Census Note 1: The Urban TurnaroundFannie Mae Foundation Census Note 01 (April 2001)* The Urban Turnaround:
A Decade-by-Decade Report Card on Postwar Population Change
in Older Industrial Cities Overview The postwar years were not kind to America's older industrial cities. Most endured substantial population decline, particularly during the 1970s when municipalities such as New York, Chicago, Detroit, and Philadelphia each lost enough residents to populate a good-sized city. This Census Note reveals that many older industrial cities have rebounded considerably from the traumatic population losses of the 1970s. During the 1970s, all of the 36 cities in our study lost population and 29 experienced their worst decade in the postwar period. By the 1990s, 15 were growing again and the remainder were losing population at often sharply reduced rates. Not one of the 36 cities experienced its worst postwar decade during the 1990s. For the first time since World War II, the combined population of these cities increased. The analysis shows that the 1990s was the best postwar decade for the 36 cities as a group. A summary index that assigns a grade to each decade based on numeric population change gives the 1990s a "B." By contrast, the 1950s and 1980s received a C+, the 1960s a C, and the 1970s an F. A grading system based on percent population change yields similar results. Methods Selecting the Cities Cities in this analysis were selected by a two-step process. The first step identified the 50 most populous cities in the nation as of 1950. [1] This initial group includes all cities with a 1950 population of 200,000 or more. The second step screened out those places that did not experience at least two decades of population decline in the postwar period. This filter excludes the following cities, most of which are rapidly growing Sunbelt metropolises: Columbus, Dallas, Fort Worth, Houston, Indianapolis, Jacksonville, Long Beach, Los Angeles, Memphis, Miami, Oklahoma City, Omaha, San Antonio, and San Diego. [2] The final list includes 36 cities that had 1950 populations ranging from 7.9 million (New York City) to 203,000 (Worcester, MA). Together, these cities contained 30 million residents and accounted for 20 percent of the nation's total population in 1950. By 2000, they were home to 25 million residents and just 9 percent of all Americans. Most of the cities on the final list are located in the Northeast and Midwest, which together accounted for almost two-thirds, or 23 of the 36 municipalities. The list includes all of the major Rustbelt cities that are frequently associated with postwar urban decline such as Cleveland, Detroit, St. Louis, Pittsburgh, Newark, and Buffalo. (The complete list of cities is provided in a table and discussed in the Results section.) Measuring Population Change Population data for each city are from the decennial censuses of 1950-2000. [3] In the numeric and percent changes in city populations were calculated for each of the five postwar decades. These measures were used to rank the five decades and identify the best and worst decade of population change for each city. [4] A summary index that works like a college grade-point average (GPA) was used to compare population performance across the decades for the entire set of cities. The summary index was created using the following steps. First, the five decades were ranked for each city, with a rank of "4" assigned to the decade with the highest numeric population increase. [5] The lowest rank of zero was assigned to the decade in which the city lost the most population. At the end of this first step, each decade contained 36 city ranks ranging from "0" to "4." Next, the 36 ranks were averaged to obtain a GPA for each decade. The GPA was then converted into a letter grade. [6] Each city receives the same five rank scores (0, 1, 2, 3, and 4) over the five decades covered. What varies is in which decade a city registers its high, low, and three middle ranks. A city's best decade for growth (or its slowest loss) gets a "4." A city's worst decade, or its biggest loss, is assigned a "0." A city's three middle decades are also ranked "1," "2," and "3." Most cities received their lowest rank during the 1970s, which is why, in large part, the decade gets the lowest GPA and a corresponding letter grade of "F." No city during the 1990s received a lowest rank of a "0," which helped lift the decade to a high GPA. Note that the cities themselves do not receive a grade. Rather, the grade is assigned to the group performance of cities by decade derived from a cumulative GPA of between "0" and "4." A city's individual population change by decade is not compared to other cities, but is instead rated against its own historical performance over the five decades examined. The grading system distills data from all 36 cities into one summary measure while dampening the effects of extreme values of population change, such as those caused by annexations. [7] Likewise, it prevents large numeric population changes for huge metropolises such as New York and Chicago from distorting the analysis. Results Growth Grades
