Home
Forgot your password?
The Impact of Katrina: Race and Class in Storm-Damaged Neighborhoods
A 2006 Brown University report analyzing New Orleans neighborhood damage through the lens of class and race
Full screen
John R. Logan
Professor of Sociology, Brown University
Director, Spatial Structures in the Social Sciences
(401) 863-2267
John_Logan@brown.edu
The Impact of Katrina: Race and Class in Storm-Damaged Neighborhoods
Early media reports about the wind damage and flooding caused by Hurricane Katrina focused
on New Orleans, and especially on the people who had been unable to escape the city before it
flooded. Images of poor and predominantly black people crowded into the Superdome and
Convention Center supported the impression that Katrina had disproportionately affected poor,
black neighborhoods. The purpose of this report is to evaluate more precisely what
neighborhoods were heavily damaged, including not only New Orleans but also the coastal
communities in Mississippi that bore the brunt of hurricane-force winds.
In brief an analysis of FEMA storm damage data shows that the storm’s impact was
disproportionately borne by the region’s African American community, by people who rented
their homes, and by the poor and unemployed.
1. More than a third of the region’s 1.7 million residents lived in areas that suffered flooding or
moderate to catastrophic storm damage, according to FEMA. The majority of people living in
damaged areas were in the City of New Orleans (over 350,000), with additional concentrations in
suburban Jefferson Parish (175,000) and St. Bernard Parish (53,000) and along the Mississippi
Coast (54,000).
2. In the region as a whole, the disparities in storm damage are shown in the following
comparisons (arranged in order of the degree of disparity):
• By race. Damaged areas were 45.8% black, compared to 26.4% in undamaged areas.
• By housing tenure. 45.7% of homes in damaged areas were occupied by renters,
compared to 30.9% in undamaged communities.
• By poverty and employment status. 20.9% of households had incomes below the poverty
line in damaged areas, compared to 15.3% in undamaged areas. 7.6% of persons in the
labor force were unemployed in damaged areas (before the storm), compared to 6.0% in
undamaged areas.
3. These comparisons are heavily influenced by the experience of the City of New Orleans.
Outside the city, there were actually smaller shares of African American, poor, and unemployed
residents in the damaged areas.
4. Closer inspection of neighborhoods within New Orleans shows that some affluent white
neighborhoods were hard hit, while some poor minority neighborhoods were spared. Yet if the
post-Katrina city were limited to the population previously living in areas that were undamaged
by the storm – that is, if nobody were able to return to damaged neighborhoods – New Orleans is
at risk of losing more than 80% of its black population. This means that policy choices affecting
who can return, to which neighborhoods, and with what forms of public and private assistance,
will greatly affect the future character of the city.
Methods of analysis
Information on affected areas is drawn primarily from FEMA, which has provided detailed maps
of zones of flooding and wind damage. Areas of flooding do not further distinguish between
areas where water was deeper or remained for a longer time period. FEMA uses a standard
system for assessing other storm damage. The “limited damage” category is applied to areas
with “generally superficial damage to solid structures (e.g. loss of tiles or roof shingles); some
mobile homes and light structures are damaged or displaced.” These areas are not treated as
“damaged” in this report. The next more severe category is “moderate damage,” where “solid
structures sustain exterior damage (e.g. missing roofs or roof segments); some mobile homes and
light structures are destroyed, many are damaged or displaced.” Areas with moderate damage or
greater are treated here as “damaged.”
Within parts of the City of New Orleans and St. Bernard Parish, the FEMA map of “full extent”
of flood damage does not correspond well with reports on the ground. This primarily affects
parts of the French Quarter, Marigny, and Bywater in New Orleans (where flooding was less
extensive than reported by FEMA) and St. Bernard Parish (where some additional areas appear
to have been flooded). We have corrected the map of flooded areas in these zones based on
remote sensing imagery from September 2, 2005, made available by the Dartmouth Flood
Observatory (http://www.dartmouth.edu/~floods).
Photo by John R. Logan, 12/15/05
The photograph above illustrates the effects of heavy flooding in the Lower Ninth Ward of New
Orleans. In other parts of the city there was little structural damage, but many houses needed to
be entirely gutted as a result of days or weeks of being submerged in water.
2
The second photograph illustrates wind damage of the type sustained along much of the
Mississippi Coast as well as in New Orleans. This example would likely be classified as
moderate, and falls within the definition of damage used in this report.
Photo by John R. Logan, 12/16/05
Information about residents of these zones is from the Census 2000, using tract-level population
data. Changes that occurred between 2000 and 2005 are not reflected here. Geographic
Information System techniques have been used to determine the proportion of land area
identified as “flooded” or “moderately to catastrophically damaged” in each census tract in the
region. Where the damaged area only covers a portion of the tract, it is not possible to know
with precision how many residents of the tract lived in that portion. The analysis is based on the
assumption that people are spread equally across the tract, and that social characteristics of
residents of the damaged portion are the same as social characteristics of the tract as a whole.
An alternative approach would be to use census data at the level of block groups, which is a finer
unit of geography. This would make it possible to line up the neighborhood data more closely
with the boundaries of the damaged zones, though the match would still not be exact. However
block group data from the census is less reliable than data for census tracts, due to techniques
employed by the census bureau to maintain confidentiality of information and to correct for
missing data.
A similar analysis has been completed by the Congressional Research Service (available at
http://www.gnocdc.org/reports/crsrept.pdf). This report reaches similar conclusions regarding
the region as a whole, but provides more detailed information about variations among
neighborhoods in the City of New Orleans, where the majority of affected persons lived.
3
The report focuses on the New Orleans and Biloxi-Gulfport metropolitan areas of Louisiana and
Mississippi. About 4500 people are estimated to live in areas in the Mobile, AL metropolitan
region (Mobile and Baldwin Counties) that meet the FEMA damage criterion used here. A wider
geographic area has been designated by FEMA as qualifying for federal disaster relief, including
assistance to individuals and to public agencies. A much larger number of people were affected
by Katrina, either because they were forced to evacuate from their homes for extended periods,
or they suffered damage to their homes, or they lost power or other basic public services
(services that in large areas were still in the process of being restored as of January 2005). There
is no reliable source of information on the number or social composition of people who were
affected in these ways. Attention here, therefore, is limited to identifying whose neighborhoods
suffered the greatest impact of the storm, in areas that will likely require the most concentrated
attention in the recovery process.
To supplement this report, Brown University’s American Communities Project has developed a
web-based map system that includes all of the information analyzed here. The site is
http://www.s4.brown.edu/Katrina/index.html. It identifies the broad zones of the three-state
region where FEMA was authorized to provide assistance. It also shows the more specific areas
classified as flooded or moderately to catastrophically storm-damaged by FEMA. At a detailed
zoom level, the user can display aerial photographs from early September 2001 that have been
made available by the National Oceanic and Atmospheric Administration (NOAA) at
http://ngs.woc.noaa.gov/katrina. The web system also displays a wide variety of population
characteristics from Census 2000 for census tracts. Within the City of New Orleans, it is also
possible to see how census tracts are grouped into neighborhoods by the city for planning
4
purposes. This system also offers the capability to download data for a specific census tract or
group of tracts, and to copy the image of any map for use in other applications.
For reference, Figure 1 above provides a map of the affected region. It shows the boundaries of
the New Orleans metropolitan area in Louisiana, including the Parishes of Orleans (the city
proper), Jefferson, Plaquemine, St. Bernard, St. Charles, St. James, St. John the Baptist, and St.
Tammany, and the Biloxi-Gulfport metropolitan area in Mississippi, including the Counties of
Hancock, Harrison, and Jackson.
5
The population in areas of extensive damage
Within New Orleans and Biloxi-Gulfport metropolitan areas, flooding was concentrated in and
around New Orleans, and wind damage was focused on a narrow band along the Mississippi
Coast. Table 1 provides estimates of the population living in these areas, categorized by
metropolitan area and county or parish.
Nearly 650,000 persons, more than a third of the region’s population, lived in heavily damaged
areas. More than half of these, 354,000, lived in Orleans Parish. Another 175,000 lived in
adjacent Jefferson Parish, with 53,000 in St. Bernard Parish, and about 54,000 in the 3-county
area along the Mississippi Coast.
