Organizing and decoding problems

My favorite sort of problem involves the interactions between individuals, groups of people, and mechanical rules generated by individuals and groups. Speculating on how the rules were shaped by the experiences of the individuals and groups is a fun game to play when presented with a curious set of circumstances. Conversely, my least favorite problems are those generated by the institutional brain damage of certain kinds of groups. Software and systems shaped by regulations and the particulars of monetary exchange are tedious at best.

I find it quite amusing when answers aren’t “clean” and technological systems aren’t the best solution. Often, no amount of analysis and brilliant coding can make an improvement. What is needed is to understand what people do, why they do it, and persuade them to do otherwise.

This brings me to economics, specifically behavioral economics. If you take away all the math, economics is largely about how people behave in aggregate. It’s a useful tool in understanding how to work with systems that involve people. But there’s more to economics than explaining how people interact in markets.

Russ Roberts, in the process of explaining how economics is not useful as a mechanism for answering questions like “did the stimulus work?” or “when will the housing market recover?”, gets to what really interests me about economics and finance:

“Is economics a science because it is like Darwinian biology? Darwinian biology is very different from the physical sciences. Like economics it is a very useful way to organize your thinking about complex phenomena. But it is not a predictive or very precise science or whatever you want to call it.”

The crux of the biscuit, for me, is right in the middle. Economics is a useful way to organize an intricate and interconnected problem and figure out how to reason about it. In casually studying economics and finance over the couple years, I’ve come to form a better mental model about how the world works.

Its possible that this mental modeling is what I do when I’m coding. I’m poking how some code works, looking at its inner workings, trying to understand how it interacts with the code around it. I sift through revision logs to see how it got to where it is today, talk to others on the team about the code and why it ended up one way and not another. I’m organizing and modeling the code in my head, building up a model that describes .

Lately, I’ve been describing my work as tinkering with lines of code until numbers appear on the screen in the right order. This is invariably greeted with a comforting-to-me response like “I could never do that”. But I really enjoy this. I’m not just debugging my code; I’m sharpening the way the program is organized in my head. Once I’ve got my head around it, I decode that organization into words, code, drawings, etc.

With programs, there is some system of rules, forces, and interactions which describe how the code works or doesn’t. Economics, too, describes a system of rules, forces, and interactions that predicts how a puzzle of human beings operate. Organizing and decoding these technical and social puzzles is great fun.