"Growth grades" based on numeric population change reveal that older industrial cities, as a group, performed better during the 1990s than in any other decade since World War II (table 1). As a group, the 36 cities received a B (average ranking of 3.1) during the 1990s. This is a big improvement over the F (average rank 0.3) received by these cities during the 1970s. Switching to the grading system based on percent population change produces similar results, although under this system the 1950s and 1990s tie for the best postwar decade, with each receiving a B- (based on a 2.8 GPA). Number of Cities Losing and Gaining Population The 1990s also compare favorably to earlier decades based on the number of cities experiencing population growth and decline (figure 1). During the 1990s, 15 cities experienced population growth, a greater number than in any other decade. A comparison with the 1970s is particularly illustrative of how far these cities have come in recent decades. During the 1970s, all cities in the study lost population. Best and Worst Decades Among the 36 cities included in the analysis, not one experienced its worst postwar decade during the 1990s (figure 2). In comparison, 1 city experienced its worst decade in the 1960s, 3 each during the 1950s and 1980s, and 29 in the 1970s. Conversely, 15 cities experienced their best postwar population change during the 1990s. The second best decade was the 1950s, during which 12 cities had their best population performance. [8] Numeric Population Change As previously noted, analysis of raw population data can be distorted by annexations or large population changes for individual cities. With this caveat in mind, however, it is informative to note that such an analysis reinforces the finding that the 1990s was the best postwar decade for our 36 municipalities. During the 1990s, the cities as a group grew by roughly 580,000 persons. [9] In all other postwar decades, the combined population of the cities decreased. Total population declined most during the 1970s, when the group of municipalities lost 3.6 million persons.
Turnarounds in Individual Cities Population turnarounds in some individual cities have been remarkable (table 2). Chicago went from losing 360,000 residents during the 1970s to gaining more than 100,000 residents in the 1990s. Similarly, New York City lost 800,000 residents in the 1970s, but gained almost 700,000 during the 1990s, and for the first time has more than eight million people. Among the cities that have not returned to growth, the rate of population loss has slowed dramatically. Cleveland went from losing a quarter of its population during the 1970s to losing only five percent in the 1990s. Newark, which dropped roughly 15 percent in both the 1970s and 1980s, remained essentially unchanged in the last decade. Thus, even those cities that are still losing residents have made substantial strides toward achieving population stability, suggesting that the wave of population decline associated with postwar urban-industrial restructuring might have run its course. Analysis Factors Influencing Population Change Although it is beyond the scope of this analysis to identify the causes of population trends in the cities, several factors likely contributed to their improved population performance during the 1990s. The unusually robust economy of the mid- to late 1990s likely supported urban population growth -- or at least greater population stability. Cities with large economic bases in outperforming sectors such as advanced services were particularly well positioned to take advantage of the strong economy. Immigration also bolstered the population of many cities in the analysis. Although data from Census 2000 on the foreign-born population have not been released, several previous studies point to the important role of immigration in urban population growth (Myers 1999; Frey and Fielding 1995). For example, were it not for the arrival of more than 800,000 immigrants, New York City would have lost population during the 1980s rather than gaining a quarter of a million residents (New York City Department of City Planning 1996). Finally, differences in population coverage rates between censuses might also have artificially inflated population growth during the 1990s. Preliminary analyses indicate that census coverage improved between the 1990 and 2000 censuses (U.S. Bureau of the Census 2001b). [10] Because the net