August 2003 • The Brookings Institution • Survey Series 1Cen

“The state of the

region's schools

shows that the

way the area is

growing hurts

residents of

almost every city

and suburb.”

Center on Urban and Metropolitan Policy

■ The elementary school student popu-lation in the Miami metropolitan areais growing rapidly, but the growth isvery unbalanced. Regionwide, enroll-ment increased by 22 percent between1993 and 2001. Miami-Dade County’selementary enrollment grew by 15 per-cent, while Broward County’senrollment grew by 35 percent. Butsome outlying communities in theregion saw much faster growth—insome cases as high as 85 percent.

■ The region’s two school districtsbecame poorer over this period, andthe degree of income segregationworsened. The number of low-incomestudents in the Miami region grew 33percent between 1993 and 2001. By2001, 51 percent of the region’s totalelementary students were eligible forfree lunches, up from 47 percent in1993. Poor students were also morelikely to attend school with other poorstudents at the end of the period. Theshare of students who would have hadto change schools to achieve an identi-cal mix of poor and non-poor studentsin each building edged up two percent-age points, to 51 percent.

■ As the region’s schools became morediverse, racial segregation easedslightly but remained severe. Miami-area students became a more diverse

group between 1993 and 2001. His-panic enrollment grew by 57 percentand black enrollment grew by 17 per-cent, while white enrollment decreasedby 10 percent. Growth patterns con-tributed to lingering segregation.Approximately two-thirds of the growthin Hispanic enrollment was in Miami-Dade County schools, while nearly allof the growth in black enrollment tookplace in Broward County. The numberof white students held steady inBroward and declined 29 percent inMiami-Dade.

■ The region’s most dramatic socialchanges are taking place in the sub-urbs. While still at alarming levels,poverty and segregation rates in thecentral city are stabilizing. The mostdramatic social changes are taking placein inner suburban communities, whichoften must address growing need withdwindling fiscal resources.

The concentration of poor and minoritystudents in a particular school can fuelthe flight of middle-class families from thesurrounding neighborhood. These changescontribute to a vicious cycle of sprawl anddisinvestment from existing communities.To help reverse some of these patterns,state and local leaders should explorereforms in land use, taxes, and regionalgovernance.

FindingsAn analysis of race and poverty trends in Miami-area elementary schools between 1993and 2001 reveals that:

Economic and Racial Segregation in Greater Miami’s Elementary Schools:Trends Shaping Metropolitan GrowthMyron Orfield, Anne Discher, and Tom Luce1

I. Introduction

As both a southern city and thenation’s gateway to LatinAmerica, Miami has long hada majority of children of color

in its schools. Bolstered by its historyand geography, along with goodweather and a healthy economy, theMiami metropolitan area grew rapidlyin the 1990s. By 2000, almost 3.9 mil-lion lived in the region’s two counties,Miami-Dade and Broward. Theregion’s school enrollment grew evenfaster, becoming poorer and moreracially diverse in the process. Rapid,unbalanced growth—coupled with theend of decades-old desegregationplans in the region’s school districts—is contributing to the segregation ofthe greater Miami schools by incomeand race. No part of the region isimmune from its harmful effects. Newchildren—mostly poor and Hispanic orblack, many of them immigrants—have disproportionately enrolled instruggling, sometimes deterioratingschools in the region’s older communi-ties. At the same time, middle-classfamilies of all races have retreated tonew neighborhoods on the region’sedge. These people find themselvescontending with the overcrowdedschools, strained budgets, and trafficcongestion that often accompany rapidgrowth.

This report is intended to highlightthe social changes underway inMiami-area schools and discuss theirimplications for metropolitan growthpolicies. As throughout the UnitedStates, patterns of school segregationare supported by incentives built into awide variety of public policies. Trans-portation and infrastructureinvestment patterns subsidize sprawl-ing development on the suburbanfringe. The fragmented political natureof the metropolitan area makesthoughtful, efficient land-use plan-ning—an important mechanism forassuring that all residents have accessto jobs and affordable housing—nearly

impossible. Tax policies encouragelocal governments to engage in waste-ful competition for the most affluentcitizens. The way the region respondsto this increasing polarization repre-sents a powerful portent for its future.

