FINDING #1 UNACCEPTABLY HIGH UNEMPLOYMENT REMAINS IN ONE IN SIX CENTRAL CITIES. Despite a dramatic drop in the overall urban unemployment rate over the past 6 years, many central cities face unemployment rates much higher than that of the Nation as a whole-1 out of 6 central cities (17 percent) has an unemployment rate 50 percent or more above the national rate,9 about 1 in 12 has a rate 75 percent or more above the national average, and about 1 in 15 central cities has a rate 100 percent above the national average. A total of 95 central cities had rates of 6.75 percent or more in 1998 (50 percent above the national average rate for that year), 64 had a rate of 7.9 percent or more (75 percent above the national rate), and 37 had rates of 9 percent or more (100 percent above). In other communities, including economic giants such as New York City and Chicago, low citywide unemployment rates mask pockets of very high unemployment in particular neighborhoods. And although unemployment rates have dropped in the six years of the recovery in most of these cities, they remain unacceptably high in the context of the lowest peacetime unemployment in over 40 years for the Nation as a whole. In part, these patterns reflect the loss of both households and jobs to suburban areas over the past quarter century. Contrary to the narrow image of distress only in the "rust belt," the high unemployment areas, most of which are small or mid-sized cities, are found throughout the Nation, from farm cities to former industrial giants, from timber towns to former mining centers: Yuma, Arizona saw its unemployment rate fall modestly, from 19 percent in 1992 (pre-recovery) to 18.8 percent in 1998; Madera, California's rate fell more than 5 percent over the last six years but remains high at 18.3 percent; North Chicago, Illinois saw a modest drop from 10.7 percent in 1992 to 10.3 percent last year; and unemployment in Yakima, Washington has declined from 13.7 percent to 10.2 percent, still more than twice the national average in 1998.
The Nation's central cities have seen sharper drops in unemployment than either suburbs or rural areas over the past six years that encompass the economic recovery. The main reason for this is that cities were home to a greater number of available but jobless (or underemployed) workers-the "reserve labor force"-at the start of the recovery. And many cities have shown dramatic declines in part because they had so far to drop; their rates were extraordinarily high in 1992, for example.
Figure 1 Unemployment Has Fallen Nationally,
Source: Bureau of Labor Statistics
Note: for a discussion of "doubly burdened" cities, see Finding #4.
It is remarkable, then, that 95 central cities still faced unemployment rates that were 50 percent higher than the national average rate of 4.5 percent in 1998 (see Appendix, Table 8 and Figure 3). These are measures of relative distress, to be sure: Flint, Michigan's 1998 rate of 9.8 percent, while still too high, is far better than the 20.0 percent rate facing Flint's residents in 1992, prior to the recovery. Many other cities, large and small and in between, show this pattern of steep declines over the same period that nevertheless leave rates well above the nationwide average, example: Bakersfield, California (11.6 percent to 8.9 percent); Miami, Florida (15.0 percent to 9.6 percent); Joliet, Illinois (13.2 percent to 7.4 percent); Newark, New Jersey (16.6 percent to 9.9 percent); and McAllen, Texas (16.5 percent to 12.7 percent).
Central City Unemployment Rate by Population Size
Central cities in a second category started the recovery with high rates and have seen those drop only modestly since 1992. These communities appear in a variety of regions and reflect a diverse array of economic activities and challenges, for example: Kennewick, Washington, where unemployment dropped only from 8.5 percent to 7.4 percent in the last six years; El Paso, Texas (11.2 percent to 9.6 percent); St. Louis, Missouri (8.3 percent to 7.2 percent); Kansas City, Kansas (8.5 percent to 7.3 percent); North Chicago, Illinois (10.7 percent to 10.3 percent); Albany, Georgia (11.2 percent to 8.9 percent), Washington, D.C. (unchanged at 8.6 percent); Yuma, Arizona (19.0 percent to 18.8 percent); and Florence, Alabama (9.7 percent to 8.3 percent). Whether struggling with a decline in farm employment, the loss of timber revenues, outdated manufacturing processes, or other hurdles, these cities and others must find a competitive niche-a role in the larger regional race to compete globally.
