The implications of extreme concentration

Concentration in equity markets has reached unprecedented levels, particularly in the United States.(1) A select few mega-cap stocks, colloquially referred to as the "Magnificent 7,"(2) now dominate market indices, reflecting a convergence of technological innovation, speculative enthusiasm, and the allure of generative AI. While this has driven remarkable returns for a narrow slice of the market, it raises critical questions about diversification, valuation, and risk for equity investors. At the same time, the growth of the market value of these firms largely reflects prospective (rather than realised) business growth, so they become increasingly reliant on anticipated or forecast future earnings. The resulting valuation multiples are very high (and seemingly growing), making the market itself heavily tilted to the style of expensive growth.

Note that this also suggests that the potential for industry disruption – through new and more advanced AI tools, lower costs of production, more advanced hardware – could create very large sensitivity to news. If competitive pressures do indeed yield disruptors, then these stocks will be highly susceptible and could sell off sharply or see significant volatility.(3)

In summary, we can therefore say that we are facing two fronts of concentration risk - market cap and style. The concerns here are manifold:

A “diversified” exposure to equities (especially US equities) taken via a market capitalisation-weighted (market cap) index fund or ETF will in fact give a concentrated exposure to only a small number of names.

The valuation multiples of these stocks are very high, which means that (again) a “diversified” exposure to equities involves purchasing a strong bet on expensive growth. Market cap index exposure cannot ever really be considered “style neutral”, but we believe it is now an extreme style bet.

“Stock picking” by fundamental active managers becomes very difficult. Setting aside for the moment the question of whether these active managers can consistently outperform, it becomes almost impossible in the current environment to outperform a benchmark (that is, in total return) without taking very concentrated bets in a small number of stocks.

Equity risk is not rewarded. The yield on equities should reflect their embedded risk when compared to debt, and so should attract a “risk premium” – in other words, a higher return. Instead, at the moment the average trailing earnings yield for the top 10 US stocks (which currently make up almost 40% of S&P500 market cap) is approximately 2.5% (as they trade on Price to Earnings (PEs) of about 38). Forward earnings yields – probably a better measure – are not much better, at 29x and 3.5%. US treasury yields are more than 4%, which means that the risk premium is negative.(4)

This issue is more or less fully confined to the US. The concentration in emerging markets and in Australia has remained fairly static for many years (not shown here but charts available on request).

One further point: the impact that concentration has on “stock picking” alpha generation is for investors looking for total return outcomes. There are simply too few stocks which will outperform on their own. However, if the intent is to accept the benchmark (with its concentration issues) and look purely for alpha above a benchmark, then diversified active strategies based around a benchmark are perfectly suitable.(5) All that is needed is for overweight positions to relatively outperform underweight positions.

Before moving to provide more detail on this, we need to step back and think of how this issue might be avoided. As the issue is one of concentrated market cap and excessive valuation, then the simplest and most effective approach is to build a portfolio that does not use market cap in its construction.

RQI’s approach to Value investing deliberately avoids chasing this sort of concentration in market cap. We build a Core portfolio based on the reported dollar value of a small set of accounting variables. Price (and so market cap) is not used. This means that the starting point (6) of our Value strategies is already much more diversified/less concentrated than the usual market cap weighted benchmark.

US equity dominance charts - plus other markets

The US equity market has rapidly increased its proportion of developed markets equities, as shown by the breakdown by country in the MSCI World benchmark in Chart 1 below.(7)

Chart 1: Proportion of MSCI World benchmark in various countries

The USA has increased its proportion from 49% at the beginning of 2000, to 57% at the beginning of 2015 to finally 74% at the end of 2024. The run up in weight post the Global Financial Crisis (GFC) is very strong and has accelerated further recently.

For the same dates, Australia (‘AU’) has moved up from 1.2% to 2.7% and back down to 1.7%, the United Kingdom (‘GB’) has fallen from 9% to 8% to 3.4% and the rest of the developed markets (‘RODM’, excluding US, Australia, UK, Canada, Germany and France) has fallen from 28.2% to 20.4% and finally to 13.1%.

At the same time, concentration of stocks within the US (and so MSCI World – see below) has risen sharply. To show this change, we start by looking specifically at the S&P500 in two charts(8):

1. The number of stocks needed to reach 50% of the market of the index (fewer stocks, more concentration)(9)
2. The aggregate percentage weight of just the top 10 stocks in the index (larger weight, more concentration)

Chart 2 below shows these for the S&P500 over time. Before 2020 the data is every three years rather than annually.

