As I was writing the last post, I started thinking about humans’ natural resistance to change and its implications. Bear with me, this could be counter-intuitive. Could resistance to change be a dominant factor causing economic and social bubbles and crashes? Are we building structures in our own image that are resistant to constructive evidence and positive change?
This is wonky and may not be right.
When we start from scratch, we try wildly different ways of doing something to achieve the initial desired result. Our feedback loop is simple – something either works or it does not. We change our method based on that binary feedback without hesitation.
When we find a reliable means of getting from A to B, the decision tree branches out to the next desired result, C. Naturally, we first try the same method that took us from A to B. This is the first iteration of the ”historical noise bit” that gets injected into our decision-making process. Twenty five decisions later, the cumulative historical components of the decision-making process evolve into intricate theories, largely based on correlation, not causation. The problem is that each iterative historical bit only has information on that one step from point D to point E. The combination of feedback bits DE and EF does not have causal information on going from D to F (I told you this is counter-intuitive).
By concentrating on the intermediary steps and analyzing past intermediary successes and failrues, we loose track of the entire long-term A-to-Z process and become resistant to fresh alternative solutions, because they do not fit into our history-ridden framework of pseudo-causation. Stuck inside the K-to-L bit, we do not see the objective Z clearly.
Although the historical feedback input is necessary to avoid reinventing the wheel all the time, it makes us resistant to try other things. When we find a pattern that fits well and gets intermediate results, we tend to iteratively over-use it to the point or going into a direction away from Z. Given a highly successful and self-reinforcing pattern of methods, the short-term opportunity cost of trying something else becomes huge. By steering away from Z, we are inflating the bubble that eventually pops.
This framework is apparent in the financial derivatives crisis. The first step was to securitize 100 mortgages and sell them to someone who wants an asset-backed product to diversify their portfolio. Everyone is happy — we got from A to B. Let’s say our ultimate ”Z” goal is capital growth and minimization of risk. One thousand iterations later, the financial instrument we are purchasing is almost completely decoupled from the actual asset. It consists of a bouquet other securitized products that are impossible to trace to the actual party who carries the obligation (your own car loan could potentially be part of the product). All of these assets separately have performed well in the past, so our historical noise is overwhelming. Then, a critical mass of consumers default on their mortgages and a domino effect ensues. Although the pattern of decisions worked nicely with every step, it slowly steered us away from our ultimate objective of capital preservation and growth.
Whoa. As with anything in life, there is no definitive solution. We obviously cannot discount historical information, because we ourselves are products of iterative historical process. But we need to find a way to reliably keep track of the ultimate objective: survival and prosperity.
