Introduction

When our preliminary report found that the Twin Cities is among the most segregated communities in the country, people were shocked. When the report asserted that if nothing changes, the Twin Cities will soon look very much like Milwaukee, Detroit, or Cleveland, again there was disbelief. But now there is evidence that the Twin Cities is further down the path to urban decline than our earlier findings suggested.

While national poverty trends have generally stabilized, poverty rates in the Twin Cities have dramatically increased. The national poverty rate for African Americans is at its lowest in history, but the poverty rate for African Americans in the Twin Cities metropolitan area is the second highest in the nation. Almost half of all people of color living in our central cities are in poverty -- the highest rate among the 25 largest metropolitan regions. The Twin Cities has one of the largest income disparities between whites and people of color in the nation.

Further, while the national growth in concentrated poverty is quite high -- more than doubling over the last two decades -- concentrated poverty in the Twin Cities has at a rate more than twice the national average. Twin Cities schools are reflecting this racial and economic segregation and this is leading to disturbing academic results. This spring it was announced that 91% of all African American 8th graders in Minneapolis failed to pass basic skills tests in reading and math.

Twin Citians have reached a time to consider what future they will create for themselves. This community has already begun to creep down the path so familiar to many larger urban areas. If the Twin Cities are to reverse their course, they must begin now. This community must seriously examine and confront what is happening here, and it must come together to implement real, proven solutions. There is hope. Other communities have dramatically improved their condition; the Twin Cities can do the same.

Organization of the Final Report

To understand the causes and consequences of racialized concentrated poverty in the Twin Cities, this report turns first to an examination of national trends. Part one of this study will survey scholarly literature and report current trends in other urban centers in the United States. Specifically, in part one, the report will overview the various indices by which metropolitan centers can be described as segregated in the United States. Part one concludes that urban centers in this country are polarized by race and income, and this polarization has had negative effects on the life chances of those harmed by this polarization.

After describing the current demographic and spatial patterns of race and class in major metropolitan regions, the report then turns to describing how segregation was created and why it continues to exist. Significant evidence demonstrates that institutional forces of government and the private sector created and perpetuate racial segregation and concentrated poverty. The third section of part one details the consequences of racial segregation and concentrated poverty on a national level.

Part one ultimately generates a series of conclusions about racial segregation and concentrated poverty in the United States. These conclusions then form the bases for hypotheses tested to see if the trends that are found elsewhere in the country are present in the Twin Cities.

Part two of the report focuses on the Twin Cities metropolitan region. Following the pattern of analysis of part one, part two will first document the levels of racial segregation and concentrated poverty in the neighborhoods and schools of the Twin Cities. Part two will also discuss many of the causes of these residential and educational patterns. The final section of part two will detail some of the consequences of this segregation and poverty for the Twin Cities.

Part three contains the results of our racial attitudes survey. The survey includes questions about racial mix preferences in the schools and neighborhoods. The survey also asks what factors motivate how we choose new neighborhoods, or what might prevent us from moving at all.

Finally, part four of the report turns to solutions for the Twin Cities. Are any of the solutions or strategies found in other metropolitan areas applicable to the Twin Cities? The answer is yes, and this part of the report offers some suggestions based on the successes and failures of policy innovations tried elsewhere.

Note: When referring to racial segregation, this report will mean the segregation of racial minorities recognizing that schools or neighborhoods that are all-white are also racially segregated.
 

I. National Trends

During the 1970s and 1980s a word disappeared from the American vocabulary. It was not in the speeches of politicians decrying the multiple ills besetting American cities. It was not spoken by government officials responsible for administering the nation's social programs. It was not mentioned by journalists reporting on the rising tide of homelessness, drugs, and violence in urban America. It was not discussed by foundation executives and think-tank experts proposing new programs for unemployed parents and unwed mothers. It was not articulated by civil rights leaders speaking out against the persistence of racial inequality; and it was nowhere to be found in the thousands of pages written by social scientists on the urban underclass. The word was segregation.

Douglas Massey and Nancy Denton

American Apartheid:

Segregation and the Making of the Underclass

Section A: A Fractured Nation
 
 

This country remains divided by race and income. This stark reality is demonstrated most powerfully in our urban centers. Metropolitan regions, particularly those in the northern rustbest, remain heavily segregated, and in some instances, they are becoming more segregated.

Segregation by race and income has been a part of our national fabric for most of this century. There are few signs that it will substantially decrease any time soon. In addition to this persistent segregation, a new phenomena has emerged, a social condition, unique to the past few decades, present in the center cores of our metropolitan regions. This new condition is concentrated poverty. A direct descendant and inevitable extension of racial and economic segregation, concentrated poverty raises its own dangerous conditions. And while overall rates of poverty and racial segregation among peoples of color may be slowing down, concentrated poverty is exploding.

The dramatic expansion of concentrated poverty will have profound effects not only on the lives of isolated, low-income communities of color, but on the health of entire metropolitan regions. How our society responds to this growing reality, and how our nation addresses persistent segregation today, will have long lasting effects. Although we do not know everything about segregation and concentrated poverty, we do know that conditions are likely to expand if left unchecked.
 
 

Housing

Housing is more than shelter. It is a critical determinant of opportunity in our society. The spatial location and form of housing are important indicators of individual and collective access to social, economic, and political resources. As Professor John Calmore states, housing "provides social status, access to jobs, education and other services, a framework for the conduct of household work, and a way of structuring economic, social, and political relationships." (Calmore 1993).

Given the important functions of housing, persistent residential segregation in the United States means continued limitations to social, economic, and political opportunities for many communities of color. As formal discrimination has diminished in our society, new mechanisms have emerged to maintain patterns of racial and economic inequality.
 
 

1. The Growth of Segregated America

At the beginning of this century, there was a mass movement of people of color to metropolitan America. In search of factory jobs and new economic opportunities, African Americans in particular left the rural south for the urban hubs of the north. By the mid-1900s, metropolitan regions throughout the country were experiencing significant population growths. Throughout this expansion, the center cities housed most metropolitan residents. For example, in 1950, seventy percent of America's metropolitan population resided in central cities. (Rusk 1993).

In conjunction with an increasing percentage of people of color in center cities came an unprecedented growth in segregation. Before 1900, little segregation by race existed. By the mid-1900s, however, "areas of acceptable black residence became more and more narrowly subscribed." (Massey & Denton 1993).

Following World War II, a second mass movement began: the migration of whites with economic means and businesses to the suburbs. This movement dramatically altered the composition of metropolitan America. By 1990, almost 70% of metropolitan residents lived in the suburbs.

During this second mass movement, segregation increasingly became defined by political boundaries. The Kerner Commission, established in the wake of the race riots of 1968, found that between 1950 and 1966, 98% of African American population growth occurred within center cities. Conversely, between 1960 and 1966, 78% of white population growth occurred in suburbs. (Kerner Commission 1968). Where once metropolitan America had been racialized along neighborhood lines, now it is now racialized along the municipal borders of city and suburb. As of 1990, more than two-thirds of African American metropolitan residents lived in central cities as opposed to a mere one-third of white metropolitan residents. (Boger 1993).
 
 

2. Measuring Racial Segregation

a. Dissimilarity Index

The standard measure of segregation historically has been the dissimilarity index created by Karl Taeuber. The dissimilarity index assesses the "evenness" of a minority population's distribution throughout a community. For example, the black-white dissimilarity index gives the proportion of all blacks who would have to move to achieve an even (integrated) residential configuration with whites, that is, one where each census tract replicates the racial composition of the metropolitan area as a whole. Dissimilarity is measured on a scale of 0.00 to 1.00, with a score of 1.00 indicting total segregation of blacks and a score of 0.00 indicating total integration. Hence, a higher dissimilarity index means greater segregation within a community. (Rusk 1993).