Chart 2: Number of stocks in S&P500 required to make 50% weight, and total weight of top 10 stocks. Data is end of year.

We see a very sharp increase in concentration from the end of 2022. At that point, the top 47 stocks aggregated up to 50% of S&P500 index weight. At the end of 2024, this number had almost halved, to 26 stocks. At the same time, the top 10 stocks now account for 37% of the index weight, up from 24% at the end of 2022.

Chart 3 draws a similar conclusion, but now for MSCI World. Panel A shows that the top 10 stocks increase in weight from about 10% at the beginning of 2015 to around a startling 25% today. The next two size ranges (11-50 and 51-100) do not grow over this period. By the beginning of 2024, more than 50% of the weight in MSCI World comes from less than 100 stocks, up from 40% in 2015.

Panel B repeats this just for the US stocks in MSCI World(10) . Here the concentration is much greater. The top 10 stocks account for about 15% in 2015 but have risen to over 34% by the end of 2024. The next two size ranges again show little growth over this period. In 2015 approximately 95 stocks made up the top 50%; today that number is less than 50. Clearly the increased concentration in Developed Markets (that is, the MSCI World benchmark) is driven by the US.

For comparison, we repeat this exercise with Emerging Markets and Australia (not shown here). We did not see an increase in concentration.

Chart 3 Panel A: Weight of stocks in three size ranges for MSCI World. Data is monthly.

Source: RQI, Factset. Data as at 31 December 2024

Chart 3 Panel B: Weight of stocks in three size ranges for just US stocks in MSCI World. Again, the data is monthly.

Valuation and the negative risk premium

The valuation multiples of US stocks in the MSCI World and the top 10 stocks in the MSCI World, has returned to extraordinary levels. In the case of the top 10 stocks, these levels have not been seen since the peak of the COVID pandemic when reported earnings fell to extremely low levels for most stocks.

Chart 4 shows the changes in PE multiples in recent times. Aside from the COVID spike, it is the sharp increase in the last two years that catches the eye. In particular, the top 10 names now have an average PE of 38x. Even including all the remainder of the US stocks in the MSCI World, the average PE today only falls to 32x. It too has trended up strongly. MSCI World is now trading on an average PE of about 26.

Chart 4: PE multiples over time

In principle, it is really expected future earnings (forecasts or estimates from analysts), not historical earnings, that matter for many stocks, so we use these going forward.

As mentioned above, one consequence of the dramatic run up in prices ahead of future earnings has been the increase in price to forward earnings (PfwdE) and the decrease in its reciprocal, the forecast earnings yield (fwdEY). (These are based on next 12 months forecast consensus earnings.)

Investing in equities carries risk of course. This risk is generally regarded as being greater than the risk carried by investment in fixed income, notably bonds, as debtholders (a) are paid interest before equity holders receive dividends in a corporate income statement and (b) in the event of liquidation of a company, debtholders are “senior” (get paid before) equity holders.

So, if risk is greater, we would expect return to be – on average – greater for equities than for bonds, as compensation for the extra risk. This is known as the risk premium.

The risk premium should almost always be positive. There will be occasions, especially with market disruptions like the GFC and COVID, where this may not be the case, but these should revert in normal conditions.(11)

So - are current conditions normal? Current 1 year US treasury bill yields are sitting at 4.2%, and the upwards sloping yield curve shows yield for 10-year treasury notes at 4.7% (according to Factset data as at 31 December 2024). These are the annualised yield to maturity returns that an investor would receive.

See Chart 5. Using only the US stocks in the MSCI World (blue line), current PfwdE multiples are around 22.6, so yield is sitting at about 4.4%, down from about 6% (PfwdE of around 18) at the start of 2015.

This means that current forward earnings yield (EY) is only very slightly more than equivalent term bond yields (one year) – the red line. Before COVID, when yields were much lower, the gap was very wide. The same applies to the broad MSCI World benchmark (orange line), although it has not fallen as far at US equities – PfwdE is about 19.5x which equates to a yield of 5.1% and a positive risk premium even today.

However, now look just at the top 10 US stocks (which make up 37% of the S&P500 and 25% of MSCI World) are trading on a weighted average PE of about 29, which corresponds to an EY of about 3.5%. That is, the largest stocks are trading with a negative risk premium – earnings yield is below the US 1 year T-bill rate. Said another way, the expected earnings return on holding these stocks (net of any further price moves) is below the return from holding debt for an equivalent term.

