Stocks for the long run? A Handful of Businesses Accounts for Much of Equity Market Returns

This article first appeared on Advisor Perspectives.

                  The Lindy Effect: The expected remaining life of an item is proportional to its past life.

“The Lindy effect is one of the most useful, robust, and universal heuristics I know.”

                                                                                                                                   – Nassim Taleb

In our last article, we presented the equity market returns as proxied by the S&P 500. We concluded that using U.S. equities as the proxy for equity market returns is incorrect. The primary issue is that proxying overall equity returns via U.S. equity returns is beset by survivorship and lookahead biases. U.S. equities in year 1900 were a much smaller component of the global markets than they are today. As of the date of this article, U.S. equities accounted for nearly 60% of global equity market capitalization. Generalizing from the U.S. relies on the extraordinary success its economy and financial markets enjoyed over the past 100 years, information that is available only in hindsight.

Shifting fortunes of sectors and industries

This leads us to another important risk that investment portfolios comprised of individual equities face. Investors are commonly told that they should buy and hold equities for the long-term. You only have to look at Warren Buffett for the success buy and hold investors can have.

If only it was that easy.

Much as the shifting fortune of countries, those of businesses and industries are ever evolving. Equity market benchmarks are constantly changing such that their composition, both in terms of industries and businesses, is ever evolving. Consider that in the U.S., railroads accounted for 63% of the total stock market value in 1900 and about 50% of the market value in the UK. Since then, railroads lost their prominence. They account for less than 1% of the market value in the U.S. and nearly 0% in the UK.

Figure 1 shows the composition of U.S. and UK’s equity markets in the years 1900 and 2015 as reported by Dimson, Marsh, and Staunton¹. Even outside of railroads, there were significant shifts in the market’s composition. Indeed, of the U.S. firms listed in 1900, more than 80% of their value was in industries that are today small or extinct, with the corresponding number for the UK at 65%.

Figure 1: Relative sizes of word equity markets, 1900 vs 2016²

Wealth management is a multi-generational affair. As you consider the current sector mix of the U.S. equities as seen in Figure 2, keep that perspective in mind. That mix will change.

Figure 2: Sector mix of the S&P 500 as of June 30, 2020³

Survivorship – Very few stocks and businesses age well, if at all

Whereas the U.S. stock market has generated an extremely healthy inflation-adjusted return of 6.9% over the past 150 years⁴, the average stock fared much differently. Since 1926, the average stock failed to beat the return of the US 1-month T-bills. In a 2018 study⁵, Hendrik Bessembinder reported that since 1926, the majority of common stocks generated lifetime buy-and-hold returns that were less than one-month T-bill returns.

It is an extremely important study that has meaningful bearing on portfolio construction processes. When constructing equity portfolios, be aware of the base rate. For portfolios comprised of single-name equities, the appropriate base rate isn’t the stock market return or the proportion of years market returns were positive versus negative. Instead, the base rate applicable to equity portfolios, especially those that have long-term holding periods, is that more than half of stocks generate lifetime returns that are lower than the one-month T-bill returns.

High probabilities of permanent losses

Indeed, what Bessembinder found is that the single most frequent outcome over the full lifetime of a typical stock is a loss of 100% (Figure 3). Nearly 12% of all firms suffered complete losses as measured by their lifetime returns including reinvested dividends. Another important data point reported by Bessembinder was the median lifetime buy-and-holdreturn of a typical stock: -2.29%. Interestingly, the study found that individual common stocks tend to have rather short lives with a median lifetime of 7.5 years as a publicly listed company.

Figure 3: Lifetime buy-and-hold returns of stocks – most frequent outcome is a 100% loss⁶

A very few winners – a positively skewed distribution

A very small fraction of all companies accounted for all the net wealth generation. Consider that fewer than 0.5% of all companies accounted for over half of the net wealth creation. Just about 4% of all companies accounted for all the net wealth that was created in the US stock markets. Figure 4 shows the number of firms that accounted for all the lifetime returns. As is seen, the top 1,100 companies out of a population of 25,300 companies accounted for all of the net wealth creation.

Figure 4: Lifetime buy-and-hold returns of stocks – A Handful Account for Much of Wealth Generation⁷

Not all is as gloomy as it seems

Note that 4% of companies accounting for net wealth creation is not the same thing as saying that only 4% of companies generated above-market lifetime returns.

What is at the heart of this dichotomy?

Think of the process of constructing a building. To construct a building that rises above the ground level, you first must dig up a hole. The depth of that hole can be equated to the loss of net wealth represented by businesses that generated poor lifetime returns. Now from the very bottom of that hole, structures must be built that will eventually allow for the building to rise above the ground level.

Because there are many more value destroyers among individual stocks, the hole that has been dug up is rather deep. Accordingly, a very significant amount of the construction effort is then spent just getting back to the ground level.

So even though only 4% of companies accounted for nearly all the wealth generation, there were many more that generated meaningful lifetime returns. It is just that their returns are hidden inside that hole, i.e.., by the value destroyed by losing positions. This can be seen in the fact that about 1/4 of stocks (26.1%) generated lifetime returns that were greater than the equal weighted portfolio of all common stocks over matched time intervals.

