Are Center Cities the Engines of Growth for Their Suburbs? A Re-Evaluation of the Economic Relationship Using Evidence from Virginia's Metropolitan Areas
Shuai, Xiaobing, Business Economics
For decades, center cities of metropolitan areas were regarded as the growth engines of their suburbs. However, this paradigm has been shifting in the past twenty years in Virginia, where suburbs have been growing faster than center cities. Consequently, there is a need in economic development communities to re-evaluate the economic relationship between center cities and their suburbs. This paper develops a statistical test to determine the cause-effect ties between these two economies and concludes that Virginia's cities are no longer the growth engines of their suburbs. The opposite is almost true: suburbs are on the verge of becoming the leaders for city economic growth. To further the understanding of what drives the city/suburban economy, the study also tests the cause-effect relationship between population growth and employment growth and finds that population growth still exerts a strong influence on employment growth. This knowledge can be useful in designing policies to promote the economic growth of both cities and suburbs.
Economists advising economic development officials are often asked the following questions: Where should a region's economic development effort be focused? Should it be concentrated on center cities or suburbs? Do jobs follow people or do people follow jobs? What options do economic officials have to accelerate job growth in their regions?
To answer those questions, it is necessary to understand the economic interactions between center cities and their suburbs. For decades, center cities of metropolitan areas were regarded as the growth engines of their suburbs. However, recent data show that suburbs are growing faster in terms of both population and jobs. Take eleven Virginia metropolitan statistical areas (MSA) as an example. From 1990 to 2004, the average annual employment growth rate for center cities was 0.21 percent, while that of suburbs was 1.76 percent (BLS, 2005). Do cities still have strong influences on the economic well-being of their suburbs? Has the interaction paradigm between cities and suburbs shifted from city-leading-suburb to suburb-leading-city?
Several academic papers have addressed these questions empirically (Voith, 1992 and 1998; Brooks and Summers, 1997; Soli-Olle and Viladecans-Marsal, 2004). The general conclusion is that the growth of suburbs is positively related to that of center cities. However, a positive correlation does not imply that center cities are the engines of suburban growth. Both city and suburban economies are affected by macroeconomic trends, national policies, and business cycles; thus they can move in the same direction. In addition, even if the positive correlation reflects a causal relationship, it could be that suburban growth leads to city growth. Another possibility is that a positive correlation is the result of a bi-directional cause-effect relationship. In that case, cities can cause suburbs to grow and vise-versa. The purpose of this paper is to develop a statistical causality test to determine the existence and the direction of the cause-effect relationship of growth of center cities and their suburbs. Specifically, using data from 11 Virginia metropolitan areas spanning 15 years, Granger causality tests will be performed to reach that goal.
Determining the economic relationship between center cities and their suburbs is not the only objective of the study. Finding the policy recommendations that can maximize the effect of the regional development effort is an additional objective. After determining the cause-effect relationship between cities and their suburbs, subsequent analysis will suggest how cities and suburbs can improve their economic performance.
The next section will describe the growth experience of Virginia's cities and their suburbs in the past 15 years. A simple correlation analysis will then be performed. Next, a series of Granger causality tests will be conducted to determine the economic relationships between center cities and their surrounding counties. …