Portfolio Construction
Thoughts on how to construction a portfolio for a non-expert amateur stock-picker
After massive market turmoil in 2020, my portfolio took a massive beating showing me the importance of proper portfolio construction. Although the best discretionary investors out there concentrate on their ideas, retail know-nothing investors need to have broad diversification to compensate for their lack of skill. But what about all of us non-finance retail investors who are willing to put some time into researching ideas and trying to pick securities? How should we think about this art?
Portfolio construction is a compilation between the investor’s temperament and ability, underlying business risks and uncertainty, and the overall cash flow characteristics of the portfolio. Let’s breakdown these 3 elements.
Investor’s temperament and ability
I once received advice from Francis Chou, whom I’m paraphrasing, “one of the biggest mistakes that a beginner makes is taking too large a position.” Hearing but not listening to this piece of advice when I started off, I made this mistake over and over again until I learned this lesson well. I suffered from the classic problem of knowing a little and being overconfident (the Dunn-Kruger effect). However, until I faced these losses, it is difficult to understand how I would feel watching these errors and how I would react to them. I think having these losses are important to recognize how to design a portfolio strategy that works for me in the long-term. I think that is why successful investors and traders have such a breadth of concentration vs diversification and turnover strategies. It depends on your own personal psyche, and what works for one person may not work for another.
Now that I recognize this important point, what would I have done instead?
First, I would have to accept the fact that my ability to select the right businesses that will appreciate in value with time, is limited. I’m competing with individuals with significant cognitive resources who dedicate their lives to this pursuit full-time. I’m competing with individuals with a data advantage from their privileged access and social networks. I’m competing with teams with better analytical abilities. The market, in aggregate, will be smarter than me.
Secondly, lacking investment experience, my ability to think rationally to avoid the fear-of-missing-out, sunk cost effect of not investing after spending time reading the filings, and becoming more risk tolerant especially with price declines through anchoring on purchase prices, averaging down when I don’t recognize a secular loss of business momentum, or selling just because the prices decline. With no mentor or team of investing friends, making the right decisions is really hard.
If I could go back in time, the solution is diversification. However, because I have a day job, tracking 100 positions by myself is similarly untenable. What I would have done at the beginning is to just buy a broad market index and dollar-cost average a percentage of my pay cheque that I won’t need to access for 10 years. If and when, I identified a business I wanted to invest in, I would take a 1% position. For a beginner investor, it is more likely that I would find many seemingly “attractive” businesses than an expert investor due to the lack of well-thought out selection standards. Although this behaviour is undesirable, it actually can be very useful for the learner. By having many chances to bat, it provides ample feedback for the beginner to learn and evolve over time without having catastrophic permanent loss of capital and losing out due to opportunity cost not being in the market.
The beginner can look back retrospectively after a few years of practice to see how they’ve performed. Using a journaling system and a larger data set, the learner can gain some valuable insights into what successful decisions look like. This gives one insight into how their emotions affect their trading actions and figure out how skillful one is employing their past investment framework. Hopefully, with proper reflection, one can work up to sizing larger positions over time with better outcomes via better processes.
I have been trying to manage my portfolio over the past 10 years with marginal quantitative success. However, qualitatively, I’ve learned a lot. I guess that is the cost of an education (a very expensive one). So today, I’m trying to heed to smarter people’s advice a lot more. My emotions and selection filters still lets in a lot of ideas. I recognize my impatience to start new positions after doing my version of “deep” work. I recognize that I often run out of liquidity and feel regret not having the firepower to take advantage of broad market dislocations. I recognize that I’m a bit of deer-in- headlights with large price declines. I recognize that I get distracted from the key business performance indicators with price volatility. I recognize that I feel significant pain with unrecognized losses. I recognize that recent past successes make me more risk tolerant and overconfident. I now aim to have much smaller position sizes in the range of 1-5%. I can tolerate the emotional pain from price volatility better in 2-3% positions sizes. I aspire to have ~10% in cash to avoid market timing but have some degree of optionality to take advantage of lower prices. Only in the rare situations do I plan to deviate from this overall framework.