Using the dissimilarity index, urbanologist David Rusk measured segregation in major metropolitan areas in the United States. (Rusk 1993). Among the most segregated cities in the country according to Rusk are Detroit, Cleveland, and Milwaukee.

Most Segregated U.S. Cities
Using Dissimilarity Index

 
Although most past studies of residential segregation have relied primarily on dissimilarity indices, these indices have significant limitations. First, dissimilarity indices are poor measures of the extent of actual contact between blacks and whites. Similarly, this index does not distinguish between instances where black neighborhoods are tightly clustered to create one large contiguous enclave and those where they are scattered in a checkerboard fashion. Also, the dissimilarity index does not indicate whether black households are concentrated within a very small area or live in lower density communities. Lastly, the dissimilarity index provides no information on whether black residents are highly centralized, whether they reside in a sector that extends from the central area to the periphery, or whether they are concentrated in neighborhoods located at the periphery of the metropolitan areas. (Smith 1991).

As a result of these limitations, others have expanded the list of indicators used to compute residential racial segregation. The most comprehensive measure of segregation appeared in the pathbreaking work of sociologists Douglas Massey and Nancy Denton.
 
 

b. The Massey & Denton Test

According to Massey and Denton, we can understand metropolitan segregation by looking at five separate factors: (1) racial unevenness; (2) racial isolation; (3) racial clustering; (4) racial centralization; and (5) racial concentration. "Unevenness" refers to whether blacks are over- or under-represented in some areas in comparison to where blacks would locate in a metropolitan area if they were randomly distributed. "Isolation" refers to the degree to which blacks occupy neighborhoods that are not similarly occupied by whites. The fewer whites in a black neighborhood the "more isolated" the blacks. "Clustering" refers to the degree blacks are spatially located in one or more contiguous neighborhoods of a metropolitan area as opposed to living throughout a region. "Centralization" refers to the degree to which blacks live in and around an urban core. Lastly, "concentration" refers to the degree that blacks are confined to one small area in a metropolitan area and how contiguous pockets of segregation are.

Using these five factors, Massey & Denton have created a statistical model that measures segregation. For each of the first three factors, the segregation of blacks is scaled from 0 to 100, while for the last two factors, the scale is from

negative 100 to 100. A score of 100 would indicate severe segregation. Massey & Denton define an area as "hypersegregated" if it has a score of 60 or more for four of the five factors.

Given this measure of hypersegregation, Massey & Denton find that sixteen cities qualify as hyper segregated, based upon 1980 census data: Atlanta, Baltimore, Buffalo, Chicago, Cleveland, Dallas, Detroit, Gary, Indianapolis, Kansas City, Los Angeles, Milwaukee, New York, Newark, Philadelphia, and St. Louis. Massey and Denton conclude that "13% of all African Americans in the United States live under conditions of intense racial poverty." (Massey & Denton 1993).

Moreover, Massey & Denton examine segregation in major metropolitan areas between 1930 and 1980 to determine what patterns exist. They conclude that from 1930 until 1970 segregation remained very high in America, with blacks actually becoming more isolated in many cities during that time. More significantly, from 1970-1980, racial segregation and isolation of blacks remained high, with the old "rustbelt" regions of the country being some of the most segregated areas in the United States. (Massey & Denton 1993). Overall, through 1980, America's metropolitan areas remained segregated, and in some areas, segregation increased.

Based on Massey & Denton's five factor measure of segregation, Harrison and Weinberg used the 1990 U.S. Census data to compute segregation for all the metropolitan statistical areas (MSAs) in the United States. They found that segregation persisted overall. Milwaukee was the most segregated city in the United States, with many of the other sixteen cities Massey & Denton labeled as hypersegregated in 1980 remaining hypersegregated in 1990. Using the Massey & Denton index, the ten most segregated metropolitan regions in the United States according to the 1990 Census were: Detroit; Milwaukee; Cleveland; Newark; St. Louis; Chicago; Indianapolis; Philadelphia; New York; and Minneapolis-St. Paul.

10 Most Segregated Metropolitan Regions
in the United States

 
Conclusion #1: Major cities and metropolitan areas in the United States remain highly segregated by race.
 
 

3. The Explosion of Concentrated Poverty

The rise of concentrated poverty is substantially increasing the isolation of low-income communities of color. Concentration of poverty causes a lack of economic resoures so severe that neighborhoods themselves are in poverty. Even though individual poverty for whites and people of color has diminished over the past few decades, we have seen an explosion in community poverty.

As indicated above, community poverty occurs when the neighborhood itself is poor. These are neighborhoods where 40% or more of residents are at or below the poverty line. Such concentrated poverty means more than just a high percentage of individuals in poverty. In neighborhoods with concentrated poverty, all residents face substantial limitations to life chances as the neighborhood cannot sustain itself economically.

Neighborhoods with concentrated poverty are marked by increasing numbers of poor people coupled with an out-migration of middle income residents and commercial resources causing a decreasing density. This results in an increase of vacant lots, dilapidated buildings, and abandoned housing which all act as magnets for crime and depressed housing values. Neighborhoods, and eventually cities and older suburbs, that become characterized by concentrated poverty suffer the disability of having growing needs and shrinking resources.

This new social condition has arisen out of the growing economic and racial segregation of metropolitan areas along city/suburb lines. In 1960, the per capita income of central city residents exceeded that of suburbanites. By 1990, however, central city per capita income was only 84% of the average for suburbs. In many large metropolitan areas such as New York, Saint Louis, and Minneapolis, the disparities were even greater with incomes of city dwellers at about two-thirds the level of suburban residents.

Not surprisingly, the proportion of people living in poverty is much higher in central cities. In 1990, 18% of all central city residents had incomes that were lower than federally prescribed poverty levels, compared with 8% of suburban residents. By 1993, this disparity widened even further as central cities experienced an average poverty rate of 22% compared to 10% in the suburbs. Again, aggregate numbers obscure much greater disparities in individual metropolitan areas. For example, the poverty rates for Philadelphia and Chicago exceeded 20% in 1990 whereas their suburbs had poverty rates that were less than 5%. (Frey 1993).

To understand the specific dimensions of poverty, both Jargowsky and Wilson argue that neighborhood-level analysis is a more appropriate and exact measure. (Jargowsky 1997; Wilson 1996). Relying upon U.S. Census data, Jargowsky focuses his study upon census tract analysis of poverty in all MSAs in the United States.

a. The Language of Poverty

High-poverty, or concentrated poverty, neighborhoods are defined as those neighborhoods with forty percent (40%) or more of their population at or below the poverty level. Poverty is determined according to household size. In 1990, for example, the poverty level for a family of three was an annual income of $10,419.

As a final note, for definitional purposes, a "ghetto" neighborhood refers to those census tracts that are predominately black in which at least forty percent of the households live below the poverty level. Similarly a "barrio" is a predominately Hispanic neighborhood in which at least forty percent of the households live below the poverty level.
 
 

b. The Demographics of Concentrated Poverty

Individual poverty rates have remained fairly stable over the past two decades. A massive racial gap still exists between whites and people of color, as seen in the table below, but poverty rates within racial/ethnic groups have remained relatively stable.
 