Analysis in this study centers onschools because they act as a kind of“canary in the coal mine”—an institu-tion that tells a lot about both thecurrent health and future well-being of the community surrounding it. This is true, first, because communitystability depends greatly on the per-formance of schools. Deepeningpoverty and other socioeconomicchanges appear in schools before theydo in neighborhoods and in elemen-tary schools before secondary schools.When the perceived quality of a schooldeclines, it can set in motion a viciouscycle of middle-class flight and disin-vestment.2 Schools often experiencethis social change faster than neigh-borhoods do because families with nochildren in the public school system(empty nesters, the young, and fami-lies with children in private schools)will often remain in a neighborhoodpast the time when most families withschool-aged children in public schoolsflee. This can ease the increase inoverall poverty rates. But ultimately, inmost cases, when schools in a commu-nity reach certain thresholds ofpoverty and segregation, middle-classhouseholds of all types (i.e., house-holds with residential choices) willchoose to live in other areas.

II. Methodology/Definitions

This report examines changesin the racial and economiccomposition of elementaryschools in the Miami region

between 1993 and 2002. The region—also referred to as “greater Miami”—isdefined in this report as the U.S. Cen-sus Bureau’s Miami ConsolidatedMetropolitan Statistical Area. TheMiami CMSA includes Miami-Dadeand Broward counties (Map 1).

Because Florida has established asingle school district for each of itscounties, greater Miami, unlike mostcomparably sized metropolitan areas,is home to only two school districts:Miami-Dade County and BrowardCounty. They are the nation’s fourthand fifth largest districts, behind onlythose of New York, Los Angeles andChicago (Table 1).

For analysis, this report relies ondata from the Common Core of Data,an annual, national database compiledby the National Center for EducationalStatistics, a division of the U.S. Depart-ment of Education.3 This survey usesdata for two years, 1993 and 2001. The database provides information onindividual schools, including totalenrollment, the number of students eligible for free lunch and a break-down of enrollment by racial or ethnicgroup. Because together they comprisethe vast majority of students in theMiami region, this report focuses onblack, Hispanic and white students.

August 2003 • The Brookings Institution • Survey Series2 CENTER ON URBAN AND METROPOLITAN POLICY

Table 1. Largest U.S. School Districts, 2001

Rank School System Total enrollment1 New York City Public Schools 1,066,9452 Los Angeles Unified 721,3463 City of Chicago School District 435,2614 Miami-Dade County School District 368,3565 Broward County School District 251,129

Source: National Center for Educational Statistics, Common Core of Data, Department of Educa-

tion, 2002.

The report uses eligibility for freelunch as a proxy for poverty. Freelunches are available to children offamilies whose household income is ator below 130 percent of the federalpoverty line. In 2001, the maximumannual household income for an eligi-ble family of three was $19,019.4 Thisstudy defines high-poverty schools asthose with free lunch eligibility rates

that are at least 25 percent above theaverage rate for the region. In theMiami area, high-poverty schools in2001 were those with at least 63 per-cent of its students eligible for freelunches.

Data at the elementary-school levelare used for several reasons. First, theyoffer finer-grain analyses of neighbor-hood trends because there are more

elementary schools than middle- orhigh-school buildings. In addition,there is some evidence that elemen-tary-level free lunch eligibility data aremore accurate than data from highergrades because eligible elementarystudents are more likely to enroll inthe free-lunch program than older eli-gible students.

To assess broader patterns withinschool districts, school-level data wereaggregated using geographic informa-tion system (GIS) software to assigneach school building to the municipal-ity or county unincorporated areawhere it is located.

Finally, this report makes use of dis-similarity indexes to measure the levelsof racial and income segregationamong area elementary students.These commonly used statistics meas-ure the degree to which two groupsare evenly distributed in a given geo-graphic area. In this case, they can beinterpreted as the percentage of one ofthe student groups that would have tochange schools to achieve a perfectlyintegrated enrollment—for example,an equal mix of black and non-blackstudents, or poor and non-poor stu-dents, in each building.5

III. Findings

A. The elementary school studentpopulation in the Miami metropoli-tan area is growing rapidly, but thegrowth is very unbalanced.Between 1993 and 2001, the numberof elementary school students in theMiami region grew by 22 percent, orover 56,000 students. By 2001, totalelementary enrollment was 310,578(Table 2). These trends reflect broaderpopulation changes in greater Miami.The region grew by 21 percent, or683,798 people, in the 1990s. Thiswas comparable to the growth rate inmetropolitan San Antonio (20 percent)but slower than that of greater Atlanta,where population grew 39 percent. Ofthe two counties comprising greaterMiami, Broward County, which makes

August 2003 • The Brookings Institution • Survey SeriesCENTER ON URBAN AND METROPOLITAN POLICY 3

Map 1MIAMI REGION

up the northern half of the region,added more residents and added themat a faster pace. It grew by approxi-mately 367,000 residents or 29percent. Miami-Dade County, theregion’s southern county, grew by 16percent or 316,000 people.