St. Bernard and Orleans Parishes had the highest shares of their populations in damaged zones,
both above 70%. More than 45% of residents of Hancock County, MS, lived in such areas, as
did more than 38% of residents of Jefferson Parish.
Table 1. Population in damaged and undamaged areas (Census 2000)
Population
Population in
in
Total
% of population in
damaged
Metropolis
County/Parish
areas
other areas
population
damaged areas
Biloxi-
Gulfport
Metro Total
54,424
309,564
363,988
15.0%
Hancock
19,398
23,569
42,967
45.1%
Harrison
27,779
161,822
189,601
14.7%
Jackson
7,247
124,173
131,420
5.5%
New Orleans
Metro Total
589,101
748,624
1,337,725
44.0%
City: Orleans
354,045
130,629
484,674
73.0%
Suburbs: Jefferson
174,694
280,771
455,465
38.4%
Plaquemines
4,235
22,522
26,757
15.8%
St. Bernard
53,029
14,200
67,229
78.9%
St. Charles
813
47,259
48,072
1.7%
St. James
0
21,216
21,216
0.0%
St. John the Baptist
0
43,044
43,044
0.0%
St. Tammany
2,285
188,983
191,268
1.2%
Region Total
643,525
1,058,188
1,701,713
37.8%
Note: "Damaged" areas are those identified by FEMA and the Dartmouth Flood Observatory as flooded or
with storm damage ranging from moderate to catastrophic.
6
Social composition of the population in damaged areas
Who are the people who lived in the most damaged areas? One way to answer this question is to
compare the social composition of damaged areas with that of the remaining less damaged parts
of the region. Table 2 does this for the following areas: the City of New Orleans itself, the larger
New Orleans metropolitan region including the city, and the Biloxi-Gulfport metropolitan region
in Mississippi. In the final rows are presented the data for all of these areas combined.
Table 2. Population social composition in damaged and undamaged areas (Census 2000)
Percent
Total
Percent Percent
Percent
Percent
>65
with
population
black
poor
unemployed renters disability
Biloxi-Gulfport
Metropolis
Damaged
54,424
14.8%
13.8%
6.1%
39.2%
6.8%
Other
309,564
20.4%
13.9%
6.0%
29.4%
5.2%
City of New Orleans
Damaged
354,045
75.0%
29.2%
10.4%
52.8%
5.7%
Other
130,629
46.2%
24.7%
7.2%
55.1%
5.3%
New Orleans suburbs
Damaged
235,056
9.1%
10.3%
4.5%
37.3%
5.7%
Other
617,995
25.2%
14.1%
5.7%
25.5%
4.6%
Total region
Damaged
643,525
45.8%
20.9%
7.6%
45.7%
5.8%
Other
1,058,188
26.4%
15.3%
6.0%
30.9%
4.8%
The data for the total region show that in several respects the neighborhoods of social groups
with least resources were the ones most affected by Katrina. The population of damaged areas
was nearly half black (45.8% compared to 26.4% black in the rest of the region), living in rental
housing (45.7% compared to 30.9%), and disproportionately below the poverty line (20.9%
compared to 15.3%) and unemployed (7.6% compared to 6.0%).
Certainly the storm struck the neighborhoods of many people of all backgrounds. If we examine
the absolute numbers behind these percentages, we note that there were almost as many non-
Hispanic whites as blacks in damaged areas (294,000 compared to 295,000), and the number of
persons in these neighborhoods above the poverty line was much larger than the number of poor
persons. This means that the suffering from the storm partly cut across racial and class lines.
But the odds of living in a damaged area were clearly much greater for blacks, renters, and poor
people. In these respects the most vulnerable residents turned out also to be at greatest risk. As
will be discussed in the conclusion, poor and black people also have fewer resources for
returning and rebuilding.
Some attention in the aftermath of Katrina was drawn to elderly persons who were stranded in
nursing homes or in their own residences. Particular concern has been expressed for elderly
persons who are disabled and therefore have the least mobility. Table 2 includes information on
the percentage of persons who are over 65 and report at least one disabling condition as defined
in Census 2000. Here, too, there is some disparity, but it is modest: 5.8% of persons in damaged
areas meet these criteria, only slightly more than the 4.8% in other neighborhoods.
Behind the overall pattern are some large differences within the region. The disparities by race
and poverty/unemployment can be traced more specifically to the City of New Orleans. In the
city, for example, 75% of residents of damaged areas were black, compared to undamaged areas
7
where a majority of residents were white. But in the suburbs, including both Jefferson Parish
and St. Bernard Parish that were hard hit by Katrina, as well as on the Mississippi Coast, the
affected neighborhoods were more likely to be predominantly white, and neighborhoods with
larger shares of black residents were actually less likely to be heavily damaged. In these parts of
the region, wealthy neighborhoods were as likely, in some cases even more likely, to be damaged
as were poor neighborhoods. In Jefferson and St. Bernard Parishes, for example, the
neighborhoods with higher educational and professional levels tended to be the hardest hit (this
is not shown in the table).
Disparities were reduced along the Mississippi Coast because many of the most desirable pre-
Katrina locations were located near the shoreline and therefore exposed to both high winds and
storm surge.
With respect to housing tenure, there is little difference within the City of New Orleans between
damaged and undamaged areas. In both sectors within the city the majority of homes were
renter-occupied. The overall disadvantage of renters has two sources. First, outside of the city
the neighborhoods of renters were more damaged. Second, a larger part of the city population
(with its large proportion of renters) was affected.
8
Differential impacts within the City of New Orleans
Because flooding within the City of New Orleans accounts for such a large share of the impacts
of Katrina, the final portion of this report is devoted to analyzing variations across
neighborhoods of the city. Figure 2 is a map showing census tracts in New Orleans as well as
suburban areas to the west and south. The red vertical line extending into Lake Pontchartrain is
the boundary between the city and neighboring Jefferson Parish. Damaged areas are identified
with a yellow boundary and yellow hatch lines. The interested reader can peruse similar maps
for any part of the three-state region, showing variations by race and many other population
characteristics, at http://www.s4.brown.edu/Katrina/index.html.
This map shows that the undamaged areas of the city were mainly in two areas. One is just north
of the Mississippi River in a zone extending westward from downtown. The other is across the
river on the West Bank, in a district known as Algiers. The map shows that some predominantly
white neighborhoods in the northwest part of the city were entirely damaged. However almost
all of the neighborhoods that were in the range of 75-100% black at the time of Census 2000
were damaged.
9
For planning purposes, New Orleans is divided into 13 planning districts and 72 distinct
neighborhoods within these districts. The neighborhood names and boundaries are shown in
Figure 3 (for reference, see the website of the Greater New Orleans Community Data Center,
www.gnocdc.org.)
Table 3 cumulates data from census tracts to these larger units, reporting the total population of
neighborhoods and planning districts as well as the number and percentage of residents in
damaged areas. Planning districts are listed in order from those with the highest portion of
residents in damaged areas to those with the least damage. Neighborhoods within planning
districts are listed likewise.
The Algiers Planning District, located across the river in what is called the West Bank, suffered
little damage, and it alone accounts for about 50,000 of the 130,000 New Orleans residents in
undamaged zones. Other areas with less than the average level of damage include lightly
populated New Aurora and Village de l’Est, the French Quarter/CBD, the Central City-Garden
District, and the Uptown-Carrollton Planning Districts. There is variation within these areas,
however. In the Central City-Garden District, for example, neighborhoods like the Garden
District itself, Irish Channel, and Touro had no damage, while 80% or more of residents of
Central City/Magnolia and Milan lived in damaged areas.
