Originally published at: The first half of 2024 in markets – InvestEngine Insights
Contents
- Performance of InvestEngine managed portfolios
- Drivers of returns:
- Inflation
- Earnings growth
- Are we in a bubble?
- Question 1 – Is the market too concentrated?
- Question 2 – Is the market too expensive?
- The case for security diversification
- The case for geographic diversification
- Summary
Introduction
With the first six months of 2024 behind us, investors will likely be pleased with how markets have performed. All regional equity markets have generated strong positive returns, and high bond yields continue to provide a tailwind to fixed income investors.
This has been against a backdrop of global uncertainty, as over half the world’s population heads to the polls to elect new leadership, the war in Ukraine continues, as does the conflict in the Middle East. These headlines have belied an unusually calm period for markets, which have not only generated excellent returns, but have done so with record-low volatility.
Performance of InvestEngine managed portfolios
Our managed portfolios have performed well, generating attractive risk-adjusted returns across all risk profiles:
Investors largely have two circumstances to thank for this pleasant start to the year: falling inflation rates around the world, and strong levels of earnings growth.
- Falling inflation
We started the year with inflation heading in the right direction, but still with some way to go before central banks hit their target rate of 2%.
The UK was at 4%, the US at 3.4%, and the Eurozone at 2.9%. Since then, the UK has continued to make good progress, with inflation now down at the target level of 2% (although persisting core and services inflation are preventing the Bank of England from celebrating just yet), the US has improved only marginally at 3.3%, and the Eurozone is edging closer, now at 2.6%:
The direction of travel, thankfully, remains clear, and with debate surrounding only the timing and pace of rate cuts, the spectre of hikes is now becoming a distant memory. As long as rates are reducing, then investors feel safe to continue buying risky assets like equities, and that’s exactly what they’ve been doing – pushing markets higher.
The taming of inflation this year has been largely thanks to central banks keeping interest rates high – although the exact extent to which central bank policy has reduced inflation remains debatable. Rates may stay high for a while longer, particularly to deal with stubbornly high core inflation (which excludes more volatile food and energy prices) and services inflation (inflation in services like education and hospitality), both of which have remained higher than expected.
The unexpectedly sticky inflation, particularly in core and services, has reduced the rate at which inflation has fallen, and thus has kept interest rates higher than markets had been predicting at the end of last year:
Although the market has done a fantastically poor job of estimating the timing of rate cuts since inflation peaked at the end of 2022, it does at least seem as though the era of monetary tightening is over. While rates may remain on a plateau for a while, the market remains confident that policy will ease from here, and not tighten.
- Earnings growth
The second trend that investors have to thank for their gains so far this year has been the continued optimism around earnings growth – particularly among the leading Artificial Intelligence (AI) companies.
This has led to the seemingly never-ending streak of outperformance for the so-called ‘Magnificent 7’ stocks (NVIDIA, Apple, Amazon, Microsoft, Alphabet, Meta, and Tesla) versus the rest of the market:
These 7 stocks posted exceptionally strong earnings results in the first half of the year. Given they started the year as already large weights in the US market, these further gains have dragged the market up with them.
Particular attention has been paid to NVIDIA – the poster child for the AI boom and the driving force behind the US market’s rally. The stock has returned over 150% this year. Its size, coupled with its extraordinary performance, has meant it alone has been responsible for almost 30% of the US market’s gain this year, returning 4.64% of the market’s overall 16.13%. NVIDIA has contributed more to the US market’s return so far this year than the entire bottom 490 S&P 500 stocks:
Towards the end of the first half of the year, we saw a brief scuffle between Microsoft and NVIDIA for the title of the largest company in the world, and while NVIDIA briefly held the mantle, that honour belonged to Microsoft as at the end of June. For context, in 2015 NVIDIA was ranked in the bottom half of the S&P 500 (an index which weights its constituents by size), at number 317.
Its meteoric ascent to the top is not only testament to the company’s exceptional results, but is also an interesting demonstration that active managers are still the ones setting prices in the market – not the buyers of index-tracking funds. This concern, that the market is simply funnelling cash to the largest companies and fuelling the market’s growth by ignoring fundamentals, sounds convincing prima facie, yet remains unfounded – as the likes of NVIDIA demonstrate. (On the other side of the coin, Tesla was added to the S&P in 2020, coming in at rank number 5, and has since fallen to 13th, showing that its fall was from active selling, not passive buying. General Electric also was a top 10 name in 2015, which fell as low as number 116, and has now recovered to sit in the top 50. That doesn’t sound very passive.)
