Why Most Things Believed About Investing Are Wrong

A conversation remains stuck in my head from early 2020, when the terms ‘lockdown’ and ‘social distancing’ were largely unheard of. My meeting with a Chief Investment Officer was coming to an end. We were discussing a US automotive company, which at the time was finally being recognised by the market and thus being rewarded with massive share price growth. He leaned across the table and said, “tell me you have been selling your shares.” What struck me was not his belief that we should sell, rather that he appeared to hold it with such absolute certainty. His assertion wasn’t anything to do with the company itself, but rather the ingrained belief that when a share price goes up a lot, you should sell. This was common sense. To do different would be foolish, greedy and undisciplined. 

It is a conventional wisdom that pervades much of the financial industry. As the old saying goes ‘it’s never wrong to take a profit’. There is some validity in this approach, hence why it pervades and endures. A client is unlikely to be unhappy or indeed notice if you sell a stock that subsequently goes up significantly. That loss – of foregone upside – is not captured in performance data, but perhaps it should be. On the other hand, if the stock in question continues to be held and goes in the other direction it will become a clear detractor in performance data and you should expect to be asked, if not chastised, about it. And so, from the investment manager’s point of view, perhaps it can be said that it is never wrong to take a profit.

But what, I hope you ask, about the client? For the client, equity investing is asymmetric, the upside of not selling is near unlimited, while the downside is naturally capped. Surely, for the client it can be very wrong to take a profit? This goes to the heart of why so much of investing is wrong. Sadly, as an industry, institutional money managers seldom try to get investment right for investors. Most conventions and practices exist to serve, protect and enrich investment managers’ interests.

The realities of investment therefore are often very different from the dogma. At Baillie Gifford, we are fortunate to be informed by a range of thinkers from outside our industry that have no incentive to prop-up the myths of investment management. Instead, they deal in observable facts, not the self-serving mantras beloved by professional investment bodies. This note tries to share a few of their perspectives that have been crucial to how we invest and, in the process, demonstrate that it is often not just wrong to take a profit, but it can be the worst possible mistake.

Professor Hendrik Bessembinder

Where Returns Actually Come From

Let’s take Bessembinder’s paper Do Stocks Outperform US Treasury Bills? as our starting point. We have all been told the answer is yes because stocks carry significantly more risk, so, of course the factual answer is no. Nearly 60 per cent of global stocks over the past 28 years did not outperform one-month treasury bills. 

Equity investing as a whole though is thankfully still worthwhile. This is because of a small number of superstar companies. Bessembinder notes that a mere 1 per cent of companies accounted for all of the global net wealth creation. The other 99 per cent of companies were, it turns out, a distraction to the task of making money for clients. The capital asset price model (CAPM) so beloved by the financial industry is therefore nonsense because the normal distribution of stock returns that underpins it is imaginary.  

This should shake the very foundations of the investment industry. It provides not an opinion but a collection of facts as to where returns come from and what investors should focus on. The entire active management industry should be trying to identify these superstar companies since nothing else really matters. Investing is a game of extremes. 

But, here lies the problem, it requires a vastly different mentality to that displayed by the financial industry today. It requires focus on the possibility of extreme upside, not the crippling fear of capped downside. This requires genuine imagination should there be any hope to grasp the potential of superstar companies. 

In addition to imagination, Bessembinder makes it clear that it is the long-term compounding of superstar companies’ share prices that matters. Investing thus requires patience to deal with the inevitable ups and downs such companies experience as well as the ability to delay significant gratification. Sadly, such behaviours are difficult and wholly inconsistent with the incentives and annual bonuses of traditional finance. Nevertheless, they are prerequisites. After all, the point of superstar companies is that they can go up five-fold and then go up five-fold again. If you sell after the share price merely doubles, crow and take your profits, you undermine the whole point of identifying companies with extreme return potential in the first place. 

Bloomberg Pictures of the Year 2019: Extreme Business. Masayoshi Son, chairman and chief executive officer of SoftBank Group Corp.

Let’s take a practical example. In early 2000, the founder of SoftBank, Masayoshi Son, made what may have been the greatest investment in history. He invested $20m in a Chinese ecommerce company. Two decades later his remaining investment is worth in excess of $180bn. A wonderful example of an extreme return.

This is a well-known story and one I’ve been lucky enough to be told first-hand by Masayoshi Son. However, less well-known is the story of Goldman Sachs. Goldman invested in the same company a year before Son on far better terms. Shirley Lin, who worked for its private equity fund, had an agreement to invest $5m for a 50 per cent stake. Unfortunately, her colleagues deemed $5m too risky and so they opted for investing a ‘safer’ $3m. Five years later their stake was worth $22m, a seven-fold return. At this point, the decision was taken to sell under the guise it’s never wrong to take a profit. In many ways, this was a remarkably successful investment, until you realise that today those shares would be notionally worth more than $200bn before dilutions are taken into account. 

