Statistical Smarts: Corrupting the Data About Growth:
How the CT Metropatterns Report Misuses Census Data & Grossly Inflates CT's Development Trends

March 19, 2004
William H. Ethier

All the hubbub about sprawl, development patterns and the use of land has led some people to rally around "smart growth.”

Smart growth, as it is most often talked about here in Connecticut, is a collection of policies, laws and regulations intended to revitalize the cities and inner ring suburbs, limit the development of our outer ring suburbs and rural towns and preserve for future generations the 72 percent or more of Connecticut that remains undeveloped.

In discussing smart growth proposals, such as providing higher priority for state funding resources in urbanized areas, providing new regulatory authority over land use decisions to regional entities or providing municipalities with even more "tools” to stop development, many proponents of these land use changes appropriately start by defining the problems they want to fix.

Some of these problems are very real, such as the ever-widening income gap between city and suburban dwellers and the resulting concentration and isolation of the poorest of our citizens in certain city neighborhoods. But addressing non-land use issues, such as making the streets safe and improving schools would do much for this problem. Reforming city bureaucracies, as Hartford has begun to do, and getting out of economic development’s way while outlawing suburban exclusionary zoning practices would also alleviate much of this problem.

Traffic congestion problems are said to be created by the land use choices people make that require them to drive their cars to stores and prevent them from walking or taking the train to work. But first recognizing that rail transit has severe limits and most people love the freedom and greater economic opportunities their cars provide them would go a long way to solving these issues.

Unfortunately, other land use "problem” statements prey on people’s emotions by using inflammatory language about the condition we’re in and where we’re headed. For example, Myron Orfield, a former Minnesota state senator who has written and talked widely about regional planning and smart growth, recently touted his conclusion that "we are developing land ten times faster than our population growth” and that we are headed toward looking like New Jersey.

The bit about New Jersey is amusing and it hopes that people up here know New Jersey only from what they’ve seen driving on the northern end of the New Jersey Turnpike. Right now, there is probably a New Jersey advocate who is warning New Jerseyites that they’ll become like Connecticut if they don’t adopt certain land use reforms, hoping that their view of us is only what you can see driving on I95 through Bridgeport.

However, professing that Connecticut is developing land ten times the population rate is a great example of misusing statistics to fit a pre-determined agenda. Mr. Orfield used U.S. Census Bureau data to develop conclusions about the growth of developed land in Connecticut. The Census Bureau divides each state into numerous census tracts and defines "urbanized” census tracts as those with more than 500 people per square mile – roughly equivalent to one housing unit per four acres.

The data shows that the number of "urbanized” census tracts increased by more than 100 percent between 1970 and 2000 while our population grew by 12 percent. However, contrary to what Mr. Orfield states, this does not mean that developed land has increased by 100 percent. For example, if in 1970 you have two census tracts in a town, one with 450 people and the other with 500 people per square mile, you’ve got one that does not meet the definition of "urbanized” and one that does. Then in 2000, both census tracts have 500 people per square mile – one grew a little bit and the other did not grow at all. You could accurately say you have doubled the amount of urbanized census tracts or increased them by 100 percent. But you have not increased developed land by that amount – not even close – because, in this example, you have added only 50 people per square mile in one of the tracts. Looking at the growth of census tracts that meets the "urbanized” definition grossly inflates the real growth in developed land area.

Comparing the growth of urbanized census tracts to population growth also completely ignores the fact that homes are not built to accommodate population growth but to accommodate the growth in the number of households. Reliance on population growth data makes for a more stark comparison but the same data source shows that the number of households grew by 40 percent over the same time period.

Household formations have outpaced population growth because of social and demographic trends, such as increased rates of divorce, single parent households, vacation or second home purchases, and purchases of homes by non-married individuals, all of which have dramatically dropped the average household size over the past thirty years. Sprawl doesn’t cause these trends – sprawl is the result of them.

So, is the correct conclusion from the data that we are overdeveloping our land or are we just keeping pace with real life and the needs of the marketplace?

People make choices every day about where they want to live, work, shop and play. Unfortunately, the misuse of statistics and hyperbolic statements can lead policy makers to pass land use laws and regulations that limit the choices people can make. While it may not fit the agenda of some smart growth proponents, smart growth should be about increasing the choices that all people can make. Many of our land use laws and regulations need serious fixing but questionable use of data to support predetermined agendas accomplishes nothing, or worse.

William H. Ethier, CAE, is the CEO of the Home Builders Association of Connecticut. He’s been a land use and environmental attorney since 1983 and previously worked for environmental organizations.

Copyright 2018 Home Builders & Remodelers Association of Connecticut, Inc.
Web site disclaimerSite Map.
Powered by Wild Apricot Membership Software