The prolonged magnificence of these seven stocks has, understandably, led many market participants to question whether we’re now in an ‘AI bubble’. The combination of increased market concentration, higher valuations, and overly-optimistic earnings expectations has led some to compare today’s environment to the dot-com bubble in 2000.
Are we in a bubble?
As the US market has become increasingly concentrated in a few AI-related companies, this has led some to worry that the broader market’s reliance on these companies increases the risk to equity investors.
Because these few stocks have generated such fantastic returns, they have inevitably become larger weightings in investors’ portfolios. These 7 stocks now account for 36% of the S&P 500 and 22% of the global market.
Any earnings disappointment in these few behemoths will therefore likely result in large drawdown, the argument goes. So should investors be worried?
We’ll attempt to answer that question by answering two questions: 1) Is the market too concentrated? and 2) Is the market too expensive?
Question 1 – Is the market too concentrated?
Firstly, it’s true the US market has become more concentrated in recent history. As of the end of June, the top 10 holdings in the S&P make up 35% of the index. This is higher than at any point in the last 30 years:
Understandably this is giving some investors pause. Yet, this is not so unusual.
If we look elsewhere in the world, a concentrated equity market is the norm. In fact, if we look at how concentrated the US market is relative to other major markets around the world, it actually ranks relatively low on the concentration scale:
Our very own FTSE 100, while not exactly a poster-child for a booming market, has almost 50% in the top 10 companies. Switzerland has over 30% of its market in only 2 stocks – Nestle and Novartis.
Having said that, paying attention to US market concentration does seem to have more relevance for investors – nobody has 60% of their portfolio in Swiss stocks. But the point remains that concentrated equity markets are the norm.
And there’s a reason concentrated equity markets are the norm. According to research, markets are concentrated because only a tiny handful of companies are responsible for a market’s gains over the long run. Most stocks end up being terrible investments, and markets only rise because a small number of exceptional companies produce returns high enough to offset the vast majority of underperforming stocks.
In fact, market returns are so skewed that only 2.4% of global stocks accounted for the entirety of global stock market wealth created between 1990 and 2020. Not only that, but the majority (55%) of stocks actually underperformed 1-month US treasury bills over the period.
This is a fascinating finding, and one which clearly demonstrates just how difficult individual stock selection is. Active investors, those investors who try to outperform the market, are tasked with finding those tiny minority of stocks which will generate all the wealth. And if they fail to find them, then the active manager will, with almost mathematical certainty, underperform the broad market (which, by definition, will always hold those super-stocks). Finding the needle in the haystack is an incredibly difficult task, and almost impossible to do repeatedly on a consistent basis. Investors holding broad global-market funds, on the other hand, will always hold those tiny minority of stocks. They, as John Bogle famously said, simply buy the haystack.
So markets being driven by a small handful of stocks is not necessarily something to be worried about, and not a reliable indicator of a bubble.
But what about market valuations? Are they becoming untethered from reality, as happened before the dot-com crash?
Question 2 – Is the market too expensive?
Looking at the S&P’s current valuation, it’s clear the US is trading some way above its long-term average:
Based on this, those who point out that the US market is expensive are correct. The S&P is undoubtedly expensive relative to its history – but an expensive market alone doesn’t cause a market crash. There are several issues with simply arguing a market must fall because it’s expensive:
- As the chart above shows, markets can remain expensive (or cheap) for decades before reverting towards their long-term average. There’s no guarantee they can’t remain this expensive, or become even more so, over the next decade.
- A reversion to a long-term average multiple from an overvalued position can also take the form of flat market prices and rising earnings – growing the denominator of the P/E ratio rather than reducing the numerator. In this instance of mean-reverting valuations, there would be no crash.
- To confound matters further, the long-term valuation average changes over time, and a sustained period of higher valuations pulls the long-term average higher.
- In addition, the state of the world changes over time. The US market looked very different in 1900 than it does today, and it may not be fair to compare valuations today to valuations from over a hundred years ago.
- There are some interesting arguments to be made for the idea that the world we live in justifies a state of permanently higher valuations. For example, investors today may be willing to accept lower returns than in the past, because it’s so much cheaper and easier to invest. Investing today is infinitely more simple than it was in 1940, and today’s investors may be happy to accept lower returns as a result. Secondly, today’s companies are very different from those of the past. Valuations attributable to railway companies in the 1900s may not be comparable to valuations for today’s technology companies. Today’s tech giants may well deserve their high valuations.
