQ would be down 20% this year has been correctly disputed, it nevertheless makes a point. Look for example at the equal weighted and market cap weighted indices for the NASDAQ and the S&P500. While the S&P500 equal weight is actually flat on a one year view, it is down 12% on 5 years, while the NASDAQ equivalent is down almost 25%.
Market Cap weight been winning this year
On the one hand, one might ask why this isn’t a good idea? After all if the biggest stocks are always going to outperform the rest then why not have a portfolio heavily concentrated in a small number of tech stocks? The problem here of course is one of Portfolio Construction, that decision may be perfectly valid, but it is an really an active decision with associated risks, most obvious of which is the concentration risk, lack of diversification and simple mean reversion. Look for example at what happened to the equal weight versus market cap weighted indices in the aftermath of the dot com crash.
But watch out for mean reversal – a return of smart beta?
Indeed, the period after the DotCom crash not only saw a dramatic reversal of the equal weighted to market cap weighted indices, but also the growth of smart-beta as a concept. This took the notion that the market cap weighted benchmark was an inefficient portfolio and introduced a whole spectrum of different ways to ‘weight’ portfolios other than market cap. Given the power of the trend shown in the graph, almost everything ‘worked’, which was both good and bad, good for (mainly) hedge fund managers who had a Golden Era charging 2 and 20 for outperforming a market cap weighted benchmark, but bad in so far as there was no clear agreement on an alternative, leaving the majority of savings still tracking the S&P500 or the MSCI World, FTAll-Share or Stoxx. None of which are efficient portfolios for long term investors.
A US passive index for example has an exposure to internet/software/computers and semi conductors of around 37%. By contrast a European equivalent passive benchmark has, literally, one tenth of that, at 3.7%, the dominant components being Pharmaceuticals, Oil and Gas and banking. The situation is similar for the UK, except here mining is also a large component, which is really an exposure to Australia! As for the Australian market, this has 37% biased towards Banks and Mining, with the only real tech component being a 6.7% exposure to BioTech – itself almost entirely down to a single stock, CSL. As we say, none of this represent sensible investment portfolios.
Smart Beta did however, lead to an enormous number of new indices becoming available as benchmarks and more recently an equally enormous number of ETFs to track them. While these will be part of our solution to the problems we see in Asset Management, they are also causing some problems, as we will discuss in the next post when we look at the Problem for Active Equities.