undercount rate actually increased between 1980 and 1990, changes in census population coverage likely have the greatest effect on comparisons between the 1980s and 1990s. [11] Implications of Population Change for City Health Renewed population growth has several positive implications for the health of older industrial cities. First, it could lead to increased representation and government funding. By supporting more robust housing demand, renewed population growth could also lead to reclamation of deteriorating or abandoned housing stock and renewal of inner-city neighborhoods. Population gains also induce other forms of economic activity, such as the formation of businesses that meet the consumer needs of new residents. Furthermore, population growth increases city tax revenues, although its net fiscal impact hinges on the socioeconomic characteristics of new residents in addition to their numbers.  It is important to note, however, that even in this post-1970s period of urban turnaround, most of the cities in this study (21 out of 36) are still losing population. This is not surprising, given that the places in this analysis were selected because they had experienced at least two decades of population loss over the past five, and are thus prone to decline. But even as many cities continue to lose people, the rate of loss has slowed dramatically (table 2). Only five cities in the 1990s experienced double-digit percentage population decline. By contrast, 29 cities had double-digit losses (four by more than 20 percent) during their worst decade. Most 1990s city population losses were slight and are accounted for by shifts in household structure rather than the wholesale abandonment of dwellings. It is also important to put urban population decline in regional context. [12] Places such as Detroit, Cleveland, and St. Louis are at the center of slow-growing metropolitan areas. Even as they lose people, their relative position in the region may slip very little. In a few instances, the population gains made by cities in the 1990s actually kept pace with, or, in the case of New York (U.S. Bureau of the Census 2001c) exceeded, regional growth rates. The encouraging findings of the study are also tempered by the limitations of the analysis. Total population change is a single -- and admittedly limited -- dimension of city health. Changes in resident characteristics are likely as important as changes in total population. For example, scholars have pointed to growth in city poverty populations and extremely poor neighborhoods as signs of mounting urban distress (Kasarda 1993; Frey and Fielding 1995). Kasarda (1993) showed that some of the same cities that we found to have improved population trends also experienced substantial increases in the number of their residents living in high poverty neighborhoods in recent decades. Aside from trends related to total population and resident characteristics, long-term changes in social, economic, and fiscal conditions also need to be analyzed prior to drawing firmer conclusions regarding the revitalization of older industrial cities. As noted previously, improvements in city population trends occurred during the longest economic expansion in American history. The population turnaround also happened during a period in which changing immigration policies renewed an important source of urban population growth that had been dormant for decades. Interruption of these important forces, or strengthening of countervailing trends such as the settlement of new immigrants directly in suburban communities, will affect future prospects for urban population growth. Thus, an additional consideration in interpreting the findings of this analysis is whether the recent reversal in city population fortunes can be sustained in a less favorable environment. Authors Patrick A. Simmons is Director of Housing Demography at the Fannie Mae Foundation. He is editor of the book Housing Statistics of the United States published by Bernan Press. Robert E. Lang is Director of Urban and Metropolitan Research at the Fannie Mae Foundation. He is author of the book Edgeless Cities: Exploring the Elusive Metropolis to be published early next year by the Brookings Institution. The authors thank Frank Popper and Elvin K. Wyly for invaluable comments on a draft of this Census Note. *About the Census Notes Series The Fannie Mae Foundation's Census Notes series provides timely analyses of Census 2000 data in order to stimulate discussion and further research. Although Census Notes are reviewed internally and on an informal basis externally, they have not been subject to the formal process of external peer review that is used for the Foundation's research publications. Therefore, they should be viewed as works in progress and their