10
Table 3. Social characteristics of neighborhoods and planning districts in New Orleans
Percent
in
Population
damaged
in damaged
Total
Percent
Percent
Percent
Percent
Planning District
Neighborhood
areas
areas
population
black
renter
poor
unemployed
Mid-City
Total
100.0%
79,438
79,441
82.9%
72.9%
44.4%
16.4%
Bayou St. John
100.0%
4,861
4,861
68.1%
67.7%
32.0%
9.2%
Calliope Project
100.0%
4,339
4,339
98.7%
94.7%
69.2%
27.7%
Fairgrounds/Broad
100.0%
6,575
6,575
69.5%
54.4%
16.9%
7.1%
Gerttown/Zion City
100.0%
4,748
4,748
95.0%
76.1%
48.6%
42.5%
Iberville Project
100.0%
2,540
2,540
98.3%
100.0%
84.2%
44.9%
Mid-City
100.0%
19,909
19,909
64.7%
71.7%
32.1%
9.5%
Seventh Ward
100.0%
16,955
16,955
94.1%
67.0%
38.0%
13.8%
Sixth Ward/Treme/Lafitte
100.0%
8,850
8,853
92.8%
78.3%
56.9%
21.4%
St. Bernard Area/Project
100.0%
6,427
6,427
97.9%
83.0%
66.0%
20.8%
Tulane/Gravier
100.0%
4,234
4,234
78.4%
81.2%
56.2%
16.0%
New Orleans East
Total
99.2%
79,192
79,808
86.8%
44.7%
18.9%
7.7%
Pines Village
100.0%
5,092
5,092
87.9%
36.2%
18.3%
9.7%
West Lake Forest
100.0%
9,596
9,596
95.9%
76.2%
27.2%
11.0%
Plum Orchard
99.3%
6,957
7,005
93.7%
42.3%
33.2%
8.5%
Read Boulevard East
99.3%
8,180
8,240
74.2%
11.4%
11.2%
4.0%
Edgelake/Little Woods
99.0%
43,871
44,311
86.8%
48.6%
17.4%
7.6%
Read Boulevard West
98.8%
5,496
5,564
80.3%
14.9%
10.5%
6.3%
Gentilly
Total
96.5%
42,597
44,133
69.5%
28.2%
15.3%
7.1%
Gentilly Woods
100.0%
4,387
4,387
69.1%
24.3%
14.4%
9.0%
Gentilly Terrace
100.0%
10,542
10,542
70.4%
30.4%
16.1%
5.7%
Fillmore
100.0%
6,983
6,983
57.5%
15.4%
11.6%
6.4%
St. Anthony
100.0%
5,318
5,318
58.9%
40.0%
20.6%
7.1%
Milneburg
100.0%
5,640
5,640
76.2%
28.7%
14.4%
6.6%
Dillard
100.0%
6,471
6,471
89.7%
43.3%
20.6%
10.1%
Pontchartrain Park
99.9%
2,627
2,630
97.3%
7.8%
10.2%
6.6%
Lake Terrace/Lake Oaks
29.1%
629
2,162
19.4%
4.8%
1.9%
5.5%
Lower Ninth Ward
Total
92.6%
18,077
19,515
95.7%
46.0%
34.4%
13.6%
Lower Ninth Ward
99.9%
14,000
14,008
98.7%
41.1%
36.4%
13.5%
Holy Cross
74.0%
4,078
5,507
88.0%
58.0%
29.4%
13.7%
Lakeview
Total
89.8%
23,259
25,897
2.3%
33.8%
6.3%
2.7%
Navarre
100.0%
2,908
2,908
3.4%
45.1%
8.5%
4.4%
Lakeview
100.0%
9,875
9,875
0.8%
29.4%
4.9%
2.0%
City Park
100.0%
2,813
2,813
9.9%
58.4%
12.3%
6.7%
Lakewood
100.0%
1,961
1,962
1.7%
9.3%
1.3%
0.0%
Lakewood/West End
87.2%
4,120
4,724
1.7%
39.4%
9.1%
1.9%
Lakeshore/Lake Vista
43.8%
1,582
3,615
0.7%
14.4%
2.7%
1.3%
Bywater
Total
85.4%
35,139
41,163
83.4%
58.4%
38.7%
13.2%
Desire Project
100.0%
660
660
99.1%
83.8%
62.5%
28.8%
Florida Area
100.0%
3,171
3,171
98.7%
41.6%
36.2%
14.8%
St. Roch
100.0%
11,975
11,975
92.1%
57.9%
37.1%
14.4%
Percent
Total
Percent
Percent
Percent
Percent
Planning District
Neighborhood
Damaged
damaged
population
black
renter
poor
unemployed
Bywater (continued)
Florida Project
100.0%
1,604
1,604
98.4%
89.5%
79.6%
53.2%
Desire Area
99.8%
3,785
3,791
94.5%
52.0%
35.7%
9.1%
St. Claude
94.8%
11,112
11,721
90.9%
55.7%
39.0%
13.7%
Bywater
43.6%
2,223
5,096
61.3%
61.8%
38.6%
9.6%
Marigny
19.4%
609
3,145
18.3%
66.4%
24.1%
7.9%
Viavant/Venetian Is
Total
78.6%
11,270
14,342
47.8%
48.3%
33.1%
10.6%
Viavant/Venetian Isles
78.6%
11,270
14,342
47.8%
48.3%
33.1%
10.6%
Uptown-
Total
60.9%
40,850
67,083
46.6%
53.3%
24.3%
7.3%
Carrollton
Broadmoor
100.0%
7,232
7,232
68.9%
51.7%
31.8%
9.6%
Dixon
100.0%
1,772
1,772
95.0%
57.2%
31.1%
15.0%
Freret
100.0%
2,446
2,446
83.4%
64.3%
33.5%
20.6%
Hollygrove
100.0%
6,919
6,919
95.4%
45.7%
28.4%
9.7%
Marlyville/Fontainbleau
100.0%
6,740
6,740
28.4%
50.9%
12.9%
4.7%
Leonidas/West Carrollton
70.5%
6,315
8,953
75.9%
58.3%
31.5%
10.5%
Audubon/University
40.8%
6,085
14,898
5.5%
46.7%
17.9%
3.9%
East Carrollton
30.0%
1,332
4,438
31.9%
57.7%
24.5%
4.8%
Uptown
29.1%
1,947
6,681
36.3%
56.8%
23.9%
6.1%
West Riverside
1.2%
63
5,232
36.7%
57.1%
18.1%
6.8%
Black Pearl
0.0%
0
1,772
36.7%
61.9%
26.4%
8.3%
Central City/
Total
46.8%
22,599
48,327
67.8%
73.6%
39.7%
12.3%
Garden District
Milan
96.9%
7,247
7,480
74.2%
65.6%
28.6%
9.2%
Central City/Magnolia
79.8%
15,215
19,072
87.4%
83.7%
49.8%
20.4%
St. Thomas Area
2.0%
122
6,116
34.8%
73.5%
28.5%
6.9%
St. Thomas Project
0.4%
12
2,957
93.3%
93.0%
69.1%
24.2%
Touro
0.1%
2
3,242
18.6%
70.4%
15.5%
1.8%
East Riverside
0.0%
0
3,220
64.1%
57.1%
36.9%
11.5%
Garden District
0.0%
0
1,970
3.0%
51.6%
11.3%
2.4%
Irish Channel
0.0%
0
4,270
69.0%
62.9%
41.1%
12.4%
French Quarter/
Total
12.2%
729
5,970
13.3%
76.4%
16.9%
6.5%
CBD
Central Business District
37.9%
681
1,794
33.6%
79.1%
32.3%
12.0%
Vieux Carre
1.1%
48
4,176
4.5%
75.5%
10.8%
4.8%
Village de L'Est
Total
10.0%
222
2,213
83.2%
54.3%
7.8%
4.3%
Village De L'Est
10.0%
222
2,213
83.2%
54.3%
7.8%
4.3%
Algiers
Total
1.2%
612
51,110
56.6%
48.3%
24.1%
7.3%
Algiers Whitney
7.1%
183
2,564
85.4%
50.0%
29.3%
14.7%
Algiers Naval Station
2.0%
57
2,902
64.1%
49.4%
21.8%
5.1%
Tall Timbers/Brechtel
1.2%
142
12,177
55.0%
64.6%
19.4%
6.3%
Aurora/Walnut Bend/
Huntlee Village
0.9%
144
15,807
31.3%
26.3%
9.9%
4.9%
Algiers Point
0.7%
17
2,381
25.3%
51.7%
17.3%
3.5%
Behrman
0.7%
68
10,430
78.0%
53.1%
33.4%
9.8%
Fischer Project
0.0%
0
2,034
99.2%
88.5%
88.2%
24.7%
McDonogh
0.0%
0
2,815
88.2%
52.9%
48.3%
16.3%
New Aurora
Total
1.1%
62
5,672
68.2%
27.0%
24.8%
10.8%
River Park/Cut Off/
Lower Coast
1.1%
62
5,672
68.2%
27.0%
24.8%
10.8%
City of New Orleans Total
73.0%
354,045
484,674
67.2%
53.5%
27.9%
9.4%
12
At the other end of the spectrum, Mid-City, New Orleans East, Gentilly, the Lower Ninth Ward,
Bywater, and Lakeview all had 90% or more of their residents in damaged areas.