Underlying Business risks and uncertainty
Starting with the end in mind, I realize now that the larger the position size means that I need to be increasingly right about the future. I think it is a bit silly for influential investors to tout concentration is what separates good vs bad investors without the caveat that it really depends on the individual investor and their investing environment.
Charlie Munger says 3 stocks are all he needs. He owns Berkshire, Costco, and has the rest in Li Lu’s funds. But he’s Charlie Munger. Furthermore, Berkshire is a diversified conglomerate, Costco is large cap retail/financing business that has already achieved scale and Li Lu is his smart personal friend. Munger also has significant influence over the business direction in formal and informal ways. He has the ability to de-risk his investments. I don’t have those advantages and likely will never have them.
Given that I have trial-by-fire business analytic skills, only have access to publicly disclosed information, and no privileged access to management all the while doing this part-time, what would be a reasonable approach to portfolio construction?
The two advantages that I have is time arbitrage and no specific investing style mandate. If I could couple these edges with better behaviour, I think I could achieve reasonable returns. The most important thing to focus on is value-for-effort spent on an idea in the context of my limitations.
There are several questions that can serve as a guideline to portfolio construction:
1) the range of future fundamental outcomes as it pertains to the business (eg. proven products or services, past management execution, a tangible market)
2) historical data available (eg. tenure of key people, financial data)
3) your own willingness to obtain information to reduce the uncertainty over time (which may require access to suppliers, customers, and management)
4) the ability to exit a position if wrong.
I think the best way to illustrate this is from some examples.
Scenario 1:
Mid-to-large cap business that has been around for 20 years with large public database set of financial metrics with management vetted by successful investors that you admire with lots of share liquidity is currently suffering from a global market meltdown.
In this situation, it is possible that the range of future outcomes is narrower given that this business is at scale and others have validated management. There is also lots of historical financial data to model out what the range of prices would represent a good return on investment. Although, it would be highly unlikely to have access to suppliers or management for businesses at this scale, usually there is enough publicly available information to judge the business direction from earnings transcripts, investor presentations, and filings. If this is a business that one could understand easily, a larger position size is fine given that this bet can be easily reversed if wrong.
Scenario 2:
Micro-cap to small-cap business that has been around for 10 years with a small customer base with 10 years of publicly available financial data but no earnings transcripts. However, these are some recorded presentations on YouTube from past investor’s conferences and free access to up-coming presentations on-line as well as some transcribed annual meeting discussions. The founder is still in place since the beginning and owns a substantial amount of shares. The share liquidity is thinly traded on a stock exchange.
In this situation, the range of future outcomes is much wider than the prior example. They could lose a customer, their product has likely found a niche and potentially could be more widely adopted or not. There is some insight into management’s thought processes and some demonstrated ability to execute. They have skin in the game and could be aligned with outside shareholders. The sizing decision here is dependent on one’s willingness to dig into the business and its ecosystem to resolve these forecasting uncertainties. If this is a part-time endeavour, but the narrative and numbers seem enticing, these uncertainties will only be resolved with time if one is unwilling to scuttlebutt. Furthermore, the cost of being wrong is high given the illiquid shares and inability for outside activist investors to displace management (or even if possible might not look after outside passive investors). A small position size is more appropriate in this situation.
Scenario 3:
A private investment with a business founder that you have a direct relationship with. This relationship has been cultivated over the past 10 years. The founder is honest, admits to their past fallibility, and you trust them to treat you fairly. The business they bought has been around for 20 years. The financials are available for you to look at and the founder’s business strategy makes sense to you and the probability of completing it is high.
In this situation, the range of outcomes should be easier to define given your personal insight into the founder’s past behaviour and their capacity to adapt to future business challenges. If the business is one that you can understand, then you can be more confident in its investment merit. The biggest questions are what will happen if the investment doesn’t work out, will the pain be shared fairly, and what are the future potential liquidity events? The answers to those questions will governing the position size. It is arguable that because this is an established business, the strategy to improve is reasonably easy to achieve, and you might believe that the founder will treat you fairly. As such, a larger position size could be allocated despite its illiquidity.