So while it’s true the market is expensive, that doesn’t necessarily mean it’s likely to fall.
But what about those who suggest these high valuations might be taking us into bubble territory? Are we close to being in dot-com mania?
A quick scan across some common valuation metrics suggests this isn’t the case:
| At dot-com bubble peak | June 2024 | Difference | |
| Price/earnings | 29.58 | 22.61 | -24% |
| Price/free cashflow | 56.26 | 32.66 | -42% |
| Price/book value | 5.15 | 4.96 | -4% |
While the performance of the US market has been strong, earnings have grown alongside prices. The result is a heavy discount for the current market versus the valuations seen before the dot-com crash.
Drilling deeper, the forward valuations of the Magnificent 7, with one exception, look elevated – but not in keeping with bubble territory:
If you’d ask most investors to guess which of the Magnificent 7 stocks is trading at almost double the second highest – NVIDIA would likely be the common answer. In fact, Telsa’s valuation far exceeds any of the others, at a forward multiple of 80. NVIDIA looks relatively expensive at 42x forward earnings, but its strong expected earnings have stopped its valuation heading into the stratosphere. Meanwhile, Alphabet and Meta are looking like value stocks, at least relative to the other five, both trading at multiples of under 25x next year’s earnings.
One caveat to bear in mind – these numbers are based on expected earnings growth. Should that growth not live up to expectations, then the stocks may be more expensive than they appear here (and the reverse is also possible. Returns from the AI-linked stocks have so far this year exceeded previous earnings expectations).
Looking at consensus earnings expectations for this time next year, NVIDIA has some lofty targets to meet if it’s going to justify its current valuation:
| Q2 2025 – consensus earnings growth | |
| NVIDIA | 135% |
| Tesla | 44% |
| Apple | 10% |
| Amazon | 16% |
| Microsoft | 12% |
| Alphabet | 9% |
| Meta | 12% |
At any rate, whether looking at historic or forward earnings expectations, comparing the current coterie of mega-caps to the valuations we were seeing for the largest stocks during the run-up to the dotcom crash remain starkly different. The table below shows the largest companies in the world as at 31st December 1999 and their valuations at the time:
| P/E | |
| Microsoft | 73 |
| General Electric | 29 |
| Cisco | 221 |
| Walmart | 58 |
| Intel | 35 |
| Nokia of America | 71 |
| Nokia | 80 |
| Pfizer | 40 |
| Average | 76 |
| Median | 64 |
Today’s largest companies have a high bar to clear if they’re to continue demanding such a premium, but their valuations are still some way off the vertiginous levels seen in 1999.
So market concentration isn’t necessarily cause for concern, and valuations are still some way away from bubble territory. Does this mean the current set of companies will continue to dominate? Should investors just keep buying AI stocks?
Looking at history, the answer is: probably not.
The case for security diversification
There aren’t any iron laws in investing. Proofs are impossible, because the state of the market is never constant. The competitive nature of the market means that as soon as a way to generate consistent outperformance is discovered, others flock into the trade and eliminate its premium. “As soon as you think you’ve got the key to the stock market, they change the lock”, as they say.
Ironically, this process is a good example of the closest thing the market has to a law: mean reversion. Mean reversion is the idea that things, like interest rates or valuations, tend to return to their average (mean) level over time. They may experience periods of abnormally high growth, but will eventually be pulled down to their average level. Trees don’t grow to the sky. Similarly they may underperform for a period, but will eventually be pulled up to their average.
Mean reversion applies to many areas of investing. But can we expect it to apply to the current state of the market? Has the US grown too big relative to the rest of the world? Have the Magnificent 7 grown too big relative to every other stock?