Table 3 also provides information about the social composition of these areas. The Table reveals
great variation within New Orleans. Some largely white neighborhoods of affluent homeowners
were completely flooded, while some relatively poor black neighborhoods were spared.
However there is a general tendency as shown in Table 2 for blacks and poor residents to have
greater odds of being in harm’s way.
Discussions of the racially differential impact of Katrina have often emphasized the Lower Ninth
Ward (where many homes were entirely demolished by the breach in the levee of the Industrial
Canal) and New Orleans East. Most neighborhoods in these planning districts were more than
85% black, and most residences were damaged. A majority of residents of both of these
planning districts were homeowners, though there were clear class distinctions between the two
areas. More than a third of Lower Ninth Ward residents were below the poverty line, and nearly
14% were unemployed. New Orleans East had a considerably larger middle class component,
though it was not among the city’s most affluent sections.
Many of the most segregated neighborhoods with the highest poverty rates are those identified as
“projects,” a reference to the prominence of public housing within their borders. The project
neighborhoods typically had poverty rates in the range of 60-80% of the population,
unemployment is above 20%, they were all predominantly black (with African Americans
accounting for 90% or more of their residents), and 80% or more of residents were renters.
There are six such neighborhoods in New Orleans (though there are concentrations of public
housing or Section 8 housing in other parts of the city). In five of them with a combined 2000
population of over 15,000 persons (Calliope, Iberville, St. Bernard Area, Desire, and Florida) the
entire territory meets this report’s definition of damaged areas. The Fischer Project
neighborhood in Algiers was little damaged. (A seventh project neighborhood, the St. Thomas
Project in the Central City/Garden District, was demolished in 2002, replaced by a Wal-Mart and
new predominantly market-rate condominiums. Therefore the census data from 2000 shown in
Table 3 do not reflect the population composition at the time of Katrina.)
At the other end of the class spectrum are a number of more advantaged neighborhoods with
poverty rates below 10% or unemployment rates below 5%. In the most heavily impacted
planning districts, few neighborhoods meet either criterion. These include the Lake-
Terrace/Lake Oaks neighborhood in Gentilly and the Read Boulevard East neighborhood in New
Orleans East. Most such neighborhoods are in the Lakeview Planning District, which is an area
with a small black population, mostly homeowners, and very low rates of poverty and
unemployment. Here only the Lakeshore/Lake Vista neighborhood, adjacent to Lake
Pontchartrain, was partly spared.
Few residents in the French Quarter, a predominantly white neighborhood with a poverty rate of
about 11% and unemployment below 5%, lived in tracts that were flooded.
Among other neighborhoods with a national reputation for affluence, the Garden District
neighborhood was not flooded and only 40% of the Audubon/University neighborhood (home of
Tulane University and Loyola University) was damaged.
13
Policy implications
The sheer number of people who lived in heavily damaged areas – over 640,000 – is a reminder
of the scale of Katrina’s impact. Because the storm hit large numbers of people of every race
and class, it seems likely that public support for policies to assist these people will also cut across
race and class lines. However there was also a substantial disproportionate impact on African
Americans and people with fewer resources. These disparities stem from within the City of New
Orleans itself, and more specifically from vulnerability to flooding. This is a pattern with deep
roots, and although Katrina caused the most extensive flooding in memory, prior studies by
historians (such as An Unnatural Metropolis by Craig Colten) have demonstrated that both high
ground and public investments in drainage and pumping systems consistently worked to the
advantage of certain neighborhoods in past storms.
There are major variations across the region that are likely to affect the process of recovery.
Damage was extensive on the Mississippi Coast, and the area’s largest single source of
employment – casino gambling – was knocked out of operation. In comparison to New Orleans,
however, the number of people living in areas of moderate or greater damage was small, only
about 50,000. And also in contrast to New Orleans, only a small share of these people were
black and a majority were homeowners. It is difficult to assess the importance of race in
recovery policy in Mississippi, but in a politically conservative state it could make a big
difference that white homeowners constitute the bulk of claimants for state assistance. Further,
these people are easier to serve for several reasons.
1. First, they are identifiable and – because they retain an ownership interest in their
properties – they should prove easier for authorities to contact.
2. Second, since much of the damage wrought by Katrina in this area was by wind and rain
damage, standard homeowner policies offer substantial private sector coverage of
damage losses. For those with uninsured flood damage, the state government currently
expects federal aid to be sufficient to fund payments of $150,000 to individual
homeowners.
3. Third, the low density of housing in this area means that typically even when one’s home
was uninhabitable, there was space for a trailer in the driveway. Since in addition the
loss of electrical power was relatively short-term in Mississippi, and basic public services
could be restored within a reasonable time, homeowners in this region more readily met
the requirements for a FEMA-provided trailer – space and confirmed utility hookups.
In contrast, consider the situation in New Orleans. More than half the persons in damaged areas
were renters, unlikely to be protected in any way by property insurance, and 30% fell below the
poverty line and were therefore unlikely to have personal resources to return to the city. By the
end of 2005, power was still unavailable much of the city, and actual connections to electric
power required residents to present evidence of inspection by a licensed electrician before power
would be restored to an individual home. The utility company (a subsidiary of the Entergy
Corporation) had filed for bankruptcy protection in September.
People who previously lived in public housing seem to have the least chances to return, given
current policy. All public housing in the affected areas has been closed (and special barriers
14
bolted to the doors), and residents have been allotted rental housing assistance in areas where
they have relocated for up to 18 months. Plans for reopening the projects or for constructing new
affordable housing have not become public.
For many of the same reasons that rebuilding will be facilitated on the Mississippi Coast, the
white residents of the City of New Orleans are more likely than black residents to be able to
return to their neighborhoods, even if the neighborhood is reopened. Whites are more likely to
be homeowners (55% compared to 42% among African American households), but more
important, they are much more likely to have the personal resources to reinvest in their homes or
to find a new residence in a difficult housing market. In the pre-Katrina black population, 35%
were below the poverty line and the median household income was only $25,000. Among
whites, only 11% were poor and the median income was more than twice as high – $61,000.
Therefore even among homeowners, blacks are less likely to have the means to rebuild than are
whites.
There is potential for political coalitions that cut across the racial and class divisions that have
helped structure city politics over the decades. Residents of such very different neighborhoods
as Lakeview and the Lower Ninth Ward have a shared interest in short-term assistance programs
such as subsidies for temporary housing outside the city. Up to now few city residents have
qualified for FEMA trailers outside their homes because they do not own the home, or there is
insufficient space for a trailer, or public utilities remain unavailable. As long as issues can be
framed so that demands are oriented toward FEMA (an external target) or toward demands for
services like electricity or schools or police protection that affect all segments of the public, the
appearance of unity can be maintained. The City Council’s repeated stands in favor of
rebuilding all parts of the city – a question on which the Council is unlikely to have the final
word – is a reflection of this temporary unity.
Yet variations across neighborhoods – and across race and class – are likely to support the
emergence of a sense of conflicting interests. In December 2005 conflict took the form of
opposition to proposals to locate FEMA trailers in public spaces within neighborhoods that
sustained less damage. In this case the interests of advantaged neighborhoods (advantaged by
protection from flooding and by having residents in place to express their views) were in conflict
with the interests of absent residents who have no place to return. Not surprisingly the City
Council gave its members veto power over new trailer parks in the areas that they represent.
In the longer term there is likely to be competition between damaged neighborhoods for the
supports that will be necessary for rebuilding. In mid-January 2006, the Bring New Orleans
Back Commission (a policy group formed by city government) released a planning report that
began to address this question. The Commission identified some parts of the city as “immediate
opportunity areas.” These areas are shaded in yellow on the map in Figure 4. Compared to the
damaged areas shown in Figure 2 above, it is clear that the Commission proposes rebuilding in
some areas that were flooded. They include areas along the river near the central business
district, Holy Cross in the Lower Ninth Ward Planning District, and Lakewood/West End and
Lakeshore/Lake Vista bordering Lake Pontchartrain. Generally these are zones where sustained
flooding was under about four feet.