Population losses include the extreme cases of East St. Louis, Illinois, which lost 30.1 percent of its population during 1980-96; St. Louis, Missouri (22.4 percent); Gary, Indiana (27.1 percent); and smaller cities like Youngstown, Ohio (24.3 percent); Johnstown, Pennsylvania (26.3 percent); and Wheeling, West Virginia (22.7 percent). These places, as well as central cities with significant but more modest declines, continued to lose population over the most recent period (1990-96). Washington, D.C., for example, saw its population shrink 14.9 percent between 1980 and 1996, with most of that loss-10.5 percent of the city's total population-occurring in the last 6 years alone. (See Appendix, Table 9 and Figure 4.)
Figure 2 Overall Central City Population Growth Lags Suburbs
Source: Bureau of Labor Statistics
These patterns are in stark contrast to the broader trend: the Nation as a whole grew by 17 percent, and two-thirds (65.9 percent) of central cities enjoyed population growth during the period. A number of cities in the South and West grew by 100 percent or more over the past two decades. The most rapid gains were in small southern and western cities like Jackson, North Carolina, which grew almost 310 percent between 1980 and 1996. But some big cities ballooned, too-Las Vegas by 129 percent, for example-over that period.
Population Change in Central Cities by Population Size,1980-96
Source: Bureau of the Census
A rapid and significant loss of population has many effects on local communities. First, it is a proxy for loss of the workers and consumers who sustain and grow economies. Second, it entails a loss of the tax base that cities need to invest to keep service quality high and the business climate healthy. Worse still-harder to measure but no less real-is the loss of hope and vitality, the blow to a community's confidence that it can bounce back from hard times. People move where jobs are, not where they used to be-or where they are perceived to be chronically missing.
It is not surprising, then, that "shrinking cities" tend to have higher rates of poverty and unemployment than cities with a growing or stable population base. That association-a self-reinforcing downward spiral-is clearest in the 116 central cities (1 in 5 central cities) that shrunk by 5 percent or more in just the 6 most recent years (1990 to 1996) for which census data are available (see Table 3 below). A striking 55 percent (64) of those communities are places with estimated poverty rates of 20 percent or more in 1995. And 6 years into what may be the greatest peacetime economic expansion in America's history, one-third (31 percent) of those population-losing central cities had unemployment rates at least 50 percent higher than the Nation's average rate for 1998. By comparison, just 21 percent of the population gainers over that period show high poverty in 1995, and fewer than 1 in 7 (14.3 percent) of those gainers show high unemployment, relative to the Nation, for last year. Population loss is both a symptom and a cause of economic decline.
Central City Poverty Rate and Unemployment by Population
Nationally, poverty has dropped during the recovery, and that decline has included central cities (see Table 4 below). Higher employment rates in most cities are helping many families, including former welfare recipients, to escape the deprivation and insecurity associated with a life in poverty.
Trends in Poverty Rate by City/Suburb, 1992-97
Still, the poverty rate in central cities remains more than twice as high as that in suburban areas (18.8 percent versus 9 percent as of 1997), and some central cities continue to face much higher poverty rates than the national average. For this study, we looked first at how central cities began the decade: extraordinarily high poverty rates tend to reflect structural barriers to participation in the changing economy-barriers such as a large "skills gap" in the workforce (or among workers in particular neighborhoods or racial/ethnic groups), rapidly dis-invested and blighted parts of a city that have trouble attracting investment even where significant market potential exists, and high rates of crime that are both symptom and cause of chronic economic disadvantage. High poverty rates are thus not only a sign of past distress but an indicator of hurdles that a community must overcome if its regional economy is to ride the prosperity trend enjoyed by the Nation as a whole over any period of recovery and expansion.
We then looked at 1993 and 1995 census estimates at the county level, and we estimated poverty rates for central cities based on the 1990 relationship between city and county poverty for each locality. We considered which central cities that showed high poverty rates in 1990 remained on that list for 1995, despite the impacts of the first phase of our economic recovery.
Table 5 (below) shows how many cities in select size categories had extraordinarily high rates of poverty in 1995, and Table 12 (Appendix) highlights particular cities with very high rates.