Key takeaways from the study

As we try and make sense of the relevant lessons from the study, some points are worth rehashing. Most stocks underperformed T-bills. Median lifetime return of all stocks is -2.29% with the most frequent outcome being a complete 100% loss. The median life of a publicly listed stock is just about seven years. And lastly, a small fraction of all businesses account for nearly all the net wealth creation.

Further, as we discussed earlier, nearly a quarter of all stocks generate superior lifetime returns as compared to the broader markets over comparable periods.

A sensible way to play the game – saying “no” most of the times

As you consider the discussion above, it is easy to conclude that to generate superior investment outcomes, one must focus on identifying companies that can generate extreme positive returns. Indeed, Bessembinder had that conclusion, stating that active strategies that have a concentrated portfolio of such stocks can generate very large stock selection returns. It is the same idea that Wall Street and fund salespeople regularly extol, bringing up the best performing funds of the recent past as evidence of the stock selection abilities of the fund manager.

However, such a conclusion is misplaced. The German mathematician, Carl Gustav Jacob Jacobi, offered the best process for solving difficult problems using a simple strategy: “man muss immer umkehren” (translated as “invert, always invert.”).

Indeed, the solution begins by recognizing that we do not need to dig as deep a hole as the markets on average do. As Warren Buffett plainly stated, “Rule No. 1: Never lose money. Rule No. 2: Never forget rule number one.”

Creating a different investment universe by following a process of removal

To this end, the creation of an investment universe is the first step in the direction of the solution. However, as against choosing an overall market segment defined by heuristics such as geography and/or market capitalization, it is a problem of rejections/removals. From your chosen market segment, you want to remove all businesses that have:

• questionable quality of earnings (understand the language of business – accounting )
• questionable corporate governance,
• poorly structured balance sheets
• poor capital allocation actions (understand the returns on incremental investments as well as M&A actions), and
• which deliver products or services that lead to dissatisfied customers (business has to create a net positive value for its customers).

Such a process, when consistently followed, leads to rejection of a vast majority of businesses. True, some of them will still go on to generate extreme positive returns, at least temporarily. However, you will likely end up filtering out a significantly larger portion of losers than winners; an outcome that will lead to generation of an investment universe with a superior lifetime returns distribution as compared to that of the broader markets. The reason why this should be the case is that you have successfully curtailed the left tail of the distribution while retaining much of the right tail.

Moats and high-quality asset owners – High probability fishing zones

Of the companies that remain, we segment them into five baskets: Moats, owners of high-quality assets, businesses we are interested in only at deeply discounted prices, not interested, and do not understand.

The biggest generators of lifetime returns come from two of these segments: moats and owners of high-quality assets. Previously, via a series of articles, we have offered detailed discussion of our moat evaluation processes. For high-quality assets, we look for desirable characteristics that ensure their relevance across time frames, e.g., electrical transmission and distribution lines, or producers of commodities that possess significant cost advantages, e.g., a steel production operation with first quartile industry cost curve positioning.

There are asset owners and/ or producers who do not qualify as high-quality owners as they own reasonable quality assets (but another superior asset owner exists) and or having cost of production positioning that is superior to industry’s average (but another cheaper cost producer exists). Such businesses, we look to acquire only at deeply discounted prices, such that the valuation discount compensates us for the business’s lower underlying returns.

The other two baskets are additional no-go zones. Whereas the “not interested” set represents companies that will generate poor returns, the “do not understand” set represents businesses that we cannot comprehend, i.e., our zone of ignorance. We constantly strive to limit this zone. However, there will always be a business that we do not understand that goes on to be a big winner.

Summary

Broader equity market returns serve as a poor proxy for an average stock’s return as well as that of a buy-and-hold strategy’s expected outcomes. A much larger proportion of all stocks contribute negatively to overall market’s lifetime returns. By following a process of rejections, investors can create a curtailed investment universe with significantly superior distribution of lifetime returns.

The process of looking for big winners should be applied to such a curtailed universe. This process should significantly improve an investor’s fail rates by reducing the number and extent of losers.

Multi-Act is a financial services provider operating an investment advisory business and an independent equity research services business based in Mumbai, India.

¹Long-Term Asset Returns, Financial Market History, Reflections on the Past for Investors Today, 2016

²Note: For 1900, UK data are based on the top 100 companies and US data on the total market.

³Source: https://siblisresearch.com/data/sp-500-sector-weightings/

⁴https://www.advisorperspectives.com/articles/2020/07/27/the-biased-history-that-drives-excess-allocations-to-equities

⁵Bessembinder, Hendrik (Hank), Do Stocks Outperform Treasury Bills? (May 28, 2018). Journal of Financial Economics (JFE), Forthcoming, Available at SSRN: https://ssrn.com/abstract=2900447 or http://dx.doi.org/10.2139/ssrn.2900447

⁶Bessembinder, Hendrik (Hank), Do Stocks Outperform Treasury Bills? (May 28, 2018). Journal of Financial Economics (JFE), Forthcoming, Available at SSRN: https://ssrn.com/abstract=2900447 or http://dx.doi.org/10.2139/ssrn.2900447

⁷Bessembinder, Hendrik (Hank), Do Stocks Outperform Treasury Bills? (May 28, 2018). Journal of Financial Economics (JFE), Forthcoming, Available at SSRN: https://ssrn.com/abstract=2900447 or http://dx.doi.org/10.2139/ssrn.2900447