- Michael Arbuthnot
“Three properties determine the complexity of an environment. The first, multiplicity, refers to the number of potentially interacting elements. The second, interdependence, relates to how connected those elements are. The third, diversity, has to do with the degree of their heterogeneity. The greater the multiplicity, interdependence, and diversity, the greater the complexity.”
- Gokce Sargut & Rita McGrath
In my book “Seeking Wisdom: Thoughts on Value Investing”, I wrote quite a bit about the concept of risk versus uncertainty. I defined risk as something that can be defined and measured while uncertainty cannot. For instance, risk might be seen as the chance of drawing a particular card from a full deck. We know the percent chance of the event happening. An example of uncertainty is calculating the number of hurricanes in next year’s autumn storm cycle. In this case, we can figure out how risky a hurricane might be (starting by estimating its strength with the Saffir Simpson scale), but not the number of hurricanes, simply because of the complexity of atmospheric change and storm development. In any situation, a participant must be able to ascertain what can be defined as risk and what is defined as uncertainty. One you can mitigate against is a rather specific way. The other you cannot.
Wall Street has always had a difficult time making the distinction between risk and uncertainty. Many times, analysts will use them interchangeably. This does investors a considerable disservice. Whether it be in the creation of credit derivatives or crypto blockchain, investors must be able to accurately discern between what is a risk to their assets and what is an uncertainty in their investment model. Another tool to employ is distinguishing between complicated models or operating systems and those that are complex.
Complicated versus Complex
Making the distinction between complicated versus complex isn’t helped by the fact that Roget’s Thesaurus shows them as synonymous. This most certainly is not the case when comparing complicated versus complex systems. So, before I get into why it’s essential to make the distinction, it might be helpful to define the meaning of each.
Complicated systems can have many moving parts - the more parts, the greater the complication. What’s critical to these systems is that each component reacts not only to the other but also in a deterministic manner. This means the reactions are in reliably predictive ways. Complicated systems generally operate under major laws such as chemistry, physics, etc. The responses are definable and generally consistent. These systems and their reactions/outcomes can be measured quite precisely. In this way, they are similar to our definition of risk. Much like knowing the odds of pulling an ace of hearts from a deck of cards, we can assign probabilities and percentages to reactions in complicated systems. Another essential part of complicated systems is that they aren’t impacted by unmeasurable components such as human emotions. Complicated systems don’t have a mind of their own, like financial markets or hurricane seasons.
An example of a complicated system is a discounted cash flow model. While very complicated with an impressive amount of data inputs, we can nearly always measure the reaction of some components against the actions of others. For instance, the company's valuation will drop with an increase in the ten-year treasury rate. This is due to the scientific nature of the time cost of money. The valuation drop can be calculated by the rise in the ten-year interest and the impact on the discount rate. There is nothing left to subjective or unmeasurable events. From this, we can declare the risk of investing in such an asset.
Complex systems are very different from complicated systems. Whereas a complicated system can be overcome by understanding relationships and deterministic patterns, complex systems develop their own self-interest. Complex systems don’t change on relatively set designs but evolve based on random actions. The operative word here is “evolve” rather than develop. Because of this, complex systems generally have no central control points. It’s impossible to say if I make point A zig, then I know point B will zag. This difference makes the relationship between complex and complicated systems very similar to risk and uncertainty. One is more defined, measurable, and quantifiable. The other is not.
An example of a complex system is the ecosystem in which a portfolio holding operates. It comprises companies, management teams, customers, and thought leaders, all driven by self-interest. It is nearly impossible to calculate the exact response of how a customer might react to a company’s new product or how a portfolio-holding management team responds to a competitive acquisition. Based on the player's self-interest, these responses can vary in so many ways to make a prediction a very shaky proposition. Moreover, the ecosystem in which your portfolio holding operates will evolve over time, with little central control or deterministic approach.
Why This Matters
As an investor, it is vital that you make the distinction between complex and complicated. You can spend much time and brainpower trying to solve problems or predict futures when it can’t be done well or repeatedly. In David Halberstam’s classic “The Best and the Brightest,” he details how the leaders in the Kennedy/Johnson administrations, deemed to be the smartest group of individuals since Jefferson’s time (hence the title), utterly failed to see what was happening in the war to save South Vietnam. They mistook a highly complex problem for a mildly complicated one, where grinding out the numbers could solve almost any issue. History has shown that some of the worst quagmires have been where leaders mistake complex for complicated. Value investing is no different. Several things to bear in mind when deciding whether you are faced with one or the other can be summed up as follows.
Draw a Bright Line Around Your Circle of Competence
Investors can avoid much trouble if they realize their circle of competence is smaller than they think. A circle of competence isn’t simply built around a particular industry or market cap. It also must define the limits of what you – as an investor and business analyst – can gainfully measure and use in your valuation process. Utilizing data from a complex system - thinking it is as valuable as a complicated system - can draw an investor far from any circle of competence. Knowing the distinction between the two can make all the difference between success and failure in your long-term investment performance.
Even a Stopped Clock is Correct Twice a Day
As I discussed earlier, a complicated system can regularly produce deterministic results. A complex system can seemingly produce the same results. But this is where an intelligent investor must distinguish between the signal and the noise. Just because you obtain the results you would expect doesn’t mean you can repeatedly predict an outcome with real success over the long term. Very smart people can be smug until they run into complex systems. It’s incredible how fast they (meaning the intelligent person, not the complex system!) can be humbled. Remember, a broken clock is correct twice a day, and no more. A repaired clock is accurate nearly all the time. Knowing how and why the clock is correct is just as important as knowing the correct time.
It’s Easy to Overthink Things
One of the things you learn with experience as a value investor is that it becomes easy to start overthinking things. It’s always great to be learning but not always to be overthinking. One of Charlie Munger’s great teachings is the idea of intellectual lattice works – the concept of multiple fields of knowledge, such as biology, statistics, psychology, and astronomy, layering and weaving together findings from each field. This latticework creates an increasingly complex ability to work on problems and see them from different perspectives. This idea of latticework can significantly improve an investor’s ability to identify opportunities and issues with potential investments. That said, it’s easy to get carried away and see trends or traits that have nothing to do with a company or market segment. Latticework is a great tool when utilized in a complicated model, but much less so when working in a complex environment. It’s easy to get overwhelmed by too much data that adds little understanding to the problem. Knowing whether you are dealing with a complicated or complex problem can reduce the risks of that happening.
Over my investment career, I’ve learned that ascertaining the difference between risk and uncertainty, along with complicated versus complex, is vital to avoiding permanent capital loss. Nearly every poor investment decision I’ve made over the years has been caused by incorrectly understanding each. When I thought I could calculate the level of risk to a certain degree, I found a level of uncertainty that could not be defined. Equally, I’ve thought a tremendous amount of data and elbow grease could help solve a complicated problem when dealing with a complex system. All of these situations have been defined by several characteristics. First, I was convinced I (and my team) were far smarter and more intelligent than we really were. This led me to think this was a brain power problem with an evidentiary solution. It was a complex systems problem with no clearly defined answer. Second, I placed too much capital at risk, where the odds could not be calculated reasonably. Last, when the investment case went wrong, I didn’t have the knowledge or system to identify what went wrong and the solution. In nearly every case, the answer was to exit the position and use the experience to understand better the opposite sides of very similar coins – risk/uncertainty and complicated/complex.
At Nintai, we are constantly looking to improve our decision-making processes. This includes making sure we know what we can measure, what we can control, and where we can make an impact. Understanding complicated and complex systems' role in our investment model is critical to that process.
As always, I look forward to your thoughts and comments.
 “Learning to Live with Complexity,” Gokce Sargut and Rita McGrath, Harvard Business Review, September 2011