Number of High Poverty Central Cities by Population Size
In some central cities, poverty was not only extraordinarily high at the start of the decade-it was highly concentrated in particular neighborhoods. Forty percent (40 percent) and higher rates are generally considered by analysts to reflect "extremely high" or "ghetto" poverty rates.10 Key cities in a wide array of States began the decade with high proportions of their poor living in 40 percent-plus poverty census tracts, for example: Tuscaloosa, Alabama (53.1 percent); West Memphis, Arkansas (67.6 percent); Fresno, California (41.2 percent); Hartford, Connecticut (40.1 percent); Miami, Florida (37.9 percent); Albany, Georgia (59.2 percent); East St. Louis, Illinois (69.9 percent); East Chicago, Indiana (50.9 percent); Monroe, Louisiana (83.6 percent); Detroit, Michigan (54.5 percent); Minneapolis, Minnesota (39.9 percent); Hattiesburg, Mississippi (57.9 percent); Camden, New Jersey (59.5 percent); Buffalo, New York (34.2 percent); Youngstown, Ohio (52.6 percent); Eugene, Oregon (28.8 percent); Pittsburgh, Pennsylvania (39.4 percent); Memphis, Tennessee (46.8 percent); Laredo, Texas (60.5 percent); and Milwaukee, Wisconsin (51.8 percent).
Why are estimates and other updates of those 1990 figures so important? The 1990 census, which measured such traits as household income and poverty status in 1989, is now a decade old. Most significantly, 1990 census data do not reflect the last 6 years of record economic growth in our country.
Detailed local poverty and other data will not be available for the Nation as a whole until the 2000 census is published several years from now. Between decennial census takings, however, several sources of data are available, for example:
When these data are analyzed for many of the central cities discussed in this report, it becomes clear that poverty rates tended to rise between 1989 and 1993 and mostly leveled or dropped modestly between 1993 and 1995, as the economic recovery accelerated in most regions. Still, the combined evidence strongly suggests that central cities with extraordinarily high poverty rates in 1990 continue to face unacceptably high rates (see Appendix, Table 12). Even with a strong national economy, many of these places have a long way to go. To highlight just a few of the more striking findings:
FINDING #4 ONE IN SEVEN CENTRAL CITIES FACES "DOUBLE TROUBLE." Seventy-four (74) central cities, or one in seven (14.3 percent), face continued high unemployment relative to the Nation as a whole, plus either significant long-run population loss or persistently high poverty rates or both.11 Not surprisingly, then, the new urban challenge is not limited to a handful of large cities in a few parts of the Northeast and Midwest-rather, it confronts all regions of the country and a significant number of small and mid-sized communities. These doubly burdened central cities experience both high levels of distress and, in many cases, have enjoyed much more modest recovery than the Nation as a whole over the past six years. For example: Camden, New Jersey, with an unemployment rate almost three times the national average in 1998; Gary, Indiana, with high unemployment, almost one-third of its residents in poverty, and a staggering population loss of 1 in 4 residents between 1980 and 1996; Anniston, Alabama, with a 12.7 percent loss of population over that period and poverty estimated at 28 percent in 1995; Niagara Falls, New York, which lost almost one-fifth (18.2 percent) of its population during 1980-96, had unemployment (10.4 percent) double the national average last year, and an estimated 22 percent of its residents in poverty in 1995; and cities in the West and Southwest, where poverty is high and job creation is lagging a population explosion, despite considerable new investment activity, such as McAllen, Texas, where unemployment (12.7 percent) was three times the national rate last year and poverty at 34.4 percent in 1995, alongside a 56 percent growth in population during 1980-96; or Madera, California, which grew 64 percent over that period and faced a 33 percent poverty rate in 1995, as well as unemployment (18.3 percent) four times last year's national average.
"Doubly burdened" cities are largely outside of the recovery along three dimensions: a long-run loss of people that has continued in recent years, extremely high poverty rates for the city as a whole (not merely a few isolated neighborhoods), and comparatively high unemployment that has fallen too little in the past 6 years, despite the recovery. All have unemployment rates that were significantly higher (50 percent or more) in 1998 than the average monthly rate for the Nation (4.5 percent). In addition, doubly burdened cities either face high (20-plus percent) poverty rates, or suffered significant population loss (5 percent or more) between 1980 and 1996, or both.
Our analysis identified 74 of the 539 central cities as doubly burdened cities (see Appendix, Table 13 and Figure 5). On this list are places that have attracted national attention, such as East St. Louis, Illinois, which lost almost one-third (30.1 percent) of its population between 1980 and 1996, and where a staggering 44.3 percent of that shrinking population lived below the poverty line in 1995; or Flint, Michigan, with an unemployment rate 90 percent higher than the national rate and a population loss of 1 out of every 6 people (15.5 percent) during 1980-96.