 
By contrast, concentrated poverty has expanded greatly over this same time period. The number of census tracts that contain at least 40% of persons at or below the poverty rate has more than doubled over the last two decades.
 

 
The number of people who live in conditions of concentrated poverty has also dramatically increased in the past few decades. There is an enormous racial gap in who lives in these concentrated poverty neighborhoods. More than half of all people living in concentrated poverty are African American.

 
As the above statistics indicate, there is a strong connection between race and poverty. Communities of color are disproportionately communities of concentrated poverty. Although by no means a demographic majority in the United States, African Americans represent more than half of all persons living in concentrated poverty.
 

Conclusion #2: While individual poverty rates have stabilized, concentrated poverty is growing rapidly, particularly
within urban centers.

Conclusion #3: People of color are substantially more likely than whites to live in neighborhoods of concentrated poverty.
 

Schools

For most of this century, schoolchildren in the United States have been attending schools that are racially and economically segregated. Despite the Brown v. Board of Education ruling of 1954, it was not until 1969 that civil rights groups, federal troops, and other governmental entities effectively dismantled formal segregation. This dismantling was limited to the southern states for the most part, leaving the de facto segregation of northern schools largely in tact. Recently, the little progress made in school desegregation has been diminished by resegregation trends.

1. The Segregating and Resegregating of American Schools

The last twenty-seven years have seen a substantial reversal of the trend toward school integration for African American students and the continuation of a trend toward greater segregation for Latinos. As the table below shows, between 1972 and 1995 the degree of segregation for African American students initially declined, then increased or remained stable. (Orfield 1997).

 
Approximately one third of African American students who have attended school since 1972-73 have done so in a school that was severely segregated. The percentage of African American students attending schools which are predominantly minority has held steady at about 66% and has increased slightly since 1980-81.

Latino students have become more segregated every year since 1968-69. In the 1980's, Latino students became more segregated than African American students, and in 1994-95, nearly three fourths of all Latino students in the country attended schools which were more than 50% minority. In the 1990's, more than one third of Latino students attend severely segregated schools.

Another way to measure school segregation is to look at the percentage of white students in schools attended by African American and Latino students. The table below shows that for African American students, this percentage has declined slightly since 1980, and currently stands at about 33%. (Orfield 1997). White enrollment at schools attended by Latino students, meanwhile, has declined steadily since 1970.
 

 
Segregation is most severe in this country's urban schools. The largest school districts in the country have schools made up almost exclusively of minority students. In the country's ten largest urban school districts, white students make up more than 22% of enrollment in only one district. (Council of Great City Schools 1993). In 1994-95, the Chicago and Detroit school districts had student bodies which were only 11.3% and 6.3% white respectively. (Orfield 1997). Many of the schools in these cities are nearly 100% poor and minority. (Council of Great City Schools 1993). Orfield observed, "Segregation is greatest in the large northern cities where desegregation was never accomplished." (Orfield 1993). While the formal segregation that was first challenged and struck down by the federal courts came from southern states, the current reality is that segregation persists at very high levels and is especially severe in the schools of Northeastern and Midwestern cities.
 
 
Conclusion #4: Schools in the United States remain intensely segregated by race, and some are becoming more segregated. Segregation is particularly severe in urban schools.

2. The Growing Concentration of Poverty in U.S. Schools

Schools that are racially segregated also tend to enroll large numbers of poor students. Most of the schools which are segregated by race also face conditions of concentrated poverty. (Orfield 1997). Nearly 90% of students are poor in schools with at least 90% minority enrollment. Professor Orfield's most recent analysis of school segregation in the U.S. finds, "The relationship between segregation by race and segregation by poverty is exceptionally strong. The correlation between the percent of African American and Latino enrollments and the percent of students receiving free lunches is extremely high . This means that when we talk about racially segregated schools, they are very likely to be segregated by poverty as well." (Orfield 1997).

This relationship between racial segregation and concentrated poverty can be seen when looking at individual urban school districts. In 1992-93, only 36% of public school students nationwide were minority while less than 40% received free and reduced lunches. (Council of Great City Schools 1993). In St. Louis, however, more than 80% of the students were racial minorities and 85% were eligible for free and reduced lunch. In Milwaukee, nearly three fourths of the students were minority and 64% received free or reduced lunches. (Council of Great City Schools 1993). Orfield also observed, "Desegregation is not only sitting next to someone of another race. A child moving from a segregated African American or Latino school will very likely exchange conditions of concentrated poverty for a middle class school." (Orfield 1997).
 
 

Conclusion #5: Schools which have predominantly students of color are also likely to have high levels of poverty.
 
 
Section B: The Causes of Segregation

Although many believe that the racial segregation of neighborhoods and schools has been a natural or self-selecting process, in reality segregation and concentrated poverty were constructed by a variety of powerful actors. Far from being an accidental occurrence, the current racial and economic patterns of metropolitan America are the product of actions taken by the federal government, state government and its local delegates, the housing and banking industry, and private individuals.

1. The Role of the Government

The federal government became active in the housing market at the same time that large numbers of blacks were moving from the south to the north in search of employment opportunities. The government acted in conjunction with the lending industry to frustrate the efforts of these migrants by inducing and facilitating the abandonment of central cities by middle income whites and businesses. This dual migration of blacks to the cities and whites and businesses to the suburbs is largely responsible for current metropolitan housing patterns. There is also much to suggest that these governmental actions helped to shape current preferences for segregated housing. Segregational forces continue to define metropolitan areas through government policies that support sprawl, exclusionary zoning, and institutional and private discrimination in the housing market.

a. Redlining

Government subsidies for home mortgages have played a large role in creating and maintaining racially segregated neighborhoods and perpetuating the racially divided distribution of resources. In 1933, the federal government established the Home Owners' Loan Corporation (HOLC). HOLC was established to provide refinanced or new mortgages for persons who were either in danger of foreclosure or had already lost their home to foreclosure. (Schill & Wachter 1995).

Through HOLC, the federal government sought to promote uniformity in lending among financial institutions. Part of that uniformity included developing standards for lending institutions to use when appraising neighborhoods for investment. As part of its policy, HOLC systematically undervalued racially or ethnically diverse central city neighborhoods, deeming them these areas too risky for investment. Federal authorities consequently induced the channeling of mortgage funds away from central city communities of color and almost exclusively toward white, outlying neighborhoods. HOLC's introduction of redlining left a legacy that influenced not only the private market, but also the Federal Housing Administration and Veteran's Administration who, during the 1940's and 50's steered mortgages towards white suburbs rather than communities of color in central cities. In fact, the FHA's Underwriting Manual explicitly required that loans were guaranteed only in racially homogenous neighborhoods. Thus, the federal government paid whites to leave the central cities and barred people of color from doing the same.

b. Exclusionary Zoning

Exclusionary zoning occurs when municipalities increase required standards of housing quality beyond those necessary for health and safety in order to exclude the realistic opportunity for low- and moderate income housing development and maintenance. This occurs predominantly in the suburbs and substantially limits housing choices for moderate and low income households. In doing so, exclusionary zoning, in conjunction with other forces, limits low and moderate income

Examples of exclusionary zoning regulations which exclude the urban poor (predominantly communities of color) include: (1) restrictions requiring excessive minimum square footage for new homes and lots; (2) prohibitions on multi-family housing units; (3) unnecessarily low maximum densities; (4) two-car garage requirements; (5) requiring unnecessary and costly fixtures and adornments; and (6) fees associated with the cost of development and service connection charges. These types of regulations make it nearly impossible for poor families to find suitable and affordable housing. Areas with such regulations are most often middle- and upper-class white communities.