Following a similar pattern, elemen-tary student enrollment in Miami-DadeCounty grew more modestly than theregion as a whole, 15 percent, from1993 to 2001. By 2001, 185,635 ele-mentary students were attendingschools in the district. The district cov-ers a wide range of communities,including the city of Miami; strugglinginner suburbs like Opa-Locka and ElPortal; fast-growing cities like FloridaCity, Homestead and Hialeah Gardens;and stable affluent places like CoralGables and Key Biscayne. In 2001, theschool district was home to 60 percentof the region’s students, nearly three-fourths of the region’s poor students,and 81 percent of the region’s Hispanicstudents (Appendix Table 1).

The Broward County schools experi-enced more explosive studentpopulation growth than Miami-DadeCounty schools between 1993 and2001. The county’s enrollment grew35 percent between 1993 and 2001.The district includes older communi-ties experiencing growing social strain,like Fort Lauderdale and LauderdaleLake, as well as fast-developing mid-dle-class areas like Pembroke Pinesand Parkland. Broward Countyschools, home to 40 percent of theregion’s students, enrolled 44 percentof the region’s black students and 72percent of the region’s white students.In 2001, Broward County’s elementaryenrollment was 124,943.

Broward County includes somecommunities where elementaryschools grew much more rapidly thanthe district’s average. Enrollmentwithin the city of Pembroke Pines, forexample, more than doubled. CoconutCreek schools grew by 85 percent andenrollment at one elementary schoolin Weston doubled between 2000 and

2001.6 This rate of growth causes sig-nificant strains. School officials inthese areas often scramble to accom-modate the burgeoning studentpopulation by bringing in portableclassrooms, raising class sizes andshifting attendance boundaries—attimes leaving parents uncertain aboutwhat school their children will attendfrom year to year.

B. Elementary students in theregion’s two school districts gotpoorer over this period, and thedegree of income segregation wors-ened. The number of low-income students inthe greater Miami area increased 33percent from 1993 to 2001—anincrease of nearly 39,000 students.That’s a considerably faster rate of

August 2003 • The Brookings Institution • Survey Series4 CENTER ON URBAN AND METROPOLITAN POLICY

Map 2MIAMI REGION: Percentage of Elementary Students Eligiblefor Free Lunch by School, 2001

change than the 22 percent increasein total enrollment. Miami-DadeCounty experienced the greaterabsolute growth in poor students(24,630 students) while the larger per-centage gain (48 percent) occurred inBroward County (Map 3).

Throughout the region, many poorschools got poorer, and many schoolsin the western portions of both coun-ties saw an influx of low-incomestudents, although most of these con-tinued to have relatively low povertylevels. By 2001, 51 percent of theregion’s total elementary students wereeligible for free lunches, up from 47percent in 1993.

In 2001, poor schools were clus-tered in the cities of Miami and FortLauderdale and many adjacent com-munities, including Hialeah,Opa-Locka and North Miami Beachnear Miami; and Dania, Oakland Park,Lauderdale Lakes and Lauderhill justoutside Fort Lauderdale (Map 2). Notonly did the Miami-Dade school dis-trict have the most low-incomestudents, it also had the highest con-centration of extremely poor schools.There were 38 schools where 90 per-cent or more of the students wereeligible for free meals. Of those 38schools, 35 were located in Miami-Dade, and 17 were located in the cityof Miami. Just three were located inBroward County; one each in FortLauderdale, Lauderhill and unincorpo-rated Broward County.

The maps show the concentrationof low-income schools in the city ofMiami and other older communitiesalong the Atlantic Ocean. Dissimilarity

August 2003 • The Brookings Institution • Survey Series 5CENTER ON URBAN AND METROPOLITAN POLICY

Map 3MIAMI REGION: Change in Percentage Points of Elementary Students Eligible for Free Lunch by School, 1993–2001

Table 2. Elementary enrollment by district, 1993 and 2001

1993 2001 Absolute Change Percent ChangeBroward County 92,391 124,943 32,552 35Miami-Dade County 161,978 185,635 23,657 15Region 254,369 310,578 56,209 22

Source: National Center for Educational Statistics, Common Core of Data, Department of Education, 2002.

indexes confirm that poor students areincreasingly attending school withother poor students. In 2001, 51 per-cent of the region’s poor elementarystudents would have had to changeschools in order to achieve an identi-cal mix of poor and non-poor studentsin each one. The degree of incomesegregation among students is edgingup; in 1993 just 49 percent of poorstudents would have had to move.