It would seem Goldman Sachs got the identification, and perhaps even the imagination part, right. They spotted one of the greatest superstar companies of our era early on. Yet, when asked why Goldman Sachs sold, Shirley gives a depressing but predictable answer: “they wanted quicker results”. Though this example is extreme and straddles public and private ownership, the point is clear: in investing, it is often not only wrong to bank profits, it can be the worst mistake you make. Despite this, in almost every client meeting I am asked about our sell-discipline. No one has ever asked me about our hold-discipline, which is a shame, as the greater cost to clients’ returns comes from the inability to hold onto superstar companies when their returns are ticking upwards. Investment managers are usually very good at selling.

Professor Brian Arthur

Increasing Returns to Scale

The notion that companies can even produce such extreme returns goes against much of economic theory which exhibits an overzealous equilibrium-mindset that likely has deep religious and spiritual roots. This mindset is the progenitor of a mindset of a different name common in finance, namely a belief in ‘reversion to mean’. 

Bessembinder’s data falsifies the assumption that company returns stabilise over time. Moreover, he shows when looking at nearly a century of US stock market returns the concentration of wealth creation is becoming yet more skewed towards a small number of companies. The returns are becoming extreme. The superstar companies are becoming more super. 

This is particularly odd given that economics focuses on diminishing returns to scale. It takes the work of Professor Brian Arthur of the Santa Fe Institute, to understand that this concept is rooted in observing the returns to scale of the “bulk-processing, smokestack” industrial companies of the 19th century. He notes that western economies have shifted “from processing of resources to processing of information, from application of raw energy to application of ideas”. Far from diminishing returns, today’s knowledge-based companies tend to exhibit increasing returns to scale and so, in the digital era, reversion to the mean is even less common. Returns are yet more extreme.

Professor Brian Arthur. © Corbis Historical/Getty Images.

Professor Ming Zeng

Network Companies Unleash Unimaginable Scale

Brian Arthur came up with this early explanation of emerging economic reality on America’s west coast, while observing both Silicon Valley and the rise of Microsoft further north. However, as with most things today, if we wish to better understand the economic reality of tomorrow, we must look east. To help us here we have the academic Ming Zeng. He became Alibaba’s Chief Strategy Officer in 2006, a job he took to further his studies by giving him a ringside seat to history in the making. 

For Ming, the greatest superstar companies of the future will be what he calls “smart businesses” harnessing network coordination and data intelligence. These organisations will look less like a company and more like a network. He notes:

The old, diversified conglomerate was like a complex machine of the old industrial age. It collapsed when it reached a certain complexity. But the future of business is more biological rather than mechanical… an ecological system grows and becomes more and more sophisticated, even more robust when it becomes richer and more diverse. 

Network companies, such as Amazon, Uber or MercadoLibre, coordinate millions of entrepreneurs, guiding them with data intelligence in real-time so both the network companies and entrepreneurs can adapt to conditions instantly in ways traditional companies could never have dreamt. This marries the benefits of enormous scale with rapid adaptability. Moreover, the larger these network companies become, the more data they have and thus the more intelligent and effective the network can become. If Ming is right, then it is logical to assume the importance of superstar companies will grow even further with the application of machine learning. 

Martin van den Brink

Extending Moore’s Law

Despite the above, we must not forget that the prevalence of new superstar companies will be determined by the amount of economic and social change that takes place from here. Indeed, I would go further still and posit that the single greatest determinant of whether returns will swing from growth to value is whether the pace of change in the world increases or decreases. For it is change which creates new markets and disrupts old ones. It is change that fuels the rise of superstar companies.

We therefore need to believe the conditions for change will persist. To give us that conviction we have Moore’s law. The observation and projection made by Gordon Moore, the co-founder of Intel, that, for the same price computing power would double every 24 months. This projection has now become seen as a law, given its predictive power over the last 50 years. In doing so it has set the pace for the semi-conductor industry and thus for human progress. 

What this means is that come the early 2030s computing power should be at least some 60x more powerful than today for the same cost. The implications of such a large increase are difficult to imagine. At the very least, the disruption should spread from industries like advertising and retail to those of even greater importance such as healthcare, finance, education, and many more. 

Nevertheless, it would of course be wrong for us to have confidence in past patterns, however long and persisting, without due cause. Here we are very much helped by the insights of Martin van den Brink, ASML’s Chief Technology Officer, whose lithography machines have been largely responsible for extending Moore’s law in recent times. Van den Brink makes two points. First, that Moore’s law is older than we think. It dates back not 50 years but an entire century in all but name to when computing power was vacuum tube rather than transistor based. His second point, is even more significant and far reaching, because he believes ASML already has the technology roadmap in place to ensure Moore’s Law extends into the 2030s. 

It therefore perplexes me why with the power and predictability of Moore’s law, our industry decides instead to focus far more on what interest rates or GDP growth rates mean for investing. Frankly, I think we would all be much better investors if we concentrated on the future implications of Moore’s law. A 60x increase in computing power will profoundly shape our world. The question we must grapple with is what this new world will look like.