The Commission also recommended the designation of large portions of the city as
“neighborhood planning” areas. Within these areas the Commission recommends a temporary
15
halt on issuance of building permits. Only where there is evidence within the next four months
that residents are committed to returning to each neighborhood in large numbers would
rebuilding be approved.
Figure 4. Recommended areas of rebuilding in the City of New Orleans
Source: Bring Back New Orleans Commission, Action Plan for New Orleans: The New
American City. January 11, 2006. (http://bringneworleansback.org)
The precise criteria underlying these designations have not been announced, and the
Commission’s recommendations have not been formally adopted. This is a temporary solution
to the political problem that few public officials are willing to state openly that some areas will
not be permitted to be resettled. On its face it leaves the decision to local residents. Clearly
some areas are at risk of being closed to reconstruction. In January 2006, the full-time
population of the city has been estimated at only 150,000. The analysis in this report suggests
that if the future city were limited to the population previously living in zones undamaged by
Katrina it would risk losing about 50% of its white residents but more than 80% of its black
population. This is why the continuing question about the hurricane is this: whose city will be
rebuilt?
16
File contents
John R. Logan
Professor of Sociology, Brown University
Director, Spatial Structures in the Social Sciences
(401) 863-2267
John_Logan@brown.edu
The Impact of Katrina: Race and Class in Storm-Damaged Neighborhoods
Early media reports about the wind damage and flooding caused by Hurricane Katrina focused
on New Orleans, and especially on the people who had been unable to escape the city before it
flooded. Images of poor and predominantly black people crowded into the Superdome and
Convention Center supported the impression that Katrina had disproportionately affected poor,
black neighborhoods. The purpose of this report is to evaluate more precisely what
neighborhoods were heavily damaged, including not only New Orleans but also the coastal
communities in Mississippi that bore the brunt of hurricane-force winds.
In brief an analysis of FEMA storm damage data shows that the storm’s impact was
disproportionately borne by the region’s African American community, by people who rented
their homes, and by the poor and unemployed.
1. More than a third of the region’s 1.7 million residents lived in areas that suffered flooding or
moderate to catastrophic storm damage, according to FEMA. The majority of people living in
damaged areas were in the City of New Orleans (over 350,000), with additional concentrations in
suburban Jefferson Parish (175,000) and St. Bernard Parish (53,000) and along the Mississippi
Coast (54,000).
2. In the region as a whole, the disparities in storm damage are shown in the following
comparisons (arranged in order of the degree of disparity):
• By race. Damaged areas were 45.8% black, compared to 26.4% in undamaged areas.
• By housing tenure. 45.7% of homes in damaged areas were occupied by renters,
compared to 30.9% in undamaged communities.
• By poverty and employment status. 20.9% of households had incomes below the poverty
line in damaged areas, compared to 15.3% in undamaged areas. 7.6% of persons in the
labor force were unemployed in damaged areas (before the storm), compared to 6.0% in
undamaged areas.
3. These comparisons are heavily influenced by the experience of the City of New Orleans.
Outside the city, there were actually smaller shares of African American, poor, and unemployed
residents in the damaged areas.
4. Closer inspection of neighborhoods within New Orleans shows that some affluent white
neighborhoods were hard hit, while some poor minority neighborhoods were spared. Yet if the
post-Katrina city were limited to the population previously living in areas that were undamaged
by the storm – that is, if nobody were able to return to damaged neighborhoods – New Orleans is
at risk of losing more than 80% of its black population. This means that policy choices affecting
who can return, to which neighborhoods, and with what forms of public and private assistance,
will greatly affect the future character of the city.
Methods of analysis
Information on affected areas is drawn primarily from FEMA, which has provided detailed maps
of zones of flooding and wind damage. Areas of flooding do not further distinguish between
areas where water was deeper or remained for a longer time period. FEMA uses a standard
system for assessing other storm damage. The “limited damage” category is applied to areas
with “generally superficial damage to solid structures (e.g. loss of tiles or roof shingles); some
mobile homes and light structures are damaged or displaced.” These areas are not treated as
“damaged” in this report. The next more severe category is “moderate damage,” where “solid
structures sustain exterior damage (e.g. missing roofs or roof segments); some mobile homes and
light structures are destroyed, many are damaged or displaced.” Areas with moderate damage or
greater are treated here as “damaged.”
Within parts of the City of New Orleans and St. Bernard Parish, the FEMA map of “full extent”
of flood damage does not correspond well with reports on the ground. This primarily affects
parts of the French Quarter, Marigny, and Bywater in New Orleans (where flooding was less
extensive than reported by FEMA) and St. Bernard Parish (where some additional areas appear
to have been flooded). We have corrected the map of flooded areas in these zones based on
remote sensing imagery from September 2, 2005, made available by the Dartmouth Flood
Observatory (http://www.dartmouth.edu/~floods).
Photo by John R. Logan, 12/15/05
The photograph above illustrates the effects of heavy flooding in the Lower Ninth Ward of New
Orleans. In other parts of the city there was little structural damage, but many houses needed to
be entirely gutted as a result of days or weeks of being submerged in water.
2
The second photograph illustrates wind damage of the type sustained along much of the
Mississippi Coast as well as in New Orleans. This example would likely be classified as
moderate, and falls within the definition of damage used in this report.
Photo by John R. Logan, 12/16/05
Information about residents of these zones is from the Census 2000, using tract-level population
data. Changes that occurred between 2000 and 2005 are not reflected here. Geographic
Information System techniques have been used to determine the proportion of land area
identified as “flooded” or “moderately to catastrophically damaged” in each census tract in the
region. Where the damaged area only covers a portion of the tract, it is not possible to know
with precision how many residents of the tract lived in that portion. The analysis is based on the
assumption that people are spread equally across the tract, and that social characteristics of
residents of the damaged portion are the same as social characteristics of the tract as a whole.
An alternative approach would be to use census data at the level of block groups, which is a finer
unit of geography. This would make it possible to line up the neighborhood data more closely
with the boundaries of the damaged zones, though the match would still not be exact. However
block group data from the census is less reliable than data for census tracts, due to techniques
employed by the census bureau to maintain confidentiality of information and to correct for
missing data.
A similar analysis has been completed by the Congressional Research Service (available at
http://www.gnocdc.org/reports/crsrept.pdf). This report reaches similar conclusions regarding
the region as a whole, but provides more detailed information about variations among
neighborhoods in the City of New Orleans, where the majority of affected persons lived.
3
The report focuses on the New Orleans and Biloxi-Gulfport metropolitan areas of Louisiana and
Mississippi. About 4500 people are estimated to live in areas in the Mobile, AL metropolitan
region (Mobile and Baldwin Counties) that meet the FEMA damage criterion used here. A wider
geographic area has been designated by FEMA as qualifying for federal disaster relief, including
assistance to individuals and to public agencies. A much larger number of people were affected
by Katrina, either because they were forced to evacuate from their homes for extended periods,
or they suffered damage to their homes, or they lost power or other basic public services
(services that in large areas were still in the process of being restored as of January 2005). There
is no reliable source of information on the number or social composition of people who were
affected in these ways. Attention here, therefore, is limited to identifying whose neighborhoods
suffered the greatest impact of the storm, in areas that will likely require the most concentrated
attention in the recovery process.
To supplement this report, Brown University’s American Communities Project has developed a
web-based map system that includes all of the information analyzed here. The site is
http://www.s4.brown.edu/Katrina/index.html. It identifies the broad zones of the three-state
region where FEMA was authorized to provide assistance. It also shows the more specific areas
classified as flooded or moderately to catastrophically storm-damaged by FEMA. At a detailed
zoom level, the user can display aerial photographs from early September 2001 that have been
made available by the National Oceanic and Atmospheric Administration (NOAA) at
http://ngs.woc.noaa.gov/katrina. The web system also displays a wide variety of population
characteristics from Census 2000 for census tracts. Within the City of New Orleans, it is also
possible to see how census tracts are grouped into neighborhoods by the city for planning
4
purposes. This system also offers the capability to download data for a specific census tract or
group of tracts, and to copy the image of any map for use in other applications.
