Explore This Section

Cascade: No. 93, Fall 2016

Spotlight on Research: Factors That Sustain Metropolitan Growth*

Marvin M. Smith, Ph.D.

The factors that underlie economic growth in metropolitan regions have been the subject of an increasing number of research efforts. Earlier research focused on the pace of growth. More recent investigations have been concerned with the resilience or capacity of regions to sustain growth paths, even in the face of external shocks. Chris Benner and Manuel Pastor add to the latter body of research by examining the growth spells among the 184 largest regions in the U.S. from 1990 to 2011.1 The authors are particularly interested in the role played by social equity. The following is a summary of their study.


Some of the early research has attempted to identify key variables that influence economic growth on the regional level as well as document the relationship between social equity and metropolitan growth. These efforts have considered different measures of equity such as income inequality; racial inclusion; and spatial segregation by race, income, or jurisdiction. Several studies found that social equity is correlated with economic growth. However, these studies concentrated on the pace of growth. While these studies were inspired by the international literature that reviewed the growth rates of countries, more current international analyses have examined sustained growth or “growth spells” of countries. Yet, little attention has been devoted to the length of the growth spells or the sustainability of metropolitan growth. The authors focus on this area.

Data and Methodology

The authors examine the growth spells in the largest 184 metropolitan regions in the U.S. with a population of 250,000 or more as of the 2010 Census. Benner and Pastor used data from the Quarterly Census of Employment and Wages for their measure of economic growth. They looked at quarter-to-quarter average employment to assess growth. A region was considered to have experienced growth in a quarter “if the total average employment in that quarter was greater (by any amount) than the same quarter in the previous year.”2 The authors defined a region as experiencing a full growth spell “if it experienced at least 12 quarters of uninterrupted quarter-to-quarter employment growth.”


The authors analyzed the 184 regions that had growth spells and calculated the growth in employment and real weekly earnings over the whole period. They found that “more time in growth spells generates more overall employment growth and generally higher earnings.” Given that growth spells are preferable, Benner and Pastor used regression analysis to determine the length of these spells. The authors categorized the variables used into the following domains: external shocks, political fragmentation, inequality and separation, social indicators, and employment structure and institutions.

External shocks. A typical external shock in the analysis is a national recession. The authors found that the longer the national economy was in a recession, the more likely the growth spell would come to an end. This result was statistically significant.

Political fragmentation. The authors pointed out that there are numerous studies that suggest regional collaboration may promote economic competitiveness. Furthermore, the “fragmentation of local government within metropolitan regions has been an important driver of inequality and inefficient public investments.” Benner and Pastor underscored the difficulty in measuring fragmentation. The authors relied on a measure that reflected the concentration of expenditures of all governmental units in a region and was derived by using the square root of the percentage contribution of each jurisdiction to total regional expenditures. The measure they used suggests that political fragmentation might be a drag on sustained growth and was slightly significant.

Inequality and social separation. The authors used several variables to evaluate the influence of inequality and social separation on growth spells. As one measure of inequality, they used the distribution of income among various metropolitan household income classes. They found that as the distribution of income becomes more unequal, the more likely a region’s growth spell will be shortened. Another measure used was the proportion of a region’s minorities in the middle class. The results revealed that “regions with a higher percentage of minorities in middle class income brackets are more likely to have longer growth spells.” To capture the effect of social separation, the authors used a measure of residential segregation (dissimilarity index) and found that more segregated regions have shorter growth spells. They also used the ratio of city to suburban poverty rates. This measure indicated that a higher city to suburban poverty differential was also associated with shorter growth.

Social indicators. The authors chose educational attainment and levels of immigration to represent the impact of social indicators on growth spells. The two education measures were the “proportion of the population 25 years and older with a bachelor’s degree or higher and the proportion with at least a high school degree but less than a bachelor’s degree.” Of the two education measures, the latter measure was found to be statistically significant and implied that regions with a larger proportion of the middle-educational population were less likely to have their growth spell shortened.

Another social indicator that was relied on was the extent of immigration; more specifically, the percentage of the foreign-born (immigrant) population in each region. This variable suggested that a higher proportion of immigrants had a shortening effect on the length of growth spells and was also significant.

Employment structure and institutions. The authors also considered some broad measures of industrial structure in the region, such as the percentage of workforce employed in construction, in manufacturing, and in public administration, in addition to the percentage of the workforce covered by unions, which represents the influence of an economic institution. The results of these variables were mixed. The measure for employment in public administration was associated with longer growth spells and was significant. However, the measures for employment in manufacturing and construction did not have a significant effect on the length of growth spells. While the measure for unionization was associated with shorter growth spells, it was not significant.

Integrated Regression Model

In the discussion, the authors considered the effect of the variables on growth spells separately. They then combined the various variables into a single regression. Included in this regression were controls for census region, per capita income, and metro size. The authors discovered that the variables interacted with one another in such a manner that, in some cases, their impact separately had the opposite effect in the integrated regression. Also, while some variables were significant separately, they were insignificant in the combined regression.

After making some modifications to ensure that the regression was technically sound, the authors did discover some noteworthy findings. The percentage of foreign born immigrants and the share of the workforce in manufacturing were associated with shorter growth spells. Although the authors also alluded to education levels and economic structures and institutions as being important to maintaining growth, the overriding result is that inequality has a dampening impact on growth spells and is highly significant. Thus, while several factors influence regional growth spells, the main takeaway from their research was “regions that are more equal and more integrated — across income, race, and place — are better able to sustain growth over time.”

  • * The views expressed here do not necessarily represent the views of the Federal Reserve Bank of Philadelphia or the Federal Reserve System.
  • 1 Chris Benner and Manuel Pastor, “Brother, Can You Spare Some Time? Sustaining Prosperity and Social Inclusion in America’s Metropolitan Regions,” 2015, Urban Studies, 52(7), pp. 1339–1356.
  • 2 This year-over-year measure allowed for the adjustment for seasonal variation in employment.