These trends are troubling becausethe negative effects of concentratedpoverty—everything from high crimeand poor health—don’t stop at theschool door. They discourage invest-ment in poor neighborhoods, place asignificant burden on city resources,and dramatically limit the opportuni-ties of residents. Ultimately peopleliving there, isolated from educational,employment and social opportunities,find it difficult to participate fully inthe metropolitan economy.

C. As the region’s schools becamemore diverse, racial segregationeased slightly but remains severe.Schools in the Miami area enrolledgrowing numbers of students of colorbetween 1993 and 2001. Hispanicenrollment in the region grew by 57percent and black enrollment grew by17 percent in this period, while whiteenrollment declined by 10 percent.

The changes taking place in theMiami region are strongly shaped byits exceptional role as a destination forimmigrants. In fact, the area tops arecent list of “melting pot metros” andis attracting large numbers of newarrivals from Latin America.7 At the

August 2003 • The Brookings Institution • Survey Series6 CENTER ON URBAN AND METROPOLITAN POLICY

Map 4MIAMI REGION: Percentage of Black Elementary Students by School, 2001

Table 3. Share of enrollment and segregation of free-lunch eligible students, 1993 and 2001Dissimilarity index

Percent of all students (Percent required to move to achieve parity)Free-lunch eligible students 1993 2001 1993 2001Broward County 33 36 46 51Miami-Dade County 55 61 47 47Miami region 47 51 49 51

Source: Nation Center for Educational Statistics, Common Core of Data; dissimilarity calculations by the authors

same time, as immigrants come to theregion, especially to Miami-DadeCounty, many native-born residentsare leaving, some heading northtoward Broward County. They move insearch of bigger, newer houses andless congestion and crime, and, whenpushed, many will admit to escapingthe powerful and growing Latin influ-ence in Miami-Dade.8

Regionwide, Hispanic enrollmentgrew at over 2.5 times the rate of totalenrollment from 1993 to 2001—57percent versus 22 percent. These fig-ures reflect the phenomenal increasein Latinos coming to the Miamiregion, largely from Central and SouthAmerica and the Caribbean.9 Nearlytwo-thirds of growth in Hispanicenrollment from 1993 to 2001 (or

31,214 students) occurred in Miami-Dade County. In fact, Hispanicstudents accounted for virtually all thenew students in that district over thoseeight years. By 2001, the majority ofMiami-Dade County’s elementary stu-dent population—57 percent—wasHispanic, up from 46 percent in 1993.The share of Hispanic students inBroward County also grew signifi-cantly—doubling from 10 percent in1993 to 20 percent in 2001—althoughtheir numbers remained far belowthose in Miami-Dade.

While black enrollment in theMiami region grew by 17 percent dur-ing this time period, nearly all of thosenew students enrolled in BrowardCounty schools. In fact, the number ofblack students enrolled in BrowardCounty rose 48 percent from 1993and to 2001, and by 2001 black stu-dents represented 36 percent ofstudents in the district, up from 33percent in 1993. The number of blackstudents in the Miami-Dade Countydistrict remained virtually unchangedfrom 1993 to 2001, and the share ofblack students in the district fell onepercentage point, from 34 percent to33 percent.

Maps 4 through 7 show the locationand shift of Hispanic and black ele-mentary school students in the Miamiregion. In 2001, Hispanic studentswere concentrated primarily in thesouthern area, including the southside of Miami, Hialeah, West Miami,Sweetwater, and adjacent unincorpo-rated Miami-Dade County (Map 6). AsMiami’s Hispanic community hasexpanded, the region’s establishedblack community, traditionally cen-tered in the neighborhoods of northMiami, has been moving north intoinner suburbs of Miami-Dade Countyand, increasingly, beyond them intoBroward County (Map 4 and 5).10

The changes are particularly evidentwithin Miami proper. In 1993, 46 per-cent of students in the city’s publicschools were black and 51 percentwere Hispanic. By 2001, Hispanics

August 2003 • The Brookings Institution • Survey Series 7CENTER ON URBAN AND METROPOLITAN POLICY