The use of zoning and other regulatory land use practices as a means of housing discrimination in the United States began with explicitly racial zoning. Explicitly racial zoning was abolished by the U.S. Supreme Court in the 1917 Buchanan v. Warley decision. Subsequent to this prohibition, the end of racial zoning (excluding minority residents from white neighborhoods) was pursued through the use of race-restrictive covenants: restrictions on deeds and/or agreements among property owners in a given area preventing property sales to minority households. These covenants were widespread until 1948 when the U.S. Supreme Court ruled in Shelly v. Kraemer that it was unconstitutional for any government to enforce a racially restrictive covenant.

Exclusionary zoning and regulatory land use practices were not explicitly covered in the Fair Housing Act of 1968. The role of exclusionary zoning in housing discrimination has changed from explicit and intentional racial exclusion to de facto exclusion of low-income, particularly minority, residents through a circumscription of affordable housing opportunities. While there are no conclusive studies regarding the scope of exclusionary zoning in the United States, several prominent cases highlight its continued practice. The most famous recent examples are the Mount Laurel decisions.

In the 1975 Southern Burlington County N.A.A.C.P. v. Township of Mount Laurel decision, the New Jersey Supreme Court declared that a municipality's zoning ordinance had an effect beyond its boundaries and implicated something of fundamental import: regional housing availability. The court held that, in order for a municipality to meet the state constitutional requirement that zoning powers be exercised for the general welfare of the state, the welfare of people living outside --as well as within--the municipality must be considered. The court concluded that developing municipalities must fulfill this general welfare obligation by affirmatively, through land-use policies and regulations, creating a realistic opportunity for affordable housing to a degree that satisfies their fair share of the regional need for low- and moderate-income housing. (Haar 1996).

The Mount Laurel decisions banned exclusionary zoning in order to facilitate the creation of affordable housing units. This set the precedent for what has become known as "fair share policies." These policies, enacted throughout the United States, attempt to remove exclusionary zoning barriers by encouraging numerous practices such as relaxation of acreage requirements; special tax breaks to developers who build affordable housing in suburbs; and so on. (Luckerman & Kane 1994; Yinger 1995).

Mount Laurel is the largest and most sustained effort to address exclusionary zoning and has led to the most comprehensive study of the process of implementing fair share goals in the private housing market. Results indicate that New Jersey's municipalities have been slow in meeting their "fair share" goals, as specified by the court, partly due to a housing development recession and largely due to municipal resistance leading to a proliferation of legal disputes among city planners, builders, and housing advocates over requirements passed by the state legislature. The housing that has been built through Mount Laurel has done little to change existing segregation by race and income as most units constructed have gone to white suburban residents experiencing short-term poverty cycles.

c. Federal Housing Programs

With few exceptions, the federal government has built public housing projects in low income, minority neighborhoods. Similarly, other government assistance housing programs, such as Section 8 housing, have concentrated low income racial minorities in urban areas. For example, even though promoting regional housing opportunity was one of the original purposes of the Section 8 program, during the first thirteen years of the program, eligible tenants received housing certificates from the Section 8 agency in their city of residence and were unable to use these certificates elsewhere. This practice confined low-income participants from areas of concentrated poverty to these same areas when seeking housing. Not surprisingly, a study of Section 8 programs concluded that, by themselves, Section 8 certificate and voucher programs were generally ineffective in changing the segregation and isolation of HUD-assisted families. (Abt Associates 1994). Even if these certificates were more freely transferable, the effects would be much the same as few suburban municipalities operate public housing agencies.

Although racial segregation in public housing has declined in the last twenty years, most African American public housing residents still live in concentrated poverty and predominantly African American neighborhoods while white public housing residents are more likely to live in economically stable, white neighborhoods. (Tegeler, Hanley, & Liben 1995).

A 1994 study commissioned by HUD which gathered demographic data on the location and racial composition of public housing nationwide found that fifty-five percent of all public housing family units are located in concentrated poverty census tracts. (Goering, Kamely, & Richardson 1994). These communities account for forty-seven percent of African Americans and sixty-eight percent of Hispanics (as well as eighty-three percent of all-white households) living in public housing. In census tracts with lower levels of poverty (less than twenty percent), twenty-five to thirty-seven percent of all public housing residents were found to be African American.

d. Fragmented Metropolitan Regions

Another important determinant of levels of racial segregation and concentrated poverty is the degree to which metropolitan regions are politically fragmented. Rusk has concluded that there is a positive correlation between the number of local governments and the degree of racial segregation in a given metropolitan area. (Rusk 1993). This is largely due to the fact that more localized political structures enables suburban municipalities to enact insular policies which shield them from metropolitan responsibilities.

 
In such fragmented areas there is a greater likelihood of exclusionary zoning laws and the utilization of other practices that limit the construction of low income housing and preclude low income migration into these suburban communities. Conversely, the more unified a metropolitan region is in developing metropolitan-wide housing, zoning, and other political policies, the more integrated the region tends to be.

Conclusion #6: Government has significantly contributed to the creation and maintenance of segregation.

2. The Role of Private Actors

a. private housing discrimination

The most comprehensive data on the state of housing discrimination was gathered in HUD's two national fair housing studies: the 1989 Housing Discrimination Study (HDS) and the 1977 Housing Market Practices Survey (HMPS). The HMPS used audit teams to survey forty metro areas. The HDS conducted 3,800 audits in twenty-five metropolitan areas nationwide. What constitutes discrimination in these HUD studies is the systematic differential treatment of racial minorities and whites.
 

 
HDS estimates indicate that African American and Hispanic home seekers encounter discrimination approximately half of the time that they visit a rental or sales agent to inquire about advertised housing opportunities. Housing agents showed twenty-five percent fewer housing units to minority customers, steered them to minority neighborhoods, and gave them considerably less assistance in finding housing that met their needs. These findings of the incidences of discrimination in categories of availability and treatment are consistent with other fair housing audits conducted in the 1980's.

In addition to HUD fair housing audits, many local communities have performed their own studies of racial discrimination. (Galster 1990; Turner 1992). A review of seventy-one fair housing audits conducted in individual localities during the 1980's conservatively estimated that blacks seeking homes for sale face a twenty percent chance of discrimination. Price/financing, availability, qualifying requirements, and information given are the most probable areas of discrimination for blacks seeking homes. Blacks seeking rental dwellings face a fifty percent chance of encountering discrimination, with availability, application, and information given being the most prevalent categories of discrimination.

Hispanics seeking rental dwellings face a thirty-three percent chance of discrimination. In terms of severity, the average black home seeker was told about two fewer homes for sale than the average white. The average black apartment seeker was told about two fewer units and was shown one fewer than their white counterpart and the average Hispanic apartment seeker was told about one fewer unit and shown one-quarter fewer unit.

Though these individual audits do not provide a fully representative statistical sample, they do give an indication of the persistence of a significant degree of discrimination in housing markets all-over the United States and are consistent with the findings of the HMPS and the HDS.

b. Racial Steering

The HDS yielded new findings on the marketing practices which produce racial steering. Racial steering is the practice of directing individuals toward neighborhoods which match their perceived race. For instance, racial steering would cause racial minorities to be less likely to be shown houses in white neighborhoods than in minority neighborhoods.

