161 | W. Brian Arthur on Complexity Economics
What this covers
Brian Arthur and host Sean Carroll discuss the foundations and reach of complexity economics—a framework that treats real economies as non-equilibrium systems of diverse, bounded agents whose interactions generate patterns no mathematical assumption can force into rest. The conversation traces how desktop computation in the early 1980s became a new scientific instrument, analogous to Galileo Galilei's telescope, letting economists relax the tight assumptions embedded in neoclassical economics and watch realistic phenomena—bubbles, crashes, lock-in, inequality—emerge spontaneously in simulation. Arthur argues that Equilibrium Thinking is not wrong for narrow domains (commodity markets that re-equilibrate within days) but structurally cannot address how new technologies form, reshape the Economy as Complex System, and create history itself.
The discussion moves through concrete models and their surprises: Kristian Lindgren's iterated prisoner's dilemma where no stable strategy ever emerges, the Santa Fe artificial stock market where realistic exploration rates trigger a phase transition into bubbles and crashes, and the El Farol problem where agents using diverse forecasting methods never settle but tend toward a chaotic middle. Arthur frames the core issue as one of fundamental uncertainty—agents cannot assign probabilities to outcomes because they don't know rivals' technologies or regulations—so rationality and optimization are often ill-defined, and exploration and coping become the realistic lens. The conversation extends to Path Dependence and Lock-in, where increasing returns let early chance events get magnified into dominant platforms, and policy implications, showing how computational "flight simulators" let policymakers test non-obvious strategies before deploying them.
Arthur argues that real economies are non-equilibrium complex systems of diverse, boundedly-rational agents reacting to each other, and that complexity economics—enabled by computation—reveals real phenomena (bubbles, crashes, lock-in, inequality) that the neoclassical equilibrium paradigm structurally cannot see.
- The essence of economics is agents making decisions in a situation where other agents are also deciding, which is inherently a complex system never settling to equilibrium.
- Computation is a new scientific instrument (like Galileo's telescope) that lets economists relax tight assumptions and observe emergent, realistic phenomena.
- Equilibrium assumptions implicitly deny that systems can be gamed or exploited, which is patently untrue.
Economic systems are quintessentially complex with many interdependent moving parts at different scales.
- An economy of a large industrialized country is a quintessential complex system with many moving parts related in different ways.
“if there is any system we can imagine that you would characterize as complex, it would be an economy”
Fluctuations are inevitable everyday features in systems where change occurs across all scales.
- In physical systems with a huge gap between molecular and macroscopic scales, large fluctuations are vanishingly rare, but in economic and social systems where something interesting happens at every scale, fluctuations become an inevitable everyday feature rather than a rare anomaly.
“in these economic or social situations where you have a multiplicity of scales and something interesting happening at every scale, that's exactly when fluctuations become just inevitable part of your everyday life. Is that a fair way of thinking about it?”
Optimal strategies in vast possibility spaces remain effectively unreachable through incremental improvement.
- Even if a single best strategy exists for the prisoner's dilemma, the space of possible strategies is so vast—like the space of possible genomes—that it will never be found; what agents can do is spend a very long time getting progressively better depending on their environment.
“the space of possible strategies is so large... that, who cares if there is a best one. You'll never find it”
Worldviews must complexify beyond simple equilibrium and pure-order assumptions about how systems work.
- Whether in policy or abstract thinking, we need to complexify our worldview and the way we think about how the world works, moving beyond simple equilibrium and pure-order assumptions.
“we need to complexify our world view and the way that we think about things. And that's true whether we're in policy, or just trying to be abstractly thinking about how the world works.”