Map 5MIAMI REGION: Change in Percentage Points of Black Elementary Students by School, 1993–2001

had solidified their majority: 36 per-cent of students were black and 61percent were Hispanic (Map 7). Thismirrored the larger demographic shiftsafoot in the city. The number of blackresidents in the city of Miami droppedby over 17,000 during the 1990s, andblacks represented just 23 percent ofthe city’s population in 2000, downfrom 27 percent in 1990.11

Although the region remains raciallydivided, overall levels of racial segrega-tion in schools have dipped slightly. In2001, for example, 55 percent of His-panics would have had to changeschools to achieve an identical mix ofHispanic and non-Hispanic students,down from 62 percent in 1993 (Table4). Despite this drop, many Latinoscontinue to attend racially isolatedschools. In 2001, nearly one in threeLatino students attended a school witha Hispanic enrollment of 90 percentor greater, and half attended schoolswhere 90 percent or more of studentswere either Latino or black.

Although black students, too, saw avery slight improvement in segregationcompared to 1993, they remained themost segregated of any racial group inthe Miami region in 2001. That year,64 percent of students would have hadto change schools to achieve an identi-cal mix of black and non-blackstudents, compared with 55 percent ofHispanic students and 59 percent ofwhite students. More than one in four(29 percent) black students attended aschool where more than 90 percent ofstudents were black, and two-thirdsattended a school with a total com-bined black and Latino enrollment of90 percent or higher.

As children of color have made uplarger and larger shares of the region’stotal enrollment, the number of whitestudents has decreased. Regionwide,there were 8,180 fewer white elemen-tary students in 2001 than in 1993.About 8,050 of those students leftMiami-Dade County. While the num-ber of white students in BrowardCounty held steady, the school district

grew, transitioning from a majority-white student body (56 percent) in1993 to one that was more diverse (41percent white) in 2001. In the regionas a whole, white students made up 23percent of total elementary schoolenrollment in 2001, down from 31percent in 1993. Despite these shifts,there continued to be areas with con-siderable white enrollment. White

students remained highly concentratedon the region’s relatively affluent edge,in western and southern BrowardCounty, in cities such as PembrokePines, Sunrise and Coral Springs; andto a lesser degree in and aroundPinecrest, South Miami and CoralGables in Miami-Dade County.

Overall, this study revealed higherlevels of racial segregation than eco-

August 2003 • The Brookings Institution • Survey Series8 CENTER ON URBAN AND METROPOLITAN POLICY

Map 6MIAMI REGION: Percentage of Hispanic Elementary Students by School, 2001

nomic isolation in the Miami region.For example, some of the schools withprimarily Hispanic enrollments insouthern Miami-Dade County (theyappear in red on Map 6) had relativelyfew students eligible for free lunchesin 2001 (in blue on Map 2). But thelink between race and incomeremains. In fact, 65 percent of blackelementary students attended high-

poverty schools in 2001, as did 38 per-cent of the region’s Hispanic students.In comparison, only 8 percent of whitestudents attended these schools.12

D. The region’s most dramatic socialchanges are taking place in subur-ban schools. Many people have traditionally dividedmetropolitan areas into two distinct

parts: the struggling city and itswealthy, stable suburbs. The idea thatsocial strife stops neatly at the city’sborders is increasingly out of date, associal strain sweeps into many of theregion’s older suburbs.13

Although the city of Miami doesindeed have alarming levels of povertyand racial segregation in its schools,city schools at least appeared to notget much worse from 1993 to 2001.The city’s average school poverty ratein 2001, 78 percent (compared withan average of 66 percent among cen-tral cities in the nation’s 25 largestmetropolitan areas in 1997), was actu-ally one percentage point lower thanin 1993. Likewise, the city’s overallelementary minority rate held nearlysteady, rising from 96 percent to 97percent.

As in many other U.S. metropolitanareas, the Miami region’s most dra-matic social changes are actuallytaking place in the suburbs. The aver-age poverty rate in schools in theBroward County city of Lauderhill, forexample, increased 20 percentagepoints—from 51 percent to 71 per-cent—between 1993 and 2001. Theaverage poverty rate in North MiamiBeach schools rose over 17 points—from 61 percent to 79 percent.

In 2001 schools in North MiamiBeach, North Miami, Opa-Locka,Homestead and Florida City actuallyaveraged higher poverty rates than the78 percent rate within Miami proper.The average poverty rates in three ofthem exceeded 90 percent. Racialtrends follow similar patterns. Therewere five suburban places—FloridaCity, Opa-Locka, Homestead, NorthMiami and North Miami Beach—withminority enrollments equal to orhigher than that of Miami. Schools inanother seven suburban cities hadaverage black and Hispanic enroll-ments of more than 90 percent.

