91B0FBB4-04A9-D5D7-16F0F3976AA697ED
C9A22247-E776-B892-2D807E7555171534

Levitt Fellow Dane Johnson '07 Writing Macroeconomics Computer Program


Dane Johnson '07 (Red Bank, N.J.) is on campus this summer for research into macroeconomics. The mathematics and economics double major is writing a computer program which will help Johnson study the effects of technological progress on business cycles. Advised by Professor of Economics Chris Georges, Johnson is working on a project titled "The Creation of a Computer Program which Simulates the Effect of Technological Progress on Fluctuations in the Business Cycle."

Long though the title is, it describes exactly what Johnson is doing. He is creating "an agent-based computer simulation," or computer program involving firms ("agents") which models business behavior. Generally in macroeconomics, economists study the aggregate by finding a representative agent. In at attempt to incorporate the heterogeneous nature of firms into the model, Johnson is working on a program whose economy consists of a wide distribution of firms rather than a representative agent.

Johnson takes as his starting point empirical evidence that suggests firm size is distributed by the power law. In other words, there are many small companies but very few large ones. "We're trying to get that distribution [of companies in the program]," says Johnson.

This project is important because it takes into account the large companies. In economics a perfectly competitive market implies that each firm is too small to affect market quantity and price. Empirical evidence, however, suggests imperfect competition. Take for example Walmart, which employs roughly 1% of the U.S. labor force. When Walmart has fluctuations, these have far more impact on far more people than business fluctuations in, say, the community ice-cream store. "The microeconomics of very large firms have an effect on the macroeconomics," explains Johnson.

Johnson's own description of his project is this: "I'm trying to come up with a model that studies the effects of technological progress on business cycles. In this model technological quality and efficiency are exogenous variables which are shocked randomly. We study the effect these shocks have on output and prices." He is focusing on technological improvement because it is often the motivation for economic growth.

His interest in macroeconomic simulation stems from his previous acquaintance with Georges, who specializes in agent-based simulation programs. Johnson worked with Georges during the year, checking a financial markets simulation program for "bugs." Georges is also working on a business fluctuation program; Johnson calls his work this summer "a baby version of Professor Georges.'" Of his advisor, Johnson says, "it's good learning from him."

This is Johnson's first summer of research and he is enjoying himself. "It's fun coming up with models [and] seeing if they work," he says. The hardest part of the project is actually designing the program and working in its language, for which he has also had to learn C++ programming.

On the Hill, Johnson is a TA for math and symbolic logic courses. He plans to use this summer's work for his senior thesis in economics and hopes to continue working on the program Georges is designing. After graduation, Johnson has firm plans to go to graduate school for a degree in economics or statistics. He explains that he decided to do summer research because he wanted preparation for graduate studies.

The grant which funds his work is the Levitt Research Fellowship, given by the Arthur Levitt Public Affairs Center. This grant is intended to fund research in a public affairs issue and allows a student to spend 10 weeks working closely with a faculty advisor.

- Lisbeth Redfield

Back to Top