fractal complexity

Schumpeter's theory about how capitalist economies tend to continually renew themselves through creative destruction is kind of like how forest biomes are supposed to work when they are not being mucked up by humans. Emergent systems like forests and markets have a lot of commonalities in this regard. They are not only extremely complex; they are fractally complex. In other words, the more we zoom in to study them to figure out how they work, the more we realize how little we actually understand.

Humans occasionally get these hair-brained ideas about how we can "manage" fractal complexity. This usually results from some egghead building a model of the system, demonstrating its supposed utility by accurately predicting future events, and then using that model to bend reality to our will.

All models are wrong, but some models are useful. Error is the source of unintended consequences. You can only control for the variables you know about.

When we try to manage forests, we end up with record-breaking forest fires. When we try to manage markets, we end up with anything from widespread famine to too-big-to-fail. Roughly the same phenomenon is at play in either case: We thought we had an accurate working model of the system, but we really only knew enough to be dangerous. Sometimes, the consequences are catastrophic.

Should we stop trying? Of course not! But knowing that some of the systems we try to model are fractally complex should give us pause while implementing policy; at the very least, we ought to be a little more humble in our approach. As we emerge from an era in which the experts were always right, perhaps a much-needed course correction is already underway.