Although places like these oftenretain vibrant, active neighborhoods,they also frequently strain to cover thecosts of social change with low and

August 2003 • The Brookings Institution • Survey Series 9CENTER ON URBAN AND METROPOLITAN POLICY

Map 7MIAMI REGION: Change in Percentage Points of Hispanic Elementary Students by School, 1993–2001

slow-growing property tax bases. Theircommercial districts cannot attractnew, big businesses that could easilybuild on greenfield sites, yet theseaging suburbs also lack the culturalamenities, gentrifying neighborhoodsand downtown tax base that help cen-tral cities survive despite theirproblems. As a result, these communi-ties often become poorer faster thaneven the cities they surround.

IV. The Future of Schools—and the Miami Region

This survey shows that currentgrowth patterns in the Miamiregion—rapid development in outlying communities,

coupled with concentrations of low-income households and people ofcolor in older communities—harm theentire Miami metropolitan area. In thecities of Miami and Fort Lauderdale,and, increasingly, in close-in suburbs

like North Miami, Opa-Locka andLauderhill, the effects of rapid studentgrowth elsewhere are damaging, leav-ing poor students—disproportionatelychildren of color—concentrated inschools of extreme poverty. Suchschools often suffer from risk fac-tors—everything from inexperiencedteachers to unstable enrollment—thatlower educational achievement amongstudents and diminish their prospectsfor the future.14

This pattern also has serious impli-cations in fast-growing communities atthe region’s edge—from Parkland andCoral Springs in north BrowardCounty to central Miami-Dade—where the middle class is streaminginto increasingly overcrowded, under-funded schools. To cover the costs ofnew schools, roads, parks and sewersneeded by new residents, local govern-ments compete against neighboringcommunities for the tax base, attempt-ing to lure high-end developments that

contribute more in tax revenue thanthey cost in public services. Theresulting large single-family homes,shopping centers and office parks aredevouring some of south Florida’smost productive agricultural land andcausing serious traffic congestion. Theside effects of the sprawling develop-ment—pollution and heavy wateruse—are harming the natural environ-ment, including Everglades NationalPark; one of the nation’s most uniqueand sensitive habitats.

In recent years, driven by growingalarm among parents, educators andthe broader public, policymakersacross the country have focused onimproving the academic performanceof students. Particular attention hasbeen given toward the inner-city andsome suburban schools where dismalacademic achievement has doomedstudents’ future prospects.

Solutions coming out of schoolboardrooms, state legislatures,

August 2003 • The Brookings Institution • Survey Series10 CENTER ON URBAN AND METROPOLITAN POLICY

Table 4. Share of enrollment and segregation of Hispanic, black and white students, 1993 and 2001

Dissimilarity index Percent of all students (Percent required to move to achieve parity)*

Black students 1993 2001 1993 2001Broward County 33 36 57 58Miami-Dade County 35 31 68 69Miami region 34 33 65 64

Hispanic students 1993 2001 1993 2001Broward County 10 20 27 29Miami-Dade County 46 57 53 56Miami region 33 42 62 55

White students 1993 2001 1993 2001Broward County 56 41 49 46Miami-Dade County 17 11 52 50Miami region 31 23 60 59

* Although segregation increased in each of the two counties when measured separately, the overall regional index declined because the way the two sepa-

rate county indexes combine to create the regional index changed during the period. The number of students increased more quickly in Broward County

than in Miami-Dade, increasing the relative impact of Broward County’s lower segregation rate on the overall regional index.

Source: Nation Center for Educational Statistics, Common Core of Data. Dissimilarity calculations by the authors.

Congress and the White House havegenerally focused on the classroom:improving the quality of instruction;increasing the time students spend onreading, writing and math; and man-dating standardized testing to makeschools more accountable. Butdecades of evidence suggests that sim-ply remediating struggling studentswithout changing the underlying pat-terns of regional growth that trapmany of them in underperformingschools of concentrated poverty willlikely to yield limited academicprogress.