The Business Brew podcast hosted by Bill Brewster had two episodes that were particularly insightful on this subject. The first was with Joel Greenblatt and the second with Ian Cassel. Greenblatt humbly states that he is not the most astute analyst and that his competitive edge lies in portfolio construction strategy with particular care in position sizing and forcing turnover by limiting the number of ideas in the portfolio while being fully invested. Cassel, being a concentrated microcap investor, dedicates his time digging into his ideas’ ecosystem with substantial scuttlebutting. This is his competitive advantage by uncovering information that others are not willing to obtain. Cassel confesses that turnover is particularly important if the business loses fundamental momentum and managing micro-cap illiquidity can be challenging given the sandbox he plays in.
With these concepts in mind, what is an average retail investor with a less discriminatory investment filter with a long-term horizon to do with limited time and a weak social network?
My strategy would be put all ideas into the following buckets:
1) Buy and hold established businesses at scale that have reached mid-to-large cap status with good liquidity using a value-averaging or dollar-cost averaging technique over a 1 to 1.5 year period. These investments can be sized up to a 3-5% starting position. The key here is buying when the valuations approach a sufficient long-term return on investment.
2) Buy and hold small cap businesses that have reached a minimum 7-10 years of public life with management having significant skin-in-the-game. With limited willingness to dig deep into the business’ ecosystem along with its illiquidity, a venture capital (or Motley Fool) type approach may be more appropriate. This would equate to small bets, 1% sizes with the hope that the winners in the basket will compensate for the losses, for an overall adequate return profile.
3) Buy and trade mid-to-large cap shorter cycle businesses with easily accessible historical and current performance data. After a 3 year holding period, evaluate the probability of the mean-reversion thesis coming to fruition. If the key performance indicators are pointing in the wrong direction, sell it. With this approach, a position size somewhere between 2-3% would be reasonable given the effort needed to monitor the investment.
4) Buy and trade net-nets in the cyclical service industry with plentiful historical data. Similar to bucket #3, hold for 3 years and sell if the indicators show that the thesis is broken. These investments are likely going to be in the smaller cap space with more share illiquidity, thus a reasonable position size would be ~1-2%, and maybe 3%+ if there is sufficient liquidity to exit the trade in the future.
The overall portfolio cash flow characteristics
Losses are twice as painful relative to the joy of rewards. Loss aversion is a particularly powerful cognitive bias. Without a constant in-flow of capital into the portfolio, a retail investor undoubtedly, will sell their winners and keep their losers instead of doing exactly the opposite. This diminishes the overall portfolio growth. With largely a buy and hold strategy, it is important to cut loose the losers and redeploy capital into the winners or new ideas with better long-term prospects. It is also important to allocate a portion of the portfolio to ideas that can churn over in a 2-5 year time frame. This will help raise capital to grow the portfolio. This practice is hard, and plenty of time is need to hone this behaviour even if there is continuous capital inflow from one’s day job.
At the current phase in my life, I have a steady day job and excess capital can flow into the portfolio over time. What will happen when I retire? Without the steady cash flow coming in, redeployment of capital is required in order to provide for living. A focus on dividend paying businesses and selling shares of businesses that are buying back their shares to provide for living expenses becomes more critical. Hopefully, some of the smaller cap businesses that was purchased decades ago are now large cap dividend paying and the large cap businesses that I owned have survived obsolescence and will be cannibalizing their public float at reasonable prices.
The freedom of managing my own portfolio has one major benefit…not having to answer to another’s mandate but my own. Peter Lynch’s strategy of having a mixture of growth stocks, steady eddies, a few turnarounds and some asset plays makes sense for an all-weather portfolio. This diversity in style also helps with capital raises and minimizes catastrophic risks to permanent capital loss.
Although a specialist can likely out-select winners in an industry compared to a non-expert generalist, the generalist with a carefully crafted portfolio strategy and some discipline might do just fine over time. At least that’s what I tell myself.