In the HDS cases, the minority auditor was steered to census tracts with a greater minority concentration 6.2 percent more often than his or her counterpart in the case of blacks and 4.8 percent more often in the case of Hispanics. However, the auditors derived their housing samples from units advertised in major newspapers, and such units were predominantly located in white census tracts. The HDS found that houses for rent and sale in largely minority and integrated neighborhoods are not marketed and sold in the same way. They are not advertised in newspapers and not shown to customers who inquire about advertised units. Furthermore, real estate firms that sell houses in largely black and integrated areas tend to have fewer listings than firms that sell in largely white areas. (Yinger 1995; Turner 1992; Turner, Mickelsons, & Edwards 1991).

Three different exploratory studies, in Washington, Milwaukee, and Boston, of housing marketing support the hypotheses that housing in black and integrated neighborhoods is marketed differently than housing in predominantly white neighborhoods, and that black and integrated neighborhoods are less likely to be advertised in the newspaper and less likely to be marketed through open houses. (Newburger 1992).
 
c. Private Mortgage Redlining

As mentioned earlier, redlining is the practice by banks of refusing to invest in certain neighborhoods because of specific racial characteristics. The practice of redlining has come from both public and private actors (See 1a above). Although the federal government historically has facilitated private redlining, it has recently vowed to improve lending in low-income communities of color by promoting changes in private mortgage practices through the Community Reinvestment Act (CRA), and other similar legislation.

Enacted in 1977 to motivate certain financial institutions to meet community credit needs, the CRA avowed purpose is to require federal financial supervisory agencies to use their authority to encourage regulated institutions "to help meet the credit needs of the local communities . . . consistent with the safe and sound operation of such institutions." A regulated institution's supervisory agency is to "assess that institution's record of meeting the credit needs of its entire community, including low- and moderate income neighborhoods, consistent with the safe and sound operation of such institutions."

The CRA is just one of three federal banking statutes enacted to address individual discrimination and geographic redlining, as well as the perpetuation of an atmosphere of "community disinvestment" in the financial services industry. The CRA, the Equal Credit Opportunity Act (ECOA), and the Home Mortgage Disclosure Act (HMDA) were each directed toward creating a more level playing field in the credit market. Discrimination in the residential real estate finance market has been prohibited under the Fair Housing Act since 1968, but it was not until the enactment of the Equal Credit Opportunity Act in 1974 that Congress addressed the issue of discrimination in lending.

The ECOA originally prohibited creditors from considering gender-related factors in assessing whether to grant credit. Amendments prohibiting discrimination on the basis of race, color, religion, national origin, age, receipt of public assistance benefits, and the exercise of right under the Consumer Credit Protection Act were enacted in 1976. Data collected under the HMDA, which uses disclosure to address the problem of discrimination in lending, indicate that many minorities have a statistically higher chance of have their home mortgages denied than do similarly situated whites. The ECOA and the HMDA only tangentially impact the related practices of geographic redlining and community disinvestment.

Most redlining occurs when an institution fails to reinvest in its own community due to unfounded perceptions about the credit risk of borrowers in low-income areas (unrelated to the borrower's ability or willingness to pay or the value of the collateral). The CRA was intended to begin to remedy years of redlining. Opinions on the CRA break down into two camps: individuals who support the CRA as a crucial step towards solving problems of inadequate housing and urban decay; and individuals who denounce the CRA as aggressive and heavy handed government intervention mandating suboptimal lending patterns.

In 1991, the CRA was amended to explicitly reference race and gender. It provides that when evaluating an institution's performance in meeting community credit needs, consideration should be given to any loss or contribution incurred when an institution donates, sells on favorable terms, or makes available on a rent-free basis a branch located in a predominantly minority neighborhood to a minority depository institution or a women's depository institution. In 1992, the agencies were instructed to consider in the CRA evaluation process majority owned institutions' capital investments, loan participations, and other cooperative ventures with minority-owned and female-owned financial institutions and low-income credit unions.

In July 1993, in an effort to put more teeth into the CRA, President Clinton instructed financial institution regulators to develop new regulations to enforce the CRA. The proposed amendments attempted to respond to industry complaints that the existing enforcement regime provided ambiguous performance standards and excessive paperwork, as well as to activists' and politicians' concerns that the existing CRA regime "overemphasized process and underemphasized performance." The amendments set forth a measurable performance criteria to assess whether an institution was meeting community credit needs by establishing an evaluation system that sought to reward results in community lending. Under the new criteria, an institution must demonstrate that its efforts produced measurable results: loans actually made to communities. In April of 1995, final enforcement regulations were approved after an extended comment period. Under the new performance standards for large, retail institutions, three tests are conducted to generate raw ratings that are then used to compute an institution's composite CRA rating: the Lending Test, the Investment Test, and the Service Test.

The Lending Test reviews an institution's performance in meeting community credit needs and measures the institution's home mortgage lending, small business and small farm lending, community development lending, and, if applicable, consumer lending. The rating on the Lending Test hinges on the institution's responsiveness to the credit needs of its assessment area: the degree of lending in its assessment area; the geographic distribution of loans throughout its assessment area and among individuals of different income levels and businesses; its record in meeting the credit needs of highly economically disadvantaged areas of its assessment area, of low-income individuals, or of businesses; its use of innovative or flexible lending practices in a safe and sound manner to meet the needs of low- or moderate-income individuals and geographies; and its leadership position in community development lending.

The Investment Test assesses the amount of "qualified investments" made by the institution, the innovative nature and complexity of the institution's investments, and its responsiveness to community credit and community development needs. The Service Test evaluates the availability and effectiveness of a bank's systems for delivering retail banking services and the extent and innovative nature of its community development services. The evaluator assesses the institution's distribution of branches among the geographic region it serves, its record of opening and closing branches, the extent to which it makes available alternative systems for delivering banking services to low- and moderate-income geographies and individuals, and the range of the services offered.

The three tests are used to determine the institution's composite CRA rating with the greatest weight given to the institution's performance under the Lending Test. The ramifications of evidence of discrimination or illegal credit practices are left vague, however, and the regulations are not specific as to how an institution's ratings will factor into application processes to a given agency.

It was proposed that if an institution receives an overall rating of "substantial noncompliance," the Justice Department subject the institution to enforcement proceedings but this proposal did not make the final 1995 version of the CRA. Also dropped late in the revisions of the CRA was a proposal to create a "three strikes and you're out" rule under which an institution which rated as "needs to improve" for three consecutive examinations would be classified as "substantial noncompliance." Another criticism of the CRA is that the ratings, "good," "adequate," "reasonable," and "poor," are overly broad and lending institutions assert that the evaluation is still too subjective. Countering the lending institutions concern is the awareness of activists and CRA advocates that, though the CRA tries to focus on the three tests, it never requires an institution to abandon safety and soundness in the name of good results.

National studies on mortgage lending practices performed in the last few years demonstrate that the CRA has a long way to go before reaching its goal of ending redlining. A study published by the Federal Reserve Bank of Boston in October, 1992 titled Mortgage Lending in Boston: Interpreting HMDA Data revealed that people of color are sixty percent more likely to be rejected for loans than similarly situated white applicants.
 

 
A study by Andriana Abariotes entitled Disparities In Mortgage Lending In the Upper Midwest: Does Discrimination Exist?, considered home purchase, improvement and refinancing loans. The study indicated that blacks have the highest rejection rate of any racial group (21.7% as compared to 14.1% for Hispanics, and 8.8% for Asians), and nearly triple the rate of rejection for white applicants (7.5%). Moreover, the study found that 38% of the disparity is inexplicable in terms of credit history, characteristics of the census tract where the relevant property was located, or "other measurable characteristics of the loan or borrower." (Abariotes 1994).

