Average is scarce in the stock selection process

Active investors want to do better than average. Index investors are happy with average, at a low cost. What may be lost along the way is just how scarce average is in the stock selection process. Active investors are always looking for unicorns. Maybe not billion-dollar concept stocks in 2023 – but listed companies that possess a combination of multiple attractive traits. High quality business. Strong balance sheet. Capable management, with integrity. High growth and of course a cheap valuation.

This describes most investment processes – with different investment styles emphasising different attributes as being the most important. But how common are these stocks? What is the base rate at which we can expect to find companies with the best of all these attributes? What about just being average or above?

If each of the five traits (business quality, financial robustness, management, growth, value) are independent, and assuming we just want companies that are average or better in each category, what are the odds a company is average or better, (say in the top 50%) in all five?

3%.

3.125% to be exact is the number of companies likely to meet that test (50%⁵).

An investor who follows this rigorous, benchmark unconstrained process and starts with a universe of 200 names should end up with a portfolio of about six stocks. Six stocks which are simply average or better across all five attributes. How do we get to a portfolio of 30 stocks? We either need to expand our universe of potential candidates (to 1,000) or reduce the number of attributes we care about. There are other approaches, but these are the most obvious two.

We think investors would find these numbers unintuitive. How rare it is to find companies who are just average or better – but across multiple dimensions. It indicates trade-offs investors must be making, when building concentrated portfolios within the S&P/ASX 200 universe. Seeking even higher hurdles in any attributes will dramatically shrink the population that passes the test.

3% was just to be average or better. Which of course most active managers don’t want. They want the best. The highest quality business, with the best balance sheet, run by superb, highly aligned management who can grow fast. And it needs to be purchased below fair value.

So if we only want companies in the top quartile across all of these attributes, what are the base rates? Well if we start with 10,000 opportunities, 10 will pass the test.

0.1%.

This also assumes that we can determine these attributes to be true in advance. That we don’t mistakenly believe that a company is in the top 25% which turns out to be worse. That top quartile management don’t leave. Importantly in small caps, it also assumes that the Cambrian process of natural selection for commercial fitness in emerging industries can be accurately forecast. We do not believe this is realistic with anything like the accuracy needed to pick 10 small cap stocks out of 10,000.

We observe the reality of investing in small caps is the process by which new winners emerge. The process in many ways mirrors the biology of undirected variation and evolution rather than the outcome of intentional design or directed evolution.

Source: Blackstar Funds.

Evolution by natural selection occurs when random mutations meet the current environment and those poorly suited for fitness (adaptability to the environment) fail to reproduce. Those characteristics passed on become amplified in subsequent generations. The process here is randomness, exposure to reality and failure. Evolution is not a process of positive attributes being actively selected, it is a negative process of unfit attributes being selected out and what remains amplified.

New businesses tend to go through a similar process. Tens of thousands of enterprises are started. Most fail to be fit for the public market. Even then, of the 1,000 plus Australian small cap stocks, hundreds will fail. Either actually fail (receivership) or fail to achieve an acceptable return on capital invested.

One way of seeing the small cap evolutionary struggle is to look at the skew of stock returns. It was a surprise to many when Hendrick Bessembinder demonstrated that positive skew meant the median US stock underperformed US treasuries, despite the total stock index (the average) doing significantly better. It gets even more exaggerated when you look at US small cap stocks. The median lifetime returns of the bottom third of the US market cap is actually negative, despite small cap indices doing very well.

The skew in individual stock returns has been known for a long time but is rarely discussed. One implication is skew imparts a cost to concentrated portfolios, which reduces as the size of the portfolio increases. A 1998 study (Ikenberry, Shockley and Womack. The Journal of Wealth Management) demonstrated that a 25-stock portfolio in the S&P 500 carries around 30bps per annum in skew-based performance drag. This reduces to around 5bps at 100 stocks. Given the higher skew, concentration drag is likely to be greater in small caps.

Escaping the small cap commercial primordial soup is hard and the few big winners deliver all the investment performance.

Source: Ikenberry, Shockley and Womack. The Journal of Wealth Management. 1998.