For reference, Figure 1 above provides a map of the affected region. It shows the boundaries of
the New Orleans metropolitan area in Louisiana, including the Parishes of Orleans (the city
proper), Jefferson, Plaquemine, St. Bernard, St. Charles, St. James, St. John the Baptist, and St.
Tammany, and the Biloxi-Gulfport metropolitan area in Mississippi, including the Counties of
Hancock, Harrison, and Jackson.
5
The population in areas of extensive damage
Within New Orleans and Biloxi-Gulfport metropolitan areas, flooding was concentrated in and
around New Orleans, and wind damage was focused on a narrow band along the Mississippi
Coast. Table 1 provides estimates of the population living in these areas, categorized by
metropolitan area and county or parish.
Nearly 650,000 persons, more than a third of the region’s population, lived in heavily damaged
areas. More than half of these, 354,000, lived in Orleans Parish. Another 175,000 lived in
adjacent Jefferson Parish, with 53,000 in St. Bernard Parish, and about 54,000 in the 3-county
area along the Mississippi Coast.
St. Bernard and Orleans Parishes had the highest shares of their populations in damaged zones,
both above 70%. More than 45% of residents of Hancock County, MS, lived in such areas, as
did more than 38% of residents of Jefferson Parish.
Table 1. Population in damaged and undamaged areas (Census 2000)
Population
Population in
in
Total
% of population in
damaged
Metropolis
County/Parish
areas
other areas
population
damaged areas
Biloxi-
Gulfport
Metro Total
54,424
309,564
363,988
15.0%
Hancock
19,398
23,569
42,967
45.1%
Harrison
27,779
161,822
189,601
14.7%
Jackson
7,247
124,173
131,420
5.5%
New Orleans
Metro Total
589,101
748,624
1,337,725
44.0%
City: Orleans
354,045
130,629
484,674
73.0%
Suburbs: Jefferson
174,694
280,771
455,465
38.4%
Plaquemines
4,235
22,522
26,757
15.8%
St. Bernard
53,029
14,200
67,229
78.9%
St. Charles
813
47,259
48,072
1.7%
St. James
0
21,216
21,216
0.0%
St. John the Baptist
0
43,044
43,044
0.0%
St. Tammany
2,285
188,983
191,268
1.2%
Region Total
643,525
1,058,188
1,701,713
37.8%
Note: "Damaged" areas are those identified by FEMA and the Dartmouth Flood Observatory as flooded or
with storm damage ranging from moderate to catastrophic.
6
Social composition of the population in damaged areas
Who are the people who lived in the most damaged areas? One way to answer this question is to
compare the social composition of damaged areas with that of the remaining less damaged parts
of the region. Table 2 does this for the following areas: the City of New Orleans itself, the larger
New Orleans metropolitan region including the city, and the Biloxi-Gulfport metropolitan region
in Mississippi. In the final rows are presented the data for all of these areas combined.
Table 2. Population social composition in damaged and undamaged areas (Census 2000)
Percent
Total
Percent Percent
Percent
Percent
>65
with
population
black
poor
unemployed renters disability
Biloxi-Gulfport
Metropolis
Damaged
54,424
14.8%
13.8%
6.1%
39.2%
6.8%
Other
309,564
20.4%
13.9%
6.0%
29.4%
5.2%
City of New Orleans
Damaged
354,045
75.0%
29.2%
10.4%
52.8%
5.7%
Other
130,629
46.2%
24.7%
7.2%
55.1%
5.3%
New Orleans suburbs
Damaged
235,056
9.1%
10.3%
4.5%
37.3%
5.7%
Other
617,995
25.2%
14.1%
5.7%
25.5%
4.6%
Total region
Damaged
643,525
45.8%
20.9%
7.6%
45.7%
5.8%
Other
1,058,188
26.4%
15.3%
6.0%
30.9%
4.8%
The data for the total region show that in several respects the neighborhoods of social groups
with least resources were the ones most affected by Katrina. The population of damaged areas
was nearly half black (45.8% compared to 26.4% black in the rest of the region), living in rental
housing (45.7% compared to 30.9%), and disproportionately below the poverty line (20.9%
compared to 15.3%) and unemployed (7.6% compared to 6.0%).
Certainly the storm struck the neighborhoods of many people of all backgrounds. If we examine
the absolute numbers behind these percentages, we note that there were almost as many non-
Hispanic whites as blacks in damaged areas (294,000 compared to 295,000), and the number of
persons in these neighborhoods above the poverty line was much larger than the number of poor
persons. This means that the suffering from the storm partly cut across racial and class lines.
But the odds of living in a damaged area were clearly much greater for blacks, renters, and poor
people. In these respects the most vulnerable residents turned out also to be at greatest risk. As
will be discussed in the conclusion, poor and black people also have fewer resources for
returning and rebuilding.
Some attention in the aftermath of Katrina was drawn to elderly persons who were stranded in
nursing homes or in their own residences. Particular concern has been expressed for elderly
persons who are disabled and therefore have the least mobility. Table 2 includes information on
the percentage of persons who are over 65 and report at least one disabling condition as defined
in Census 2000. Here, too, there is some disparity, but it is modest: 5.8% of persons in damaged
areas meet these criteria, only slightly more than the 4.8% in other neighborhoods.
Behind the overall pattern are some large differences within the region. The disparities by race
and poverty/unemployment can be traced more specifically to the City of New Orleans. In the
city, for example, 75% of residents of damaged areas were black, compared to undamaged areas
7
where a majority of residents were white. But in the suburbs, including both Jefferson Parish
and St. Bernard Parish that were hard hit by Katrina, as well as on the Mississippi Coast, the
affected neighborhoods were more likely to be predominantly white, and neighborhoods with
larger shares of black residents were actually less likely to be heavily damaged. In these parts of
the region, wealthy neighborhoods were as likely, in some cases even more likely, to be damaged
as were poor neighborhoods. In Jefferson and St. Bernard Parishes, for example, the
neighborhoods with higher educational and professional levels tended to be the hardest hit (this
is not shown in the table).
Disparities were reduced along the Mississippi Coast because many of the most desirable pre-
Katrina locations were located near the shoreline and therefore exposed to both high winds and
storm surge.
With respect to housing tenure, there is little difference within the City of New Orleans between
damaged and undamaged areas. In both sectors within the city the majority of homes were
renter-occupied. The overall disadvantage of renters has two sources. First, outside of the city
the neighborhoods of renters were more damaged. Second, a larger part of the city population
(with its large proportion of renters) was affected.
8
Differential impacts within the City of New Orleans
Because flooding within the City of New Orleans accounts for such a large share of the impacts
of Katrina, the final portion of this report is devoted to analyzing variations across
neighborhoods of the city. Figure 2 is a map showing census tracts in New Orleans as well as
suburban areas to the west and south. The red vertical line extending into Lake Pontchartrain is
the boundary between the city and neighboring Jefferson Parish. Damaged areas are identified
with a yellow boundary and yellow hatch lines. The interested reader can peruse similar maps
for any part of the three-state region, showing variations by race and many other population
characteristics, at http://www.s4.brown.edu/Katrina/index.html.
This map shows that the undamaged areas of the city were mainly in two areas. One is just north
of the Mississippi River in a zone extending westward from downtown. The other is across the
river on the West Bank, in a district known as Algiers. The map shows that some predominantly
white neighborhoods in the northwest part of the city were entirely damaged. However almost
all of the neighborhoods that were in the range of 75-100% black at the time of Census 2000
were damaged.
9
For planning purposes, New Orleans is divided into 13 planning districts and 72 distinct
neighborhoods within these districts. The neighborhood names and boundaries are shown in
Figure 3 (for reference, see the website of the Greater New Orleans Community Data Center,
www.gnocdc.org.)
Table 3 cumulates data from census tracts to these larger units, reporting the total population of
neighborhoods and planning districts as well as the number and percentage of residents in
damaged areas. Planning districts are listed in order from those with the highest portion of
residents in damaged areas to those with the least damage. Neighborhoods within planning
districts are listed likewise.