In another national audit, blacks were found to experience discriminatory treatment in the mortgage lending process 39% of the time, while Hispanics experienced discrimination 37% of the time. Limitation of access to mortgage credit occurs when: the lender discourages potential borrowers from submitting formal applications, the person appraising the property systematically under-appraises certain types of property, the lender discriminates against certain types of applicants, or the lender imposes harsher terms on selective applicants. (Leigh 1993).

In April 1995, U.S. News and World Report published the report of a six month investigation of banking, lending, and home insurance coverage in poor and minority communities. The study found the number of poor and minority homeowners who could not obtain full coverage property insurance is nearly fifty percent greater than for mostly white, middle-income areas. The study reported that over the last two decades the number of bank branches in white neighborhoods has increased at a rate three times greater than the rate of increase for bank branches in mostly minority areas. The study also found that middle-income black applicants from mostly minority areas were more than twice as likely to be rejected for mortgage loans as middle-income whites living in mostly white areas.

Community First, a D.C. based group reported in 1995 that African American residents of D.C. lost more than $2 billion in potential mortgage loans between 1991 and 1993. The study reported that if D.C.'s African Americans had received mortgage loans at the same rate as white home owners, an additional 20,000 loans would have been made. The study also estimated a $1 billion shortfall in loans to low and moderate income communities. In addition, it estimates that discrimination costs blacks over four billion dollars annually, with those costs including higher housing prices, mortgage payments and costs, and search costs. (Yinger 1995).
 
 
Conclusion #7: The private sector has significantly contributed to the creation and maintenance of segregation in housing.

3. The Interrelationship between Housing and Education Markets

There are at least three different ways in which school segregation and residential segregation influence one another. First, people often base their residential choices within metropolitan areas on the quality and racial composition of an area's schools. Second, states often draw school district and municipal lines in ways that create and preserve segregated communities. Finally, housing and education are related in that residential segregation produces segregated schools.

In her path breaking work, Diana Pearce (1980) examined the links between educational desegregation and housing integration in metropolitan regions and found that schools serve as a cue for housing choices. According to Pearce, school segregation supports housing segregation as suburban schools signal a place where whites can escape racial minorities. Given that school segregation promotes housing segregation, Pearce also considered whether school desegregation reinforces housing desegregation. While some argue that school desegregation actually produces white flight, Pearce noted that white flight is only a possibility within metropolitan areas that provide white enclaves.

Pearce theorized that the impact of schools on housing choices varies depending on whether schools are segregated or desegregated at the metropolitan level. She further hypothesized that the removal of white enclaves in metropolitan school systems will result in lowered levels of housing segregation.

Pearce's study looked at the impact of metropolitan school desegregation plans on housing patterns. Seven paired cities were selected. One city of each pair had a metropolitan school desegregation plan for at least five years; a minimum ten percent minority in schools; and a population of at least 100,000 in 1970. Its counterpart was a city demographically similar, but without a metropolitan desegregation plan. The different pairs were chosen to maximize variance in ethnic mix, region, length and type of desegregation. Data was gathered from interviews, school districts, newspapers, the census, visits to real estate agents, and academic reports, among other sources.

Pearce found that, in cities with segregated schools, information about schools is often used as a coded means to convey information about the racial content of communities. Because it is illegal in many cases to discuss directly the racial content of neighborhoods, real estate agents or newspaper ads often discuss schools in order to reveal information about neighborhoods.

In looking at fifty real estate ads in each area, Pearce found that in areas with segregated schools, advertisements mentioned school names two to ten times more often than in areas with desegregated schools. Furthermore, of all the ads surveyed that mentioned a school, not one ad mentioned a predominately black school and half the schools mentioned were 98% or more white. Thus, Pearce concluded that in segregated communities, the school names serve as code words for the racial composition of a neighborhood, and that whites considered predominantly white schools a desirable characteristic.

Next, Pearce studied advice given directly to home seekers by real estate agents. She found that agents often used school information as a way to discuss neighborhoods, sometimes explicitly referring to the racial composition of the school. Other times, agents used "good" and "bad" to describe the racial composition of a school as a means of conveying information about a neighborhood's racial composition.

Finally, Pearce examined differences between paired cities and found that those cities with metropolitan-wide school desegregation plans had experienced decreased residential segregation as measured by the index of dissimilarity.

Pearce also concluded:

• The effectiveness of a metropolitan school desegregation plan is directly proportional to how metropolitan-wide the school desegregation actually is (i.e., the more metropolitan the plan, the more effectively schools will be desegregated).
• Metropolitan housing desegregation is more effective in areas with growing housing markets, while metropolitan areas that are not experiencing growth will have more difficulty desegregating housing.
• School desegregation plans have benefits on the individual, neighborhood, and metropolitan level:
   
• On the individual level, desegregated schools provide positive racial encounters and reduction of fears between majority and minority parents.
• At the neighborhood level, the removal of white educational enclaves makes it difficult for whites to flee integration. Further, if we exempt racially integrated neighborhoods from busing, incentives are created to encourage people to move into these neighborhoods.
• On the metropolitan level, school desegregation often results in increased outside resources for desegregated schools and brings us closer to the principle of equal educational opportunity for all.
   
• Overall, the more metropolitan the school plan, the stronger the impact on housing integration: school desegregation has long-term, cumulative effects on housing choices, whereas school segregation, supported by housing segregation, is self-perpetuating and will not disappear on its own.

In a 1984 report, Pearce examined twenty-five metropolitan areas with central city black populations over 100,000 and employed the index of dissimilarity to examine the spatial segregation of blacks. Most of the cities studied did not have metropolitan-wide school desegregation plans. In such cities, any school desegregation occurred solely in the central city, and thus the impact of the desegregation was confined to this area and did not extend to the suburbs. The study also posited a lag of four years between school desegregation and any resultant changes in residential patterns.

The study demonstrated a clear correlation between school desegregation and housing desegregation (R²=.765), finding a "curvilinear effect, reflecting a decreasing return in housing desegregation as greater amounts of school desegregation occur." Pearce tested for spurious effects, finding region of the country was not an important variable. Only population growth seemed to be significant: while southern cities were more likely to have integrated neighborhoods as a result of school desegregation, this was often a consequence of their quicker growth. City size in 1970 and percent of blacks in schools in 1968 also appeared as significant.

Conclusion #8: There is a significant connection between housing segregation and school segregation. Furthermore, the more metropolitan-wide school districting is, the more likely it is to result in desegregated schools and housing.
 
 

Section C: The Consequences of Segregation

1. Schools

a. educational disadvantage

Racial segregation and concentrated poverty create powerful and unique educational impediments for schools. When low-income children are segregated into schools that are predominantly poor, the students confront not only their own individual poverty, but the effects of concentrated poverty on the school system itself. Poor children are less likely to attend pre-school and are increasingly arriving at school unprepared for learning. (Kozol 1991). Students in poverty are also more likely to receive inadequate health care. As a result, these students experience many health problems such as low birth-weight, learning disabilities, malnourishment and other developmental problems. (National Center for Education Statistics 1996; Council of Great City School 1993). Moreover, the parents of these students are usually much less educated than the parents of students in middle class schools. (Orfield 1997).

Children in poverty are also disproportionately affected by family instability. Children from single-parent families are concentrated in poor neighborhoods and poor schools. (Coleman 1966; National Center for Education Statistics 1996). Poor schools also concentrate teen pregnancy to the point where it is common for girls to become pregnant at least once before finishing high school. (National Center for Education Statistics 1996). Poor families in urban districts also tend to move frequently. In many urban districts, it is common for poor African American and Latino students to change residence three or more times in a year and attend three or more different schools. (Kozol 1991; Orfield 1993).