But on this list of doubly burdened central cities, too, are communities that rarely capture national headlines as places in trouble, for example: Fresno, California; Lake Charles, Louisiana; Saginaw, Michigan; Buffalo, New York; Youngstown, Ohio; Beaumont, Texas; Yakima, Washington; and Parkersburg, West Virginia. As we outline below, this list is remarkable not only for its length but for the diversity of population sizes, economic bases, and regions represented.
MOST DOUBLY BURDENED CENTRAL CITIES ARE SMALL OR MID-SIZED CITIES. Sixty-six (66) percent of the central cities that are doubly burdened as we describe above are small or mid-sized, with populations of 100,000 or less. The challenges in a handful of the larger cities in this category relate mainly to extending the recovery into the most distressed neighborhoods-the deepest pockets of poverty.
Most of the doubly burdened central cities are small communities with populations of less than 50,000 (such as Anniston, Alabama; Madera, California; Alton, Illinois; or Johnstown, Pennsylvania) or mid-sized cities with 50,000 to 100,000 persons (such as Pine Bluff, Arkansas; Albany, Georgia; Lake Charles, Louisiana; Camden, New Jersey; Youngstown, Ohio; and Galveston, Texas).
Although some of the "old urban crisis" symbols, including the 1960's riot cities of Newark, New Jersey and Washington, D.C., are on our list of doubly burdened places, most of the list belongs to regional cities that tend to have narrower economic bases, and a tougher time adapting to rapid economic change, than those metropolis cities. The small to mid-sized places are fighting hard for job-creating investment.
Table 6 Most "Doubly Burdened" Central Cities Are Small or Mid-Sized
Number of Doubly Burdened Central Cities by Population Size
The handful of very large central cities, with populations of 500,000 or more, on our list of doubly burdened places tend to differ markedly from the small to mid-sized communities. The large places have, by and large, enjoyed notable drops in unemployment during the recovery thanks to a rebound in investment and a base of skilled workers. This is true in Baltimore, Cleveland, Detroit, and Miami, for example, which posted big drops in joblessness and are perceived to be enjoying an economic renaissance. In these cities, it is the most deeply distressed neighborhoods that are struggling-enough to depress the citywide numbers that we generated and examined.
The recovery has taken hold in these large communities, but it is marching ahead unevenly. The problem is one of persistent inequality defined in distinctly spatial terms. Joblessness, poverty, and minority populations are highly concentrated. The challenge in these cities is bringing the fruits of the economic recovery to the large numbers of people living in the most distressed neighborhoods. Cities like Gary, Indiana and Trenton, New Jersey, on the other hand, are starved for investment-poverty and joblessness in those places reflect more citywide, and not merely neighborhood-level, distress.
The Nation's Capital, Washington, D.C., also a doubly burdened city by our definition, is an intermediate example, with very serious patterns of race and income segregation but more new and varied business activity than that of the formerly industrial cities of the rust belt. Much of DC's large population loss (14.9 percent) over the last two decades is out-migration to job-rich suburbs in a metropolitan region that is healthy overall, if troubled by persistent patterns of inequality. As we discuss below, however, the recovery is thus far anemic in D.C. itself-unemployment dropped just 1.3 percent (8.5 percent to 7.2 percent) in the past 6 years, and it remains well above last year's national average rate of 4.5 percent. El Paso, Texas, is, likewise, a still struggling large city. There has been strong population growth over the 1980-96 period (41.1 percent), but high poverty (an estimated 29.3 percent in 1995) persists, and the unemployment rate has dropped just 1.3 percent over the past 6 years.
ALMOST HALF OF ALL STATES ARE AFFECTED.Twenty-three (23) of 50 States have at least one central city affected by two or more distress indicators. Due to patterns of migration and economic change, some States are home to more than their share of these central cities-for example, California is home to 11; Texas to 10; Ohio to 7; Illinois to 6; Florida and New Jersey to 5; and Michigan, New York, and Pennsylvania to 4. Doubly burdened cities in the "rust belt" tend to be those former industrial powerhouses that lost investment, jobs, and population over the last two decades. Doubly burdened cities in the South and West tend to be smaller and started with a different economic base, such as farming or mining, although those communities have also seen people and jobs leave in dramatic numbers. A number of high unemployment cities in the West and Southwest tend to be gaining population rapidly, like those regions generally, but not generating sufficient job growth to overcome high poverty and unemployment rates, despite the recovery. A population explosion in many of these western and southwestern cities can present challenges quite unlike those posed by the demographic implosion of shrinking places in the Northeast, Midwest, and South.