History suggests the largest stocks don’t stay the largest for long. Looking back at the top 10 companies by decade, it’s clear the market doesn’t let stocks get too comfortable at the top:
| 1980 | 1990 | 2000 | 2010 | 2020 | June 2024 |
| IBM | NTT | Microsoft | PetroChina | Saudi Aramco | Microsoft |
| AT&T | Bank of Tokyo-Mitsubishi | General Electric | ExxonMobil | Apple | Apple |
| Exxon | Industrial Bank of Japan | NTT Docomo | Microsoft | Microsoft | NVIDIA |
| Standard Oil | Sumitomo Mitsui Banking | Cisco | ICBC | Alphabet | Alphabet |
| Schlumberger | Toyota | Walmart | Walmart | Amazon | Amazon |
| Shell | Fuji Bank | Intel | CCB | Meta | Saudi Aramco |
| Mobil | Dai ilchi Kangyo Bank | Nippon Telegraph | BHP Group | Berkshire Hathaway | Meta |
| Atlantic Richfield | IBM | Nokia | HSBC | Tencent | Berkshire Hathaway |
| General Electric | UFJ Bank | Pfizer | Alphabet | JP Morgan | Eli Lilly |
| Eastman Kodak | Exxon | Deutsche Telecom | Apple | Visa | TSMC |
The largest companies in 1980, dominated by the oil majors, largely disappeared from the top 10 by 1990. Then, in 1990, when Japan was set to take over the world, those largest companies followed the same fate, and were replaced by an entirely new set in 2000. The tech crash came for those companies, and by the next decade had been replaced by rising Chinese companies. In turn they fell too, and 2020 saw the start of today’s well-known tech dominance, with the top 10 dominated by technology companies.
If history is anything to go by, the next decade may see an entirely new set of companies rising to the top.
Putting some numbers to the phenomenon that large companies don’t stay large forever, research has shown that companies underperform the market after making it into the top 10:
While companies experience runs of superior outperformance relative to the market in the 10, 5, and 3 years leading up to a stock’s inclusion in the top 10 (which makes sense, as the largest companies are largest because they have the highest market capitalization, and a run of sustained strong performance is required for a company to break into the top 10), the performance doesn’t last. Interestingly, after a company enters the top 10, it becomes more likely to underperform. In the 5 and 10 years after entering the top 10, companies underperform the market by 4% and 14% respectively.
This is mean reversion in action – stocks generate supernormal earnings for a time, sending the share price higher. Then, eventually, the market extrapolates too far – earnings expectations become overly optimistic, competitors snip at their heels, diseconomies of scale sink in, margins contract, and the company can’t keep up with what the market has come to expect. Earnings begin to normalise at a more realistic long-term growth rate – pulling the share price down to meet reality.
History tells us that once a stock enters the top 10, its best days are usually behind it. Mean reversion takes hold, and the stock goes on to underperform. The data suggest we should remember to diversify away from simply holding the largest stocks, despite the temptation to try and capture the strong performances which took them there.
Investors should not only remember to diversify at the security level, but also across geographies. This is relevant for all investors, but seems particularly so for younger investors who have experienced nothing but US exceptionalism since starting their investing journey.
The case for geographic diversification
Mean reversion doesn’t just exist at the company level – it’s equally present at the regional level.
The chart below shows the difference between the rolling 5 year returns of the US market versus the rest of the world (using the MSCI EAFE as a proxy for the rest of the world – the index covers Europe, Australasia, and the Far East):
While the US has enjoyed a long stretch of outperformance versus the rest of the world since 2008, in the decade prior it was ex-US which outperformed. Similarly, the US outperformed for much of the previous decade, and the decade before that it was the rest of the world outperforming.
Is it now time for the rest of the world to have its day? When will the US start to underperform – if ever?
The answers to such questions require a crystal ball. However, it’s useful to understand the cyclical nature of markets, because it helps investors avoid the trap of simply buying only what’s worked well recently. While the US has done well recently, it may not always be so. If history is our guide, the need to diversify internationally remains strong.
As a surprising aside, it’s interesting to note that since the market bottom in October 2022, ex-US stocks have returned almost as much as the S&P. Something which most investors are unlikely to realise given much of the financial media has focused only on the Magnificent 7. There are plenty of opportunities for strong returns outside the US, too:
Returning to the case for diversifying outside the US, not only does the cyclical nature of markets suggest that diversification remains prudent, but a closer look at valuations further bolsters its case:
Valuation multiples are another prime example of mean reversion in markets.
Prices can only go so high relative to a company’s fundamentals (earnings, cashflow, or other business measure) before investors stop being willing to overpay and start selling the stock as it becomes overvalued. Similarly on the downside, prices can only go so low relative to a company’s fundamentals before the company becomes a bargain and investors start buying (pushing up the price). Research has shown that high valuations are often accompanied by lower long-term returns, and low valuations by higher long-term returns.
It’s interesting to see, then, that the US is the most expensive market, both when looking at forward earnings multiples and also on FCF multiples. Elsewhere in the world valuations are looking far more attractive.
China is almost in the single digits on earnings multiples, and European countries are looking relatively cheap across the board. The US looks expensive, trading at a 62% premium to Europe on a forward earnings basis and a 75% premium to Emerging markets. It’s also worth remembering that these multiples are based on expected earnings, which, as mentioned, are a high bar for these US companies to clear.
Again, if we put our faith in mean reversion and the historic relationship between valuations and long-term returns, the case for investing outside the US becomes even stronger.
A caveat
There’s always a but.
In this case, there’s a potential rebuttal to the idea that diversification away from the US is a good idea, that ex-US stocks will surely outperform one day, and that these mega-cap AI companies can’t grow forever. One rebuttal to almost all the points made above.
And that’s that AI companies are different.
The world of AI has caused investors to reconsider whether these rules of mean reversion still apply. If we’re in a new paradigm, then maybe this time it’s different. This new technology may enable companies to protect their long-run margins from competition, protect their market share, protect their size advantage, negate diseconomies of scale, and protect the oligopoly which they seem to have created.
For now, these technology companies are taking an ever greater share of the market and their earnings look impregnable. But will it last? Nobody knows.
For now, a diversified approach remains the most sensible. As the famous investor Sir John Templeton said, the four most dangerous words in investing are “this time it’s different”. Investors were similarly optimistic about the largest companies in 2010, and 2000, and 1990, and 1980, and every decade before – and those companies have now faded into the history books.
Indeed, a diversified approach doesn’t mean not owning these stocks. A global portfolio still has large weights to these AI companies. But it also includes allocations to areas of the market ripe for positive mean reversion – those companies with attractive valuations throughout Europe and the emerging markets.
Minimising potential future regret can be a useful framework for constructing portfolios, and the regret investors would likely experience from over-allocating to this minority of stocks, should their story not play out, is likely higher than the regret felt for remaining diversified but watching these stocks continue to outperform.
Summary
The first half of the year has been a comfortable one for investors. Volatility has been low and returns have been high.
Falling inflation across the globe, combined with strong earnings growth, particularly in the US mega-cap technology companies, have driven markets higher. There have been concerns around the US market becoming overly concentrated in a handful of stocks. However, the US remains less concentrated than many other markets around the world, and investors would do well to remember that concentration remains a feature of the markets rather than a bug. The overwhelming majority of the market’s returns have always been, and will always be, generated by a tiny minority of stocks – which is one reason it pays to stay diversified (“Buy the haystack”).
There have been similar concerns among some investors of the market approaching bubble territory. While the US market is indeed expensive relative to its long-term history, an expensive market does not mean a crash is imminent. Looking at today’s market leaders, these companies are generating more in cashflow and earnings relative to their prices than the companies in the dot-com bubble. Both today’s market and today’s largest companies are cheaper than they were in 2000.
While we’ve seen exceptional returns from this handful of US tech giants, it doesn’t mean we should be consigning diversification principles to the recycle bin. Mean reversion is a powerful force in markets, and the cost of holding a diversified portfolio remains low. Valuations are cheap elsewhere in the world, and there are plenty of high-quality companies located outside the US technology sector.
While it may be tempting to chase the performance of individual stocks or a single market, we would urge investors not to ignore the lessons of history. A well diversified portfolio, by definition, will always include the minority of exceptional stocks and will always have large weights to the best performing regions. But it’s also constructed to protect investors should yesterday’s winners not become tomorrow’s winners, and mitigates potential losses should investors’ bets on the current market darlings be proved wrong.
Which is why our portfolios remain diversified across all regions, sectors, companies, asset classes, and currencies. We can’t know the future, but we can construct portfolios which remain robust throughout market cycles, and which are not dependent on the performance of any single region or handful of companies.
If you would like to learn more about how we construct our portfolios, you can learn more by reading our Investment Philosophy document.
Important information
Figures taken from Bloomberg 02/07/2024
Capital at risk. The value of your portfolio with InvestEngine can go down as well as up and you may get back less than you invest. ETF costs also apply.
This communication is provided for general information only and should not be construed as advice. If in doubt you may wish to consult a professional adviser for guidance.
Tax treatment depends on personal circumstances and is subject to change, and past performance is not a reliable indicator of future returns.