Schools with high numbers of poor and minority students also see more problems associated with drug and alcohol abuse, crime, and gang activity. Whether a student's family or the student herself is the user, drug and alcohol abuse can be a serious problem in inner city schools. (Council of Great City Schools 1993). Crime is much more pervasive in the neighborhoods in which poor children live. Accordingly, the schools attended by these students tend to be marred by crime and violence as well. (Olson 1993). Urban schools have been confronting the educational problems associated with gang activity long before such problems extended beyond the poorest neighborhoods. (Kozol 1991).

Teachers and staff in racially segregated, high poverty schools are too often overwhelmed by student needs. Orfield observes, "High poverty schools have to devote far more time and resources to family and health crises, security, children who come to school not speaking standard English, seriously disturbed children, children with no educational materials in their homes, and many children with very weak educational preparation." (Orfield 1997). The famed Coleman Report made a similar observation over thirty years ago. "The inequalities imposed on children by their home, neighborhood, and peer environment are carried along to become the inequalities with which they confront student life." (Coleman 1966).

Conclusion #9: Students of racially segregated, high poverty schools face enormous educational disadvantages.

b. achievement

It is difficult to engage in a discussion of public education in America without hearing how our schools are failing or that education is in a state of crisis. The modern education crisis is most closely linked to the release of A Nation at Risk in 1983 in which a federal commission concluded "The educational foundations of our society are presently being eroded by a rising tide of mediocrity that threatens our future as a Nation and a people . . . . If an unfriendly foreign power had attempted to impose on America the mediocre educational performance which exists today, we might well have viewed it as an act of war." The federal commission found that in comparisons with students from other countries on nineteen tests, U.S. students never finished first or second and were last on seven tests. Furthermore, SAT scores for U.S. students had shown an unbroken decline between 1963 and the release of the report. Average achievement for high school students was lower in 1983 than in 1957, and there was a steady decline in science achievement dating to 1969.

Whether the problems with public schools are overstated is debatable, but there are plenty of reasons to believe that the schools are failing. Widely distributed reports on student performance usually provoke grave concern. The most compelling of these are the results from the nation's report card: the National Assessment of Educational Progress (NAEP). The 1996 NAEP results showed that 70% of 4th-graders were "not proficient" in reading and 82% were "not proficient" in math. (Bracey 1996). In addition, reports of low achievement relative to students in other countries continue to pour in. The International Mathematics and Science study showed that in 1993, U.S. students ranked fourteenth out of fifteen countries tested in math and thirteenth out of fifteen in science. (Olson 1993).

The overall low performance of students in the nation as a whole does not reveal the full extent of the education crisis. Low student performance is not uniformly distributed throughout the country, but is disproportionately found in segregated and high poverty schools in urban areas. While A Nation at Risk warned of "a rising tide of mediocrity," educational attainment in high poverty schools is far worse than mediocre. Low student performance is endemic to communities where large numbers of poor and minority students attend school. (Council of Great City Schools 1993). The results from the nation's report card in 1992 shockingly showed that only 5% of fourth grade students in disadvantaged urban communities were proficient in reading and only 3% of these students met this standard in math. Nine percent of the students in disadvantaged communities in eighth grade tested proficient on the reading test, and only 6% on the math test. In twelfth grade, 20% of disadvantaged students were proficient on the reading test and 6% on the math test. (Council of Great City Schools 1993).

Students in urban districts consistently have lower levels of achievement than students overall. The Council of Great City Schools reports that only 40% of the students in urban districts scored above the national average on reading tests. These students are also more than twice as likely as students in rural and suburban districts to drop out of school. (Council of Great City Schools 1993).

These disparate results are compounded by variations in student achievement within the urban districts. Within these districts there is a sizable and consistent gap between White and middle class students and poor and minority students (Council of Great City Schools 1993). In grades 9-12, 30.7% of African Americans in urban districts finished above the national average on standardized tests in math. Conversely, sixty-five percent of white students in urban districts exceeded the national average. In grades K-6, 31.3% of African American students in urban districts scored above the national norm in reading, whereas the rate for white students in these schools was 60.7%. Only 28.6% of African American students in urban high schools scored in the top half on reading achievement tests. Two thirds of the white students in urban high schools, meanwhile, finished above the national average. Thirty-eight percent of African American students in kindergarten through sixth grade finished above the national average on math achievement tests. Over two-thirds of the white students in these grades surpassed the national average. (Council of Great City Schools 1993).
 

Conclusion #10: While achievement levels for U.S. students remain unacceptably low, achievement is especially low for low-income students of color in urban schools.

An abundance of national research on student performance verifies that attending schools facing racial isolation and concentrated poverty causes performance to suffer. In 1966, the Coleman Report reviewed comprehensive data on more than 600,000 students in 3000 schools. The report concluded that the only factor with a significant effect on a student's academic achievement is the socioeconomic level of that student's schoolmates. (Coleman 1966).

The report's findings have been tested by many researchers since 1966, but the impact of a school's socioeconomic mix on student achievement has not been diminished. Added to this has been the effect of a school's racial composition on student achievement. (Jencks 1972). Current research on student achievement produces the same conclusion. Caldas found that the strongest predictors of a school's academic achievement was the percentage of black students in the school and the percentage of students receiving free and reduced lunch. These factors were much more highly correlated with student achievement than any other factor including attendance rate, class size, and school size. (Caldas 1993). The results led Caldas to conclude that earlier estimates regarding the impact of school poverty and student background characteristics may have been too conservative.

A wealth of research in education has found specifically that school desegregation improves student performance. These studies have consistently shown that minority students educated in a desegregated school achieve at a higher level than those in segregated schools. Furthermore, students from desegregated schools also experience long-term benefits even after their schooling is finished.

A 1972 study showed that African American students educated in a desegregated environment had higher verbal achievement than those from segregated schools. (Crain 1972). This study also found that students finished more years of school and were more likely to graduate from high school if they had a desegregated education. This study was careful to control for differences in student and school characteristics so that fair comparisons could be made. The only difference between the two groups of students was that one attended schools suffering from racial isolation and the concentration of poverty and the other attended integrated schools.

In 1983, researchers looked at 93 studies on desegregation. This meta-analysis synthesized the findings that emerged from these studies and found that desegregated schools have a positive effect on student achievement. Furthermore, these positive effects are more pronounced when students attend desegregated schools at an early age. (Crain 1983). In 1994, conclusions reached in earlier studies received further support from a pair of meta-analyses. (Crain 1994; Dawkins & Braddock 1994). Together, these analyses looked at 34 of the most substantial studies on the long-term effects of segregation. These studies controlled for student and school characteristics and looked at outcomes beyond school performance. Consistent among these studies was the finding that black students educated in desegregated schools are more likely to work in a desegregated environment, attend a desegregated college, live in a desegregated neighborhood, and possess higher career aspirations. (Crain 1994; Dawkins & Braddock 1994).

Perhaps the strongest evidence of the value of desegregation on educational attainment comes from Norfolk, Virginia. In 1986, the Norfolk school board decided to end its attempts to integrate the schools through mandatory busing and instead opted for a community schools plan. (Meldrum 1994). The gravamen of the plan was the opportunity for parents to send their children to an elementary school close to home. This was further justified by the assertion that community schools would reverse the tide of white flight. The school board predicted that white students would return at a rate of 7% per year. (Meldrum 1994). Members of the school board in Norfolk also stated that they expected the plan to close the learning gap that existed between white and African American students. Unfortunately, community schools did not produce these results in Norfolk.