The Algiers Planning District, located across the river in what is called the West Bank, suffered
little damage, and it alone accounts for about 50,000 of the 130,000 New Orleans residents in
undamaged zones. Other areas with less than the average level of damage include lightly
populated New Aurora and Village de l’Est, the French Quarter/CBD, the Central City-Garden
District, and the Uptown-Carrollton Planning Districts. There is variation within these areas,
however. In the Central City-Garden District, for example, neighborhoods like the Garden
District itself, Irish Channel, and Touro had no damage, while 80% or more of residents of
Central City/Magnolia and Milan lived in damaged areas.
10
Table 3. Social characteristics of neighborhoods and planning districts in New Orleans
Percent
in
Population
damaged
in damaged
Total
Percent
Percent
Percent
Percent
Planning District
Neighborhood
areas
areas
population
black
renter
poor
unemployed
Mid-City
Total
100.0%
79,438
79,441
82.9%
72.9%
44.4%
16.4%
Bayou St. John
100.0%
4,861
4,861
68.1%
67.7%
32.0%
9.2%
Calliope Project
100.0%
4,339
4,339
98.7%
94.7%
69.2%
27.7%
Fairgrounds/Broad
100.0%
6,575
6,575
69.5%
54.4%
16.9%
7.1%
Gerttown/Zion City
100.0%
4,748
4,748
95.0%
76.1%
48.6%
42.5%
Iberville Project
100.0%
2,540
2,540
98.3%
100.0%
84.2%
44.9%
Mid-City
100.0%
19,909
19,909
64.7%
71.7%
32.1%
9.5%
Seventh Ward
100.0%
16,955
16,955
94.1%
67.0%
38.0%
13.8%
Sixth Ward/Treme/Lafitte
100.0%
8,850
8,853
92.8%
78.3%
56.9%
21.4%
St. Bernard Area/Project
100.0%
6,427
6,427
97.9%
83.0%
66.0%
20.8%
Tulane/Gravier
100.0%
4,234
4,234
78.4%
81.2%
56.2%
16.0%
New Orleans East
Total
99.2%
79,192
79,808
86.8%
44.7%
18.9%
7.7%
Pines Village
100.0%
5,092
5,092
87.9%
36.2%
18.3%
9.7%
West Lake Forest
100.0%
9,596
9,596
95.9%
76.2%
27.2%
11.0%
Plum Orchard
99.3%
6,957
7,005
93.7%
42.3%
33.2%
8.5%
Read Boulevard East
99.3%
8,180
8,240
74.2%
11.4%
11.2%
4.0%
Edgelake/Little Woods
99.0%
43,871
44,311
86.8%
48.6%
17.4%
7.6%
Read Boulevard West
98.8%
5,496
5,564
80.3%
14.9%
10.5%
6.3%
Gentilly
Total
96.5%
42,597
44,133
69.5%
28.2%
15.3%
7.1%
Gentilly Woods
100.0%
4,387
4,387
69.1%
24.3%
14.4%
9.0%
Gentilly Terrace
100.0%
10,542
10,542
70.4%
30.4%
16.1%
5.7%
Fillmore
100.0%
6,983
6,983
57.5%
15.4%
11.6%
6.4%
St. Anthony
100.0%
5,318
5,318
58.9%
40.0%
20.6%
7.1%
Milneburg
100.0%
5,640
5,640
76.2%
28.7%
14.4%
6.6%
Dillard
100.0%
6,471
6,471
89.7%
43.3%
20.6%
10.1%
Pontchartrain Park
99.9%
2,627
2,630
97.3%
7.8%
10.2%
6.6%
Lake Terrace/Lake Oaks
29.1%
629
2,162
19.4%
4.8%
1.9%
5.5%
Lower Ninth Ward
Total
92.6%
18,077
19,515
95.7%
46.0%
34.4%
13.6%
Lower Ninth Ward
99.9%
14,000
14,008
98.7%
41.1%
36.4%
13.5%
Holy Cross
74.0%
4,078
5,507
88.0%
58.0%
29.4%
13.7%
Lakeview
Total
89.8%
23,259
25,897
2.3%
33.8%
6.3%
2.7%
Navarre
100.0%
2,908
2,908
3.4%
45.1%
8.5%
4.4%
Lakeview
100.0%
9,875
9,875
0.8%
29.4%
4.9%
2.0%
City Park
100.0%
2,813
2,813
9.9%
58.4%
12.3%
6.7%
Lakewood
100.0%
1,961
1,962
1.7%
9.3%
1.3%
0.0%
Lakewood/West End
87.2%
4,120
4,724
1.7%
39.4%
9.1%
1.9%
Lakeshore/Lake Vista
43.8%
1,582
3,615
0.7%
14.4%
2.7%
1.3%
Bywater
Total
85.4%
35,139
41,163
83.4%
58.4%
38.7%
13.2%
Desire Project
100.0%
660
660
99.1%
83.8%
62.5%
28.8%
Florida Area
100.0%
3,171
3,171
98.7%
41.6%
36.2%
14.8%
St. Roch
100.0%
11,975
11,975
92.1%
57.9%
37.1%
14.4%
Percent
Total
Percent
Percent
Percent
Percent
Planning District
Neighborhood
Damaged
damaged
population
black
renter
poor
unemployed
Bywater (continued)
Florida Project
100.0%
1,604
1,604
98.4%
89.5%
79.6%
53.2%
Desire Area
99.8%
3,785
3,791
94.5%
52.0%
35.7%
9.1%
St. Claude
94.8%
11,112
11,721
90.9%
55.7%
39.0%
13.7%
Bywater
43.6%
2,223
5,096
61.3%
61.8%
38.6%
9.6%
Marigny
19.4%
609
3,145
18.3%
66.4%
24.1%
7.9%
Viavant/Venetian Is
Total
78.6%
11,270
14,342
47.8%
48.3%
33.1%
10.6%
Viavant/Venetian Isles
78.6%
11,270
14,342
47.8%
48.3%
33.1%
10.6%
Uptown-
Total
60.9%
40,850
67,083
46.6%
53.3%
24.3%
7.3%
Carrollton
Broadmoor
100.0%
7,232
7,232
68.9%
51.7%
31.8%
9.6%
Dixon
100.0%
1,772
1,772
95.0%
57.2%
31.1%
15.0%
Freret
100.0%
2,446
2,446
83.4%
64.3%
33.5%
20.6%
Hollygrove
100.0%
6,919
6,919
95.4%
45.7%
28.4%
9.7%
Marlyville/Fontainbleau
100.0%
6,740
6,740
28.4%
50.9%
12.9%
4.7%
Leonidas/West Carrollton
70.5%
6,315
8,953
75.9%
58.3%
31.5%
10.5%
Audubon/University
40.8%
6,085
14,898
5.5%
46.7%
17.9%
3.9%
East Carrollton
30.0%
1,332
4,438
31.9%
57.7%
24.5%
4.8%
Uptown
29.1%
1,947
6,681
36.3%
56.8%
23.9%
6.1%
West Riverside
1.2%
63
5,232
36.7%
57.1%
18.1%
6.8%
Black Pearl
0.0%
0
1,772
36.7%
61.9%
26.4%
8.3%
Central City/
Total
46.8%
22,599
48,327
67.8%
73.6%
39.7%
12.3%
Garden District
Milan
96.9%
7,247
7,480
74.2%
65.6%
28.6%
9.2%
Central City/Magnolia
79.8%
15,215
19,072
87.4%
83.7%
49.8%
20.4%
St. Thomas Area
2.0%
122
6,116
34.8%
73.5%
28.5%
6.9%
St. Thomas Project
0.4%
12
2,957
93.3%
93.0%
69.1%
24.2%
Touro
0.1%
2
3,242
18.6%
70.4%
15.5%
1.8%
East Riverside
0.0%
0
3,220
64.1%
57.1%
36.9%
11.5%
Garden District
0.0%
0
1,970
3.0%
51.6%
11.3%
2.4%
Irish Channel
0.0%
0
4,270
69.0%
62.9%
41.1%
12.4%
French Quarter/
Total
12.2%
729
5,970
13.3%
76.4%
16.9%
6.5%
CBD
Central Business District
37.9%
681
1,794
33.6%
79.1%
32.3%
12.0%
Vieux Carre
1.1%
48
4,176
4.5%
75.5%
10.8%
4.8%
Village de L'Est
Total
10.0%
222
2,213
83.2%
54.3%
7.8%
4.3%
Village De L'Est
10.0%
222
2,213
83.2%
54.3%
7.8%
4.3%
Algiers
Total
1.2%
612
51,110
56.6%
48.3%
24.1%
7.3%
Algiers Whitney
7.1%
183
2,564
85.4%
50.0%
29.3%
14.7%
Algiers Naval Station
2.0%
57
2,902
64.1%
49.4%
21.8%
5.1%
Tall Timbers/Brechtel
1.2%
142
12,177
55.0%
64.6%
19.4%
6.3%
Aurora/Walnut Bend/
Huntlee Village
0.9%
144
15,807
31.3%
26.3%
9.9%
4.9%
Algiers Point
0.7%
17
2,381
25.3%
51.7%
17.3%
3.5%
Behrman
0.7%
68
10,430
78.0%
53.1%
33.4%
9.8%
Fischer Project
0.0%
0
2,034
99.2%
88.5%
88.2%
24.7%
McDonogh
0.0%
0
2,815
88.2%
52.9%
48.3%
16.3%
New Aurora
Total
1.1%
62
5,672
68.2%
27.0%
24.8%
10.8%
River Park/Cut Off/
Lower Coast
1.1%
62
5,672
68.2%
27.0%
24.8%
10.8%
City of New Orleans Total
73.0%
354,045
484,674
67.2%
53.5%
27.9%
9.4%
12
At the other end of the spectrum, Mid-City, New Orleans East, Gentilly, the Lower Ninth Ward,
Bywater, and Lakeview all had 90% or more of their residents in damaged areas.