As mentioned above, this is not a normal situation when compared to history, which can be seen by comparing the green shaded area with the rest of the chart.

Chart 5: Comparing forward earnings yield to US 1 year T-bill rate

RQI Core portfolio concentration and valuation

Using RQI’s model for core portfolio construction (recall the core model is the portfolio before the alpha signals are applied), the concentration in holdings can be very different to a benchmark constructed using market cap. The RQI process encourages stability rather than trend following, and this is evident in the concentration of the core – it is very stable over time.

In the case of Developed Markets(12) (that is, MSCI World, of which US is now about 74%), see Chart 6, which uses 12-month forward earnings yield for comparison. The Core always shows a better yield than the benchmark (as expected), and that gap has widened recently. Both have trended down recently as well.

In fact, the RQI Developed Markets (DM) Core model currently shows a clear 2% premium in terms of this yield over the MSCI World benchmark, which is the widest it has been in 10 years. DM Core is much cheaper than MSCI World.

We can repeat this for just the top 10 stocks, which we do in Chart 8, but first note Chart 7. Chart 7 shows the weight of the top 10 stocks in the MSCI World benchmark compared to DM Core. The top 10 stocks in MSCI World were about 10% of the benchmark in 2015 – they are now an astonishing 25%. At the same time, the weight contribution of the top 10 stocks in DM Core has remained quite stable at just under 10%.

Very importantly, this does not mean that the RQI Global Value portfolio top 10 is a set of totally different stocks. In fact, many of the top 10 stocks are in common: Apple, Microsoft, Amazon, Meta for example. What is different is the weight at which these are held: RQI holds less of these extreme valuation stocks (and none of the likes of Tesla and Nvidia) than the benchmark, and more of the lower valuation stocks outside of the top 10(13). So, while the top 10 stocks in DM Core become increasingly expensive in the same way as MSCI World (see Chart 8), the overall valuation does not (Chart 6).

Chart 6: 12-month forward earnings yield for the MSCI World and for the RQI DM Core model. Monthly data.

Chart 7: Weight of top 10 stocks in MSCI World compared with weight of top 10 stocks in RQI DM Core

Chart 8: 12-month forward earnings yield of top 10 stocks in MSCI World (~25% weight today) compared with the 12-month forward earnings yield of top 10 stocks in RQI DM Core (~10% weight today)

Conclusion

Concentration in developed markets equities has grown sharply in the last few years. Investors in benchmarks which supposedly get broad exposure to these equity markets are actually exposed to a small, concentrated and expensive set of stocks. The top 10 developed market stocks, all US, account for a whopping 25% of the MSCI World benchmark (currently about 1400 stocks) and about 37% of the S&P500. These are unprecedented levels of concentration.

This is solely a US market phenomenon. Emerging markets and the Australian market do not see this dramatic increase in concentration.

The valuation of the MSCI World, US stocks and finally the top 10 stocks, have run up in recent years, no matter how we measure it. Comparing forward earnings yields to T-bill rates over the same holding period, we see that the expected return to T-bills is actually higher than for the top 10 equities – which implies the rare and unsustainable case of a negative risk premium for those stocks.

RQI’s core process has much less concentration than the benchmark, and trades at a significant valuation discount. However, it does not do this by avoiding these hugely expensive names completely. Instead, it holds them at a much more balanced weight, driven by the fundamentals of the company rather than its price.

Learn more

RQI Investors build portfolios using quantitative signals based on our team's insights and market experience exploiting inefficiencies to systematically select stocks from a broad universe. Our goal is to zero in on the ones that we believe can deliver an edge. Find out more.

Managed Fund

RQI Australian Small Companies Value

Australian Shares

View

This material has been prepared and issued by First Sentier Investors (Australia) IM Ltd (ABN 89 114 194 311, AFSL 289017) (FSI AIM) RQI Investors is a brand of First Sentier Investors.

This material is general information only. It does not take into account your objectives, financial situation or needs.

Any opinions expressed in this material are the opinions of the individual author at the time of publication only and are subject to change without notice. Such opinions: (i) are not a recommendation to hold, purchase or sell a particular financial product; (ii) may not include all of the information needed to make an investment decision in relation to such a financial product; and (iii) may substantially differ from other individual authors within First Sentier Investors.

No assurance is given or liability accepted regarding the accuracy, validity or completeness of this material and we do not undertake to update it in future if circumstances change.