The evidence suggests that the con-centration of poor students is aproblem that requires coordinated,regional strategies among all levels ofgovernment, and a focus not just onschool curriculum and testing, but onbroader, regional policies that changethe distribution of opportunity withinthe region. There are at least threeareas of regional reform that can easethe growing social polarization ingreater Miami:

1. Regional land-use planning helpscommunities coordinate invest-ments in roads, highways, sewersand utilities and use land more effi-ciently. It can be used to ensure thatall communities, particularly thosewith new jobs and good schools,strengthen their commitment toaffordable housing. That helpsreduce the consequences of concentrated poverty on core com-munities and provides people withreal choices about where they live.

2. Regional tax reforms can narrowthe fiscal gap between rich and poorplaces and decrease the incentivesfor local governments to engage inwasteful competition for tax base.They also offer struggling communi-ties the resources they need forrevitalization efforts and reduce theincentives for the middle-class fami-lies living there to pull up stakesand move.

3. Accountable metropolitan gover-nance gives all communities a voicein regional decision-making.Although the region is home to oneof the oldest consolidated govern-ments in the country, Miami-DadeCounty, there is still great room forimproved cooperation among localgovernments. Miami-Dade’s current“two-tiered” organization leavesmany important functions, includ-ing land-use planning, exclusively inthe hands of the county’s 30 indi-vidual municipalities. BrowardCounty has no consolidated govern-ment. And there is no coordinationof land-use or transportation plan-ning between Miami-Dade andBroward counties, despite the highdegree of economic and social inter-connection between them.

No community within a metropoli-tan area is an island. For better orworse, the well being of different partsof metropolitan areas are linked.When social and economic disparitieswithin the region are minimized, allparts of the region benefit.15 Indeed,the state of the region’s schools showsthat the way the metropolitan area isgrowing hurts residents of almostevery city and suburb, leading to con-centrated poverty and abandonedpublic facilities in central cities andat-risk suburbs, and overcrowdedschools and strained budgets on theurban fringe.

August 2003 • The Brookings Institution • Survey Series 1 1CENTER ON URBAN AND METROPOLITAN POLICY

August 2003 • The Brookings Institution • Survey Series12 CENTER ON URBAN AND METROPOLITAN POLICY

App

endi

x Ta

ble

1.E

lem

enta

ry E

nrol

lmen

t B

y F

ree-

Lun

ch E

ligi

bili

ty a

nd R

ace,

Mia

mi-

Dad

e an

d B

row

ard

Cou

ntie

s, 1

993

and

2001

Shar

e of

Shar

e of

Shar

e of

%

of

% of

En

rollm

ent

Enro

llmen

tEn

rollm

ent

Enro

llmen

t Re

gion

al

Regi

onal

Ch

ange

Di

stri

ct

Dist

rict

19

9320

01Ch

ange

% Ch

ange

Tota

l 199

3To

tal 2

001

1993

-200

1To

tal 1

993

Tota

l 200

1%

Chan

geTo

tal

Enr

ollm

ent

Bro

war

d C

ount

y92

,391

124,

943

32,5

5235

3640

58M

iam

i-D

ade

Cou

nty

161,

978

185,

635

23,6

5715

6460

42To

tal

254,

369

310,

578

56,2

0922

100

100

100

Fre

e-L

unch

Eli

gibl

eB

row

ard

Cou

nty

30,1

3244

,447

14,3

1548

2528

3733

363

Mia

mi-

Dad

e C

ount

y88

,355

112,

985

24,6

3028

7572

6355

616

Tota

l11

8,48

715

7,43

238

,945

3310

010

010

047

514

His

pani

cB

row

ard

Cou

nty

8,99

225

,263

16,2

7118

111

1934

1020

10M

iam

i-D

ade

Cou

nty

75,0

3710

6,25

131

,214

4289

8166

4657

11To

tal

84,0

2913

1,51

447

,485

5710

010

010

033

429

Bla

ckB

row

ard

Cou

nty

30,0

8344

,508

14,4

2548

3444

9933

363

Mia

mi-

Dad

e C

ount

y57

,143

57,2

8614

30

6656

135

31-4

Tota

l87

,226

101,

794

14,5

6817

100

100

100

3433

-2

Whi

teB

row

ard

Cou

nty

51,7

4551

,616

-129

065

722

5641

-15

Mia

mi-

Dad

e C

ount

y27

,838

19,7

87-8

,051

-29

3528

9817

11-7

Tota

l79

,583

71,4

03-8

,180

-10

100

100

100

3123

-8

Sou

rce:

Nat

iona

l Cen

ter

For

Edu

cati

onal

Sta

tist

ics,

Com

mon

Cor

e O

f Dat

a

Endnotes

1. Myron Orfield is the president of Ameregis,a research and GIS firm in Minneapolis.Anne Discher is a research associate withthe firm; Tom Luce is its research director.