The percentage of white students in the district was the same in 1994 as it was in 1986. (Meldrum 1994). What the community schools plan did change, however, was the degree of intra-district racial segregation and concentrated poverty. In 1994, there were ten elementary schools that had become at least 90% African American. According to the plan, these ten racially isolated schools received extra resources including more money per student, more experienced teachers, smaller classes, more library books, a pre-school program, and a multicultural program. (Meldrum 1994). The percentage of poor students at the racially isolated schools was also much higher. In five of the schools, 100% of the students were eligible for free and reduced lunch. Not surprisingly, the level of achievement at these racially isolated schools was dismal. On three different achievement tests, a large gap persisted between black students who attended one of the racially isolated schools and black students who attended more integrated schools. Furthermore, the overall learning gap between white and black students grew after the adoption of the community schools plan. (Meldrum 1994; Ikpa 1993).

Conclusion #11: Attending a racially isolated, high poverty school adversely affects student achievement levels.

2. Economy and Employment

Segregation prevents wealth accumulation by residents of isolated, poor communities of color, thereby establishing barriers to market participation. Lack of educational opportunities, poor job accessibility, and the lack of appreciation of housing values in isolated, low-income communities contribute to this problem. Segregation geographically polarizes metropolitan areas along lines of race, income and opportunity, artificially separating urban centers and older suburbs from surrounding communities. (Jargowsky 1997; Orfield 1997; Wilson 1996; Wilson 1987; Rusk 1993).

Recent studies document the lack of an economic base in poor communities of color. (Oliver & Shapiro 1995). The lack of capital and access to capital is a serious issue for any community in a market society. For example, studies of fiscal characteristics of black and white suburbs have found that black suburbs have lower property tax bases, higher taxes, and greater indebtedness. (Schneider & Logan 1982).

Segregation also has adverse effects beyond the borders of central cities and inner-ring suburbs because cities are important to their regional economies. (Downs 1994; Keating 1994). Suburban residents and governments benefit from the unique economic opportunities and cultural amenities that large urban centers provide. (Jacobs 1984; Rusk 1993; Voith 1992). Furthermore, metropolitan areas with an employable workforce possessing basic job skills have an advantage in attracting businesses to the area. Metropolitan regions as a whole benefit from breaking down barriers such as segregation which prevent poor communities of color from full participation in the metropolitan economy as workers, consumers, and citizens. A community has a healthy economic base when people both inside and outside the community are willing to invest in the area: through purchasing homes; sponsoring or owning businesses; and making loans.

Segregated economic development in metropolitan regions has resulted in what is known as "spatial mismatch." The greatest need for jobs is in the center city where unemployed minorities live. However, job production has left the urban cores and many of the better paying jobs are now in the the suburbs, often in places difficult to reach by many in the center city because of a lack of adequate transportation. (Wilson 1987; Wilson 1996; Rusk 1993; Ihlanfeldt 1994). Some also suggest that in addition to the inaccessibility of jobs, there is a problem in terms of the types of employment available to those in center cities given their often limited skills. Poor people of color often lack strong educational backgrounds and employment records relegating them to low wage jobs with minimal advancement opportunities. (Jargowsky 1997; Wilson 1996).

Overall, for those living in concentrated poverty neighborhoods, studies indicate that these individuals have lower income, less job opportunities, and less stable family situations than those who are able to escape these neighborhoods (Wilson 1996).

Conclusion #12: Neighborhoods of concentrated poverty adversely impact the economic conditions of their residents.

3. Neighborhood Characteristics

Concentrated poverty harms neighborhoods in other ways beyond its damage to the regional economy, educational achievement, and employment opportunities. Evidence drawn from the Twin Cities suggests that crime in concentrated poverty neighborhoods is higher than in nonconcentrated neighborhoods. (Orfield 1997).

The age, quality, and availability of adequate housing in concentrated poverty neighborhoods also compares unfavorably to nonconcentrated neighborhoods. (Jargowsky 1997; Wilson 1987; Wilson 1996). Specifically, individuals in these neighborhoods are more likely to pay a greater percentage of their income for housing that is substandard, overcrowded, and/or which is experiencing falling property values in comparison to suburban housing. (Oliver & Shapiro 1995).

Family instability, in terms of high divorce rates, single parent households, family violence and abuse, and teenage pregnancies is also greater in concentrated poverty neighborhoods. (Wilson 1996; Danziger & Weinberg 1994; Mincy 1994; Garfinkel & McLanahan 1994).

Overall, individuals who live in concentrated poverty areas or who attend schools with high concentrations of poverty experience decreased educational achievement and employment prospects and are exposed to neighborhoods with high crime, substandard housing, and isolation from many employment, public, and cultural services that the rest of society takes for granted.

Conclusion #13: Neighborhoods of concentrated poverty often experience a number of adverse social conditions.

Conclusion

The major metropolitan centers in the United States remain highly segregated by race and income. With the exception of a few metropolitan areas, the degree of segregation has not changed significantly in the forty years since Brown v. Board of Education. In many cases, there is evidence of resegregation and growing concentrations of low income racial minorities into selected neighborhoods in urban centers and some inner ring suburbs.

The causes of segregation and concentrated poverty are many, including private housing market discrimination, government placement of subsidized housing in urban centers, exclusionary zoning, metropolitan fragmentation, and the interaction of residential and school segregation.

Concentrated poverty neighborhoods have grown in the last few decades, with the vast majority located in metropolitan areas. Blacks make up the majority of people in these neighborhoods. Individuals living in concentrated poverty neighborhoods experience higher crime, poorer schools, decreased housing quality, and isolation from economic, public, and cultural resources that are enjoyed by those not living in concentrated poverty. The effects of concentrated poverty on racial minorities are severe and pervasive.

Overall, the research has generated thirteen conclusions about racial segregation and concentrated poverty in metropolitan areas across the nation:
 
 Conclusion #1: Major cities and metropolitan areas in the United States remain highly segregated by race.

Conclusion #2: While individual poverty rates have stabilized, concentrated poverty is growing rapidly particularly within urban centers.

Conclusion #3: People of color are substantially more likely to live in neighborhoods of concentrated poverty.

Conclusion #4: Schools in the United States remain intensely segregated by race, and some are becoming more segregated. Segregation is especially severe in urban schools.

Conclusion #5: Schools which enroll predominantly students of color are also likely to have high levels of poverty.

Conclusion #6: The Government has significantly contributed to the creation and maintenance of segregation.

Conclusion #7: The private sector has significantly contributed to the creation and maintenance of segregation in housing.

Conclusion #8: There is a substantial connection between housing and education segregation.

Conclusion #9: Students of racially segregated, high poverty schools face enormous educational disadvantages.

Conclusion #10: While achievement levels for U.S. students remain unacceptably low, achievement is especially low for low- income students of color in urban schools.

Conclusion #11: Attending a racially isolated and high poverty school adversely impacts achievement levels of low-income students of color.

Conclusion #12: Neighborhoods of concentrated poverty adversely affect the economic conditions of their residents.

Conclusion #13: Neighborhoods of concentrated poverty experience a disproportionately high number of adverse social conditions.

Given these general conclusions, one must ask whether the Twin Cities area reflects these national experiences. In part two, this report will test each of these conclusions in the form of hypotheses regarding the Twin Cities metropolitan area.

Next Section...

Back to the Table of Contents...