Table 3 also provides information about the social composition of these areas. The Table reveals
great variation within New Orleans. Some largely white neighborhoods of affluent homeowners
were completely flooded, while some relatively poor black neighborhoods were spared.
However there is a general tendency as shown in Table 2 for blacks and poor residents to have
greater odds of being in harm’s way.
Discussions of the racially differential impact of Katrina have often emphasized the Lower Ninth
Ward (where many homes were entirely demolished by the breach in the levee of the Industrial
Canal) and New Orleans East. Most neighborhoods in these planning districts were more than
85% black, and most residences were damaged. A majority of residents of both of these
planning districts were homeowners, though there were clear class distinctions between the two
areas. More than a third of Lower Ninth Ward residents were below the poverty line, and nearly
14% were unemployed. New Orleans East had a considerably larger middle class component,
though it was not among the city’s most affluent sections.
Many of the most segregated neighborhoods with the highest poverty rates are those identified as
“projects,” a reference to the prominence of public housing within their borders. The project
neighborhoods typically had poverty rates in the range of 60-80% of the population,
unemployment is above 20%, they were all predominantly black (with African Americans
accounting for 90% or more of their residents), and 80% or more of residents were renters.
There are six such neighborhoods in New Orleans (though there are concentrations of public
housing or Section 8 housing in other parts of the city). In five of them with a combined 2000
population of over 15,000 persons (Calliope, Iberville, St. Bernard Area, Desire, and Florida) the
entire territory meets this report’s definition of damaged areas. The Fischer Project
neighborhood in Algiers was little damaged. (A seventh project neighborhood, the St. Thomas
Project in the Central City/Garden District, was demolished in 2002, replaced by a Wal-Mart and
new predominantly market-rate condominiums. Therefore the census data from 2000 shown in
Table 3 do not reflect the population composition at the time of Katrina.)
At the other end of the class spectrum are a number of more advantaged neighborhoods with
poverty rates below 10% or unemployment rates below 5%. In the most heavily impacted
planning districts, few neighborhoods meet either criterion. These include the Lake-
Terrace/Lake Oaks neighborhood in Gentilly and the Read Boulevard East neighborhood in New
Orleans East. Most such neighborhoods are in the Lakeview Planning District, which is an area
with a small black population, mostly homeowners, and very low rates of poverty and
unemployment. Here only the Lakeshore/Lake Vista neighborhood, adjacent to Lake
Pontchartrain, was partly spared.
Few residents in the French Quarter, a predominantly white neighborhood with a poverty rate of
about 11% and unemployment below 5%, lived in tracts that were flooded.
Among other neighborhoods with a national reputation for affluence, the Garden District
neighborhood was not flooded and only 40% of the Audubon/University neighborhood (home of
Tulane University and Loyola University) was damaged.
13
Policy implications
The sheer number of people who lived in heavily damaged areas – over 640,000 – is a reminder
of the scale of Katrina’s impact. Because the storm hit large numbers of people of every race
and class, it seems likely that public support for policies to assist these people will also cut across
race and class lines. However there was also a substantial disproportionate impact on African
Americans and people with fewer resources. These disparities stem from within the City of New
Orleans itself, and more specifically from vulnerability to flooding. This is a pattern with deep
roots, and although Katrina caused the most extensive flooding in memory, prior studies by
historians (such as An Unnatural Metropolis by Craig Colten) have demonstrated that both high
ground and public investments in drainage and pumping systems consistently worked to the
advantage of certain neighborhoods in past storms.
There are major variations across the region that are likely to affect the process of recovery.
Damage was extensive on the Mississippi Coast, and the area’s largest single source of
employment – casino gambling – was knocked out of operation. In comparison to New Orleans,
however, the number of people living in areas of moderate or greater damage was small, only
about 50,000. And also in contrast to New Orleans, only a small share of these people were
black and a majority were homeowners. It is difficult to assess the importance of race in
recovery policy in Mississippi, but in a politically conservative state it could make a big
difference that white homeowners constitute the bulk of claimants for state assistance. Further,
these people are easier to serve for several reasons.
1. First, they are identifiable and – because they retain an ownership interest in their
properties – they should prove easier for authorities to contact.
2. Second, since much of the damage wrought by Katrina in this area was by wind and rain
damage, standard homeowner policies offer substantial private sector coverage of
damage losses. For those with uninsured flood damage, the state government currently
expects federal aid to be sufficient to fund payments of $150,000 to individual
homeowners.
3. Third, the low density of housing in this area means that typically even when one’s home
was uninhabitable, there was space for a trailer in the driveway. Since in addition the
loss of electrical power was relatively short-term in Mississippi, and basic public services
could be restored within a reasonable time, homeowners in this region more readily met
the requirements for a FEMA-provided trailer – space and confirmed utility hookups.
In contrast, consider the situation in New Orleans. More than half the persons in damaged areas
were renters, unlikely to be protected in any way by property insurance, and 30% fell below the
poverty line and were therefore unlikely to have personal resources to return to the city. By the
end of 2005, power was still unavailable much of the city, and actual connections to electric
power required residents to present evidence of inspection by a licensed electrician before power
would be restored to an individual home. The utility company (a subsidiary of the Entergy
Corporation) had filed for bankruptcy protection in September.
People who previously lived in public housing seem to have the least chances to return, given
current policy. All public housing in the affected areas has been closed (and special barriers
14
bolted to the doors), and residents have been allotted rental housing assistance in areas where
they have relocated for up to 18 months. Plans for reopening the projects or for constructing new
affordable housing have not become public.
For many of the same reasons that rebuilding will be facilitated on the Mississippi Coast, the
white residents of the City of New Orleans are more likely than black residents to be able to
return to their neighborhoods, even if the neighborhood is reopened. Whites are more likely to
be homeowners (55% compared to 42% among African American households), but more
important, they are much more likely to have the personal resources to reinvest in their homes or
to find a new residence in a difficult housing market. In the pre-Katrina black population, 35%
were below the poverty line and the median household income was only $25,000. Among
whites, only 11% were poor and the median income was more than twice as high – $61,000.
Therefore even among homeowners, blacks are less likely to have the means to rebuild than are
whites.
There is potential for political coalitions that cut across the racial and class divisions that have
helped structure city politics over the decades. Residents of such very different neighborhoods
as Lakeview and the Lower Ninth Ward have a shared interest in short-term assistance programs
such as subsidies for temporary housing outside the city. Up to now few city residents have
qualified for FEMA trailers outside their homes because they do not own the home, or there is
insufficient space for a trailer, or public utilities remain unavailable. As long as issues can be
framed so that demands are oriented toward FEMA (an external target) or toward demands for
services like electricity or schools or police protection that affect all segments of the public, the
appearance of unity can be maintained. The City Council’s repeated stands in favor of
rebuilding all parