Bigger stocks aren't always better

Published in The Sunday Times on 28th July 2024.

There are no gimmies in financial markets. However, one that comes close is that risk and returns are correlated, and the stocks of smaller companies therefore do better than bigger ones in the long run.  

Importantly, it accords with intuition. Many of the large stocks of today were the small ones of a decade or two ago. They’re often faster-growing, more agile, innovative, and frequently under-appreciated. That they should perform better in the long run seems uncontroversial.

Can market-beating returns be simply harvested by buying smaller stocks? Yes, historically at least. The person credited with discovering this in 1981, Rolf Banz, from the University of Chicago, showed that between 1926 and 1975, the average annual rate of return from large US stocks was 8.8%, while smaller ones averaged 11.6%. Even adjusting for extra volatility, small caps performed better than large caps.

Subsequent research showed that this also held true in international markets. The so-called “size factor” became widely accepted in finance. Even the father of the efficient markets’ theory, Eugene Fama incorporated it into his models. 

However, there are no iron laws in finance. It would be nice to have a few rules that explain market dynamics much as Newton’s three laws do in physics. But in financial markets, things that worked for a long time can simply stop working. And the small cap effect seems to have stopped working – or at least temporarily disappeared for the last fourteen years. 

And 2024 has been particularly bruising for small caps. The large cap S&P 500 is +18% for the year, crushing small caps by a massive 20%. The outperformance of large caps over small caps is now at its most extreme since October 1999. 

This has led to legitimate questions about the mere existence of a small cap effect. Was the small cap effect just an artefact of over-zealous data miners? Did it ever really exist, in US markets at least? 

Prominent sceptics like Cliff Asness of AQR argue that any size effect is just a mix of other market factors (like value or quality) disguised by the small-caps wrapper. 

In 2024, there is far more research and information on smaller US companies than ever before. Any systematic size premium that might have existed because companies were overlooked and underinvested has plausibly been eroded in an age of transparency, data accessibility and immense computing capabilities.

Many point to a 1997 paper by Tyler Shumway which detailed how there was a delisting bias in the data used by Banz and others that biased upward the size effect. The CRSP (Center for Research in Security Prices) database often omitted the final return for stocks that were delisted from exchanges, particularly for smaller firms that were delisted due to poor performance. Once adjusted for, the size factor was thought to be smaller and more nuanced than originally believed.

So, there’s definitely a case to answer in terms of whether a size effect ever really existed.

When we look for proximate causes for why the small cap magic has vanished, we see several. The most obvious of which is rising interest rates.

Smaller companies are hurt relatively more by higher interest rates. Large companies mostly borrow long-term at fixed rates through the bond market, and smaller companies carry more floating-rate debt, which has become much more costly over the past two years. In addition, small caps have a larger relative debt burden — their aggregate net debt is more than three times earnings, compared with under two times for large companies, according to analysis by Furey Research Partners.

Another culprit for the disappearance of the small cap effect is earnings and quality.

Earnings/Quality

The five-year growth rate of small cap earnings per share has been 14.4%, compared with the large cap at 15.2%. You’d have to classify this as bad, but hardly terrible. However, these earnings flatter to deceive. There’s a murky trend underlying this data.

According to Goldman Sachs research, almost a third of all small caps are now unprofitable, compared with about five percent two decades ago. Since the financial crisis in 2008, the number of companies with zero revenues is up six-fold and the number of stocks in the index that are unprofitable is up almost 2.5 times. Some of this is down to the index make-up, with a proliferation of unprofitable pharmaceutical and biotech stocks making up 16% of the small cap index recently, versus five percent 30 years ago according to research by Verdad Advisers. 

Yet even when you exclude healthcare entirely, the average quality of US small caps has markedly deteriorated over the past two decades, according to Verdad.

