Think Complexity

Chapter 11. Case Study: Sugarscape

Dan Kearney, Natalie Mattison, and Theo Thompson

The Original Sugarscape

Sugarscape is an agent-based model developed by Joshua M. Epstein and Robert Axtell to investigate the economics of wealth distribution. They presented the original model in their book, Growing Artificial Societies.

The sugarscape is a virtual 2D grid where each cell has a certain amount of abstract wealth, called sugar. Agents roam the grid and accumulate sugar.

In the simplest sugarscape, each agent has a sugar reserve, a metabolism at which rate it consumes its sugar, and a range of nearby cells that it can observe. At each time step, the agent observes its nearby cells and moves to the cell with the most sugar. These rules can be expanded to include topics as varied as reproduction, death, disease, loans, and warfare. For example, a disease can be introduced to the system wherein a sick agent can infect nearby healthy agents.

Despite its simplicity, the model generates outcomes that resemble the real world. When modeling wealth with Sugarscape, a long-tailed distribution appears where some agents are vastly richer than others. Similar behavior is seen in most real economies where a small part of the population holds a large fraction of the wealth. Extreme wealth inequality is generally considered a problem, because it means there are many people barely surviving while others are fabulously rich.

The Occupy Movement

Wealth inequality has partly fueled a modern social movement ...