Josh LaVigne, Cisco Minthorn UP504

Professor Scott Campbell

Title

When Differences in City Boundaries are Equalized, Are Rust Belt Cities More Competitive with Sun Belt Cities?

Overview

Cities in the Sun Belt are generally thought of as doing better economically than those in the Rust Belt. In these same general areas of the country, people who can afford to live in suburban-type settings (i.e. detached, single-family homes with front and back yards) are generally better off financially than those who live in the city's core. The difference in these two areas is that city boundaries are often much larger in the Sun Belt than in the Rust Belt. In southern and western states, people who live in suburban settings are often technically still living within city boundaries, whereas people in the Rust Belt who live in suburban settings are actually living in the suburbs themselves. The fact that more affluent " suburbanites" live within the boundaries of southern cities but not within the boundaries of Rust Belt cities causes cities in the Sun Belt to appear as though they are more economically competitive relative to cities in the rust belt than they would if we controlled for these differences.

Here is a more visual representation of the spatial typologies we are considering and comparing:

Research Question

Does the greater proportion of a city's area to its metropolitan area evident in cities in the Rust Belt skew economic indicators such that cities in the Sun Belt seem to be doing better than they really are economically?

Hypothesis

When boundaries are equalized, rust belt cities are doing better than they appear, but still not as well, economically, as cities in the Sun Belt.

Units of analysis

The units of analysis are metropolitan areas, central cities, and suburban portions of metropolitan areas.

Data

we will need to obtain are the physical areas of cities, which can be found through the Census; land areas of metropolitan areas; indicators of city economic health (such as unemployment rates, percentage change in housing prices, tax rates, values of goods and services produced, per capita values of goods and services produced, total number of unfilled jobs, and other variables to be determined later), which can be found through the Census, local newspapers, and various city websites, indicators of central city health, which can be found through city websites, state and federal resources, and local newspapers.

Methodology

We will analyze the thirty largest metropolitan areas in the "Rust Belt", defined at first as the East North Central States, West Virginia, Pennsylvania, and upstate New York. We will also analyze the thirty largest metropolitan areas in the "Sun Belt", defined to start as those portions of the contiguous US below 37 degrees North. As our research progresses, we may compare differing definitions of these two regions to see which maximize the distinction between Rusk's elastic and inelastic cities, which provides the basis for this study. We will compare economic indicators (unemployment rates, percentage change in housing prices, tax rates, values of goods and services produced, per capita values of goods and services produced, and total number of unfilled jobs) for metro areas as a whole between the Great Lakes Watershed and the Sunbelt, and determine the differences between the two regions, as well as the differences as central city area increases as a total proportion of metropolitan area.

We will then compare the economic indicators for just the central cities in the two regions. We will also compare the economic indicators for the non-central city portions of metropolitan areas in each region. We will then analyze economic indicators for the central cities, controlling for their area as a proportion of total metropolitan area using simple calculations of central city physical areas over associatedmetropolitan physical areas. We anticipate some of these data will be easier to obtain than others. Unemployment rates for central cities, for example, are commonly used and very accessible figures. Per capita values of goods and services produced in suburban areas, on the other hand, may be very challenging to obtain, especially since we will be seeking data from the same time period for each area.

Preliminary Results

Regression:

ANOVA

Coefficients

In English: The data we have analyzed to date suggest that there is a weak negative correlation, on the edge of statistical significance, between urban elasticity (measured as the proportion by area of a metropolitan area occupied by its central city) and central city unemployment.