findings considered preliminary. Footnotes [1] 1950 is a significant starting point because it represents the population and economic peak for many of the nation's older industrial cities. The 36 cities in our study grew by almost 2.5 million people during the 1940s, before commencing a long period of population decline. [2] These places also tend to grow by urban annexation. Inter-decade expansion of city boundaries can distort the analysis by making additions of existing households appear as actual population growth. Thus, the removal of these cities dampens this distortion. [3] Data on population and land area for 1950 through 1990 were obtained from Gibson (1998). Population data for 2000 were obtained from U.S. Bureau of the Census (2001a). [4] Note that the "best" postwar decade for some of these cities was a decade in which it lost population. In such cases, the numeric or percent decrease was less than in any other postwar decade. [5] For a city that lost population in each postwar decade, the highest score was assigned to the decade in which the fewest residents were lost. [6] The cutoffs for each grade are as follows: 3.85-4.0, A; 3.50-3.85, A-; 3.15-3.50, B+; 2.85-3.15, B; 2.50-2.85, B-; 2.15-2.50, C+; 1.85-2.15, C; 1.50-1.85, C-; 1.15-1.50, D+; 0.50-1.15, D; 0.0-0.50, F. [7] Urban annexation was much more common in the 1950s and 1960s (Rusk 1993), and therefore its effect (which is mostly mitigated by the screening and indexing methods used here) tends to favor earlier decades in the analysis. [8] Half of these 12 cities increased in land area by at least 25 percent during the 1950s (analysis by the authors using data on city land area from Gibson 1998). At the time this Census Note was prepared, information was not available on changes in city land area between 1990 and 2000. [9] As shown in the next section, the huge increase in New York City is largely responsible for this aggregate gain. [10] Eugene Ericksen (2001a) found the biggest numerical improvement in the net undercount between 1990 and 2000 was in Texas and California. Together they accounted for almost two-thirds (64 percent) of the improvement of the net undercount. The 2000 net undercount was higher in 12 states in the Northeast and Midwest (Connecticut, Indiana, Iowa, Maine, Massachusetts, Nebraska, New Hampshire, New Jersey, Pennsylvania, Rhode Island, Vermont, and Wisconsin). Ericksen (2001b) also compared undercounts in four cities that are included in this analysis: Atlanta, Chicago, New York, and Philadelphia. He estimated that the biggest improvement occurred in New York, which missed 3.2 percent of people in 1990 and 2.3 percent in 1990. The rankings for these four cities would not have changed had the census coverage remained the same in 2000 and 1990. [11] However, the difference between urban population change in the 1980s and 1990s is much too large to be mostly attributed to an improvement of census coverage in the 1990s. [12] The authors are grateful to Elvin K. Wyly for this point. References Ericksen, Eugene. 2001a. A Comparison of 1990 and 2000 Net Undercount by State. Analysis provided to USA Today on figures derived from Census 2000 Redistricting Data (Public Law 94-171) Summary. March 28. Ericksen, Eugene. 2001b. A Comparison of 1990 and 2000 Net Undercount for Five Cities. Analysis provided to the USA Today on figures derived from Census 2000 Redistricting Data (Public Law 94-171) Summary. March 28. Frey, William H., and Elaine L. Fielding. 1995. Changing Urban Populations: Regional Restructuring, Racial Polarization, and Poverty Concentration. Cityscape 1(2):1-66. Gibson, Campbell. 1998. Population of the 100 Largest Cities and Other Urban Places in the United States: 1790 To 1990. U.S. Bureau of the Census, Population Division Working Paper No. 27, June. Kasarda, John D. 1993. Inner-City Concentrated Poverty and Neighborhood Distress: 1970 to 1990. Housing Policy Debate 4(3):253-302. Myers, Dowell. Immigration: Fundamental Force in the American City. Housing Facts & Findings 1(4):3-5. Washington, DC: Fannie Mae Foundation. New York City Department of City Planning. 1996. The Newest New Yorkers: 1990-1994. New York: New York City Department of City Planning. Rusk, David. 1993. Cities Without Suburbs. Baltimore: Johns Hopkins University Press. U.S. Bureau of the Census. 2001a. Census 2000 Redistricting Data (Public Law 94-171). Accessed at http://www.census.gov/clo/www/redistricting.html in March 2001. U.S. Bureau of the Census. 2001b. Report of the Executive Steering Committee for Accuracy and Coverage Evaluation Policy. Released March 1, 2001. U.S. Bureau of the Census. 2001c. Population Change and Distribution. Census 2000 Brief. Washington, DC. April 2. |
| All Content © 2007 Fannie Mae Foundation. All rights reserved. |