As the attached map (see Appendix: Figure 6) indicates, doubly burdened central cities are found throughout the Nation. Not surprisingly, many are in the "rust belt" areas of the Northeast and Midwest. Clearly, a significant number of the older, formerly industrial, small to medium-sized cities are still struggling to find their places in a changing global economy. These include such cities as North Chicago and East St. Louis in Illinois; Cleveland and Lima in Ohio; Lawrence in Massachusetts; Camden, Newark, and Trenton in New Jersey; Buffalo, New York; Johnstown, Pennsylvania; East Chicago and Gary in Indiana; Flint, Pontiac, and Saginaw in Michigan; and others like them.
Number of Doubly Burdened Central Cities by State, including the District of Columbia
Doubly burdened central cities face somewhat different challenges in the Southwest or West. Part of this can be explained by demographic trends. These are the fastest growing urban areas of the country. Though a number of cities in the growth regions are struggling economically, many of these are also adding new workers and consumers.
Except for the handful of timber or mining towns, such as Lakeview, Oregon or Leadville, Colorado, western cities have not endured the rapid loss of population and high-wage job base that buffeted rust belt cities in the Northeast and Midwest. Economic woes relate to sustaining job growth to keep up with population growth. Shocks such as the closing of military bases are apparent in communities such as San Bernardino, but it is in California's small and medium-sized agro-cities in the Central Valley, such as Fresno, Madera, Merced, Porterville, Stockton, and Tulare, that the seasonal strains of a drop in demand for farm labor relative to the number of available, low-skilled workers are seen. The economic base is undiversified. Furthermore, when farm demand slackens, the multiplier effect depresses winter retail trade and services.
These communities, and those in South Texas near the U.S./Mexico border, are fast-growing places that have not tracked national trends because they cannot grow their economies fast enough to match the explosive growth in population-82 percent in the case of Fresno between 1980 and 1996 and 80.3 percent in the case of Laredo-over that period.
In El Paso, Texas, new immigrants, though often poor, are feeding a regional job economy, bringing skills and entrepreneurial spirit to invigorate the city as generations of immigrants before them have. Despite a loss of jobs in the military, apparel, and retail sectors, that city's population grew over 41 percent during 1980-96, at the same time that Gary, Indiana's dropped 27 percent and East St. Louis' dropped more than 30 percent. Other South Texas cities reflect the same, tell-tale pattern: staggering rates of poverty at the beginning of the decade-such as Brownsville (43.9 percent) and Laredo (37.3 percent)-alongside tremendous population growth rates of 50-80 percent more, much of it owing to international immigration or to in-migration from other parts of the United States.
The regional character of the changing global economy is perhaps in most evident in these southwestern central cities. Rapid growth in Brownsville and McAllen, Texas, for example, has been spurred by proximity to Matamoros, Mexico. Although high poverty rates reflect very real strains, the area's diversifying job base includes electronics, apparel, metal fabrication, food processing, bus manufacturing, medical facilities, tourism, higher education, and agriculture.
Laredo's fast growth owes in large measure to its strategic location as the U.S. gateway on Interstate 35, the major link between the three signatories to the North American Free Trade Agreement (NAFTA). Unemployment is down, with a boom in trucking and the emergence of a regional medical center. But continued job growth and affordable housing are major challenges.
Even among the doubly burdened cities, then, there is variety in terms of the human and other assets available to stage a comeback, as well as the nature of the local challenge. South Texas and California cities are struggling to ensure opportunity for many newcomers who arrive poor. At the same time, the doubly burdened cities in Illinois, Ohio, Michigan, Missouri, New Jersey, New York, Pennsylvania, and other States have continued to lose population; too many of those who stay remain poor and unemployed despite the Nation's economic recovery. We have not invested enough in those urban cores, despite the available infrastructure, land, and labor.