The idea of complex adaptive systems might sound overwhelming, but an experiment conducted by physicist Per Bak makes the concept easier to understand.
If you drop one grain of sand at a time onto an empty table, a small, cone-shaped pile begins to form. As the pile grows, eventually a grain of sand will hit the pile and trigger an avalanche.
If you’ve ever watched sand run through an hourglass, you might have noticed this dynamic in action. As the sand pours through the top glass and the pile below grows, small avalanches of sand start cascading down the side of the cone-shaped pile.
The longer the pile avoids an avalanche, the bigger the eventual sand avalanche will be.
Bak’s experiment was designed to determine the exact conditions that would trigger that avalanche. During his experiment, Bak found the sandpile to be completely unpredictable. Avalanches sometimes occurred after hundreds of grains were added. Sometimes they happened after thousands.
Bak came to realize the timing of an avalanche was not a function of the size of the pile or number of grains of sand, but instead was related to the interactions between those individual grains of sand.
The more grains of sand in the pile, the more interactions that occur between the individual grains and the more difficult it is to predict the next avalanche. Eventually, the pile reaches a critical point (called self-organized criticality) in which the pile transforms into something more complex and gains properties that must be considered separately from the individual pieces.
In other words, you can’t study the individual grains of sand and understand the pile in its entirety.
The Problem with Predictions in Nonlinear Systems
The sand pile is a great example of a nonlinear system that does not produce the same result every time even though the inputs and conditions are the same. You never know which grain of sand is going to cause an avalanche or how big the eventual avalanche will be because each grain of sand uniquely interacts with other grains to create a pile that is slightly different each time.
Like piles of sand, financial markets are also nonlinear systems. But they are far more complex. Sandpiles are simply made up of interacting grains of sand. Financial markets are comprised of millions of interconnected inputs that adapt and learn over time.
We can’t predict when an avalanche is going to occur in a sandpile with only one set of interactive agents. Shouldn’t we expect it to be exponentially more challenging, if not impossible, to predict financial markets comprised of human and computers each seeking out different data sources and interpreting them based on their unique personal experiences and trading rules?
The answer is yes, we should expect to find those predictions nearly impossible. And yet average investors and countless talking heads insist on making market predictions.
Understanding the Financial Markets as Complex Adaptive Systems
The times I find myself referring to financial markets as complex adaptive systems most often are cases in which people try to precisely link cause and effect.
Our innate human tendency to seek clean-cut reasons behind everything around us makes us highly susceptible to linear thinking. Exacerbating this is the fact that financial media presents its viewers with Wall Street “experts” that succinctly describe past events by explaining specific causes that led to specific market movements, all after the fact.
Market “experts” sound a lot like the description of historians that Per Bak describes in his book on self-organized criticality, How Nature Works:
“Historians explain events in a narrative language where event A leads to event B and C leads to D. Then, because of event D, event B leads to E. However, if the event C had not happened, then D and E would not have happened. The course of history would have changed into another sequence of events, which would have been equally well explainable, in hindsight, with a different narrative.”
The point isn’t that cause and effect don’t exist, but that they aren’t proportional. Large fluctuations are more the result of the interactions within the complex adaptive system and less attributable to external or environmental factors.
This means that cause and effect are not neatly linked. As a result, worrying about the cause of the next crisis a futile exercise.
If we start thinking of financial markets more like piles of sand, then we can no longer assume that a given action or event will produce a given reaction. Thinking about markets in a nonlinear fashion requires us to embrace an ever-changing mix of calculus and psychology.
What All This Means for You as an Investor
Although we are just getting started in our exploration of complex adaptive systems, there are some basic takeaways we can draw from thinking of financial markets more like sandpiles:
Avalanches are infrequent, so we shouldn’t assume the next grain of sand will cause an avalanche. Stock market returns follow a non-normal distribution that has more positive outcomes than negative outcomes.
Avalanches will eventually occur, so it is important you have a plan in place. Rather than constantly worry about when or why the next avalanche is coming, you should plan on avalanches occurring with a similar frequency and magnitude as they have in the past. By carefully assessing your willingness and ability to risk, you can build a portfolio you can stick with through poor market conditions.
Cause and effect are not neatly linked. People place way too much emphasis on a few specific data points that allow for a narrative that closely link cause and effect. Complex adaptive systems take on additional characteristics that can’t be accounted for by simply weighing the individual parts. This is a big reason it’s impossible to make accurate predictions about the market.
Ignore predictions from Wall Street “experts.” Good investing is boring, but the media creates a sense of urgency and encourages bold predictions as if they are the source of an informational edge. Nobody is going to get major media attention for saying, “there’s no way to know what markets will do next.” If we begin to view the market as a complex adaptive system, you might be more likely to simplify things and leverage your natural advantages to achieve investment success.