Private pools of capital starving the small cap universe

The Private Equity (PE) industry has been blamed for this. According to BCA Research, the outperformance of the whole small cap space comes entirely from a few exceptionally successful companies that experience extremely positive returns and transition into becoming large stocks — a phenomenon it calls “migration”. But this migration across size buckets has decreased to almost half of what it was at the turn of the century because of the availability of larger pools of private capital.

Many of the small companies that would have gone through an IPO (initial public offering) to become a small publicly traded stock in the past are no longer doing so. Instead, they are either acquired by larger companies, staying private longer, or IPO as a large cap. An entire swathe of high-quality companies that would decades ago have been listed and members of the small cap index, now exist inside a PE fund.

The argument is that only the worst companies are left with IPO as an option. Research by Verdad indicates that more IPOs of weaker companies and frequent exits by stronger ones explain much of the membership churn and deteriorating quality of US small caps. As a result, small cap indices suffer from an adverse selection problem, and the quality of indices has deteriorated as a result. 

So, what’s the counter-argument? 

They are cheap

At a price-to-book ratio of just 2, small caps are currently cheaper than the 10-year average of 2.2, and the 40-year average of 2.1. In contrast, the S&P 500’s price-to-book ratio is close to the record 5 times it touched in 2021 and 2000.

Over the long term, earnings growth for small cap compares very favourably to large cap. This must be the case — if small caps have produced a premium of about 2% per annum relative to the market over the long term, they cannot produce this without superior earnings growth, otherwise, their valuation multiples would have to rise in perpetuity.

According to a research piece by Furey Research Partners, 100% of the underperformance of small caps in the past decade is due to multiple compression in small and multiple expansion in large. That means that the earnings growth of small cap has kept pace with large cap, but the market has consistently discounted the price it is willing to pay for that growth. It’s reasonable to think that this trend will reverse at some point. 

Leverage

Furey partners show that despite rising levels of aggregate leverage amongst small cap, interest coverage ratios for small caps are near all-time highs. 

And, while it is true that there is a higher percentage of variable rate debt in small cap, that ratio hasn’t materially changed in 15 years. In fact, research from Dimensional does not show a drastic change in the weighted average profitability of small companies over the past 20 or 30 years, in contrast to the BCA research. 

Dimensional does make the point that small companies with a high relative price and low profitability or high asset growth dramatically underperform the rest of the small cap market. Excluding such companies from the eligible universe enhances expected returns.

The standard benchmark for US small caps is the Russell 2000 index. Its construction is formulaic and the methodology for index changes is transparent. 

But according to Dimensional, a very meaningful part of the underperformance of US small caps as measured by this index has to do with what it refers to as the ‘reconstitution effect’. Owing to the transparency and predictability - hedge funds and trading firms can front-run weighting changes, demotions and promotions meaning greater volatility and heightened transaction costs for small-cap funds that must rebalance on the day. According to Dimensional, this adds up to a 0.86% average annual headwind for the Russell 2000’s performance across 1979-2023. That is a considerable drag over time. 

So, if you’re persuaded by these arguments, then applying a quality filter to the small cap universe would seem to make sense.

Conclusion

Whenever markets appear to be at extremes, it’s always tempting to try to bet on a return to normal — or reversion to the mean in finance parlance.

But with Keynes’ famous aphorism "The market can remain irrational longer than you can remain solvent" ringing in my ears, this may be a moment to be brave — at least a little. 

Can large cap growth stocks dominate indefinitely? I’d be willing to wager a lot of money that the answer to that question is “No” (eventually). 

Is there a difference between being right in the end, and being wrong?  If you are managing someone else’s money, the answer to that is, no.  But that doesn’t mean that managers should ignore the potential for small caps indefinitely either.  Nor the importance of diversification away from the current large cap winners.

Source: Data is sourced from Bloomberg as at market close 29th December, returns are based on total indices in local currency terms, unless otherwise stated.

Gary Connolly is Investment Director at Davy. He can be contacted at gary.connolly@davy.ie or on twitter @gconno1.