No Light Rail in Vancouver!

More on Modeling: Cities Are Queerer Than We Can Imagine

Jan 10

2007

Some planners and economists once built a model of their city. They assumed all the jobs were downtown and people wanted to minimize the combined cost of housing and commuting. How far, on average, would people live from work?

The model said, “One mile.” But census data showed that people actually lived an average of seven miles from work.

The planners and economists had totally opposite responses to this answer. The economists assumed there was something wrong with the model, and set about refining it. Instead of a monocentric model in which all jobs were downtown, they created a polycentric model that spread jobs across several different job centers. The revised model said people would live a little more than two miles from work.

“Naturally we don’t expect the real world to fit the model perfectly,” wrote the economists, “but being off by a factor of seven or even three is hard to swallow.” The economists concluded that the model needed much more refining.

In contrast, the planners concluded instead that there was something wrong with reality. The cities are inefficiently designed, they decided, leading people to waste their time commuting. Typically, they came up with entirely the wrong solution for reducing commute lengths. They proposed to create a “jobs-housing balance” that would put as many jobs in each neighborhood or suburb as there were workers. This, they imagined, would allow more people to drive shorter distances or even walk to work.

But if going from one job center to several job centers doubles projected commuter miles, think what going to many job centers would do. As UC Berkeley planning Professor Robert Cervero discovered, many San Francisco Bay Area communities “are nearly perfectly balanced, yet fewer than a third of their workers reside locally, and even smaller shares of residents work locally.”

Economist Tom Bogart’s new book, Don’t Call It Sprawl, notes that even the polycentric model fails to come close to the complexity of modern cities. In most American urban areas, only about 30 to 40 percent of jobs are located in downtown or other job centers. “The majority of employment is dispersed throughout the metropolitan area,” says Bogart.

This means the polycentric models account for only about a third of the jobs in many urban areas. No wonder their projections were wrong.

The planners’ ideal jobs-housing balance doesn’t work because people locate their homes for many reasons other than being as close to work as possible. They may want to be in an area with good schools, nice parks, or restaurants and shops that fit their tastes. There is even evidence that people don’t want to live close to work. UC transportation researcher Patricia Mohktarian found that people prefer to live an average of 16 minutes away from work, perhaps so they could use the commute to mentally prepare for work and home environments.

Biologist J.B.S. Haldane once wrote that “The universe is not only queerer than we suppose; it is queerer than we can suppose.” In 1993, forest ecologist Jerry Franklin paraphrased this in a presentation to President Clinton, saying that “old-growth forests are not only more complicated than we imagine, they are more complicated than we can imagine.”

In the same way, cities are not only more complicated that we model, they are more complicated than we can model. Chaos theory teaches us that complicated systems like cities or forests are chaotic, meaning (among other things) that tiny differences in initial conditions can lead to huge differences in outcomes. That is the butterfly effect, the notion that a butterfly flapping its wings in Biejing can change the weather in New York.

In such situations, models will not work and plans based on those models are likely to do more harm than good. This means that planning is impossible and suggests that we need to find other ways to solve the problems that planners claim to address.

16

Trackback  •  Posted in Regional planning, Why Planning Fails  

Reprinted from The Antiplanner