2. See James S. Coleman and others, “Equal-ity of Educational Opportunity,” (U.S.Department of Education, Office of Edu-cational Research and Improvement,National Center for Educational Statistics,Washington, D.C.,1966); Gary Burtless,ed., Does Money Matter? The Effect ofSchool Resources on Student Achievementand Adult Success (Washington: BrookingsInstitution, 1996); and James Traub,“What No School Can Do.” New YorkTimes Magazine, January 16, 2000, p. 14.

3. The Common Core of Data can be foundat http://nces.ed.gov/ccd/.

4. “The 2001 HHS Poverty Guidelines,” available at aspe.hhs.gov/poverty/01poverty.htm (July 2003).

5. For more information on school and resi-dential segregation in U.S. metropolitanareas, see John R. Logan, “Choosing Segre-gation: Racial Imbalance in AmericanPublic Schools, 1990-2000” (Albany: Uni-versity at Albany, Lewis Mumford Centerfor Comparative Urban and RegionalResearch, 2002).

6. Daniel de Vise, “Faulty Growth ProjectionsTest, Tax Schools In West.” Miami Herald,December 17 2001, p. A1.

7. “Melting pot metros” are those with lowerthan average concentrations of whites andgreater than average concentrations of atleast two minority groups. See William H.Frey and Ross C. DeVol, “America’sDemography in the New Century: AgingBaby Boomers and New Immigrants asMajor Players” (Santa Monica, CA: MilkenInstitute, 2000). For additional informa-tion on growing suburban diversity, seeWilliam H. Frey, “Melting Pot Suburbs: ACensus 2000 Study of Suburban Diversity,”(Washington: Brookings Institution, 2001).

8. William Booth, “A White Migration NorthFrom Miami,” The Washington Post,November 11, 1998, p. A1.

9. U.S. Census Bureau, Supplementary Sur-vey, 2000, table P039.

10. Andres Viglucci, “Miami neighborhoodswin, lose in 2000 Census counts,” TheMiami Herald, April 8, 2001; and AndresViglucci, “Dade blacks exiting inner city,”The Miami Herald, April 22, 2001.

11. U.S. Census Bureau, 2000.

12. Data from National Center for EducationalStatistics; calculations by the authors.

13. Myron Orfield, American Metropolitics:The New Suburban Reality (Washington:Brookings Institution, 2002.

14. Gary Orfield and John T. Yun, Resegrega-tion in American Schools (Cambridge, MA:Harvard University, 1999).

15. A growing body of research shows theinterdependence of different parts of met-ropolitan areas. Larry C. Ledebur andWilliam R. Barnes, for example, found thatmedian household incomes of central citiesand their suburbs move up and downtogether in most regions and that thestrength of this relationship appears to beincreasing. They also found that metropoli-tan areas with the smallest gap betweencity and suburban incomes had greaterregional job growth. See Ledebur andBarnes, “All In It Together: Cities, Suburbsand Local Economic Regions,” (Washing-ton D.C.: National League of Cities,1993); and Barnes and Ledebur, “City Dis-tress, Metropolitan Disparities, andEconomic Growth,” (Washington D.C.:National League of Cities, 1992). RichardVoith found that in large metropolitanareas, income growth in central citiesresults in income growth and house-valueappreciation in the suburbs. See Voith, “DoSuburbs Need Cities?” Journal of RegionalScience 38 (8) (1998): 445-464.

August 2003 • The Brookings Institution • Survey Series 13CENTER ON URBAN AND METROPOLITAN POLICY

For More Information:Ameregis(612) 379-3926

Myron Orfieldmorfield@ameregis.com

Anne Discheradischer@ameregis.com

Tom Lucetluce@ameregis.com

For General InformationBrookings Institution Center on Urban and Metropolitan Policy(202) 797-6139www.brookings.edu/urban

Acknowledgments:The Brookings Center on Urban and Metropolitan Policy would like tothank the John D. and Catherine T. MacArthur Foundation for their support of our work on metropolitan growth dynamics in the South Florida region.

The Brookings Institution

1775 Massachusetts Avenue, NW • Washington D.C. 20036-2188Tel: 202-797-6000 • Fax: 202-797-6004

www.brookings.edu

Direct: 202-797-6139 • Fax/direct: 202-797-2965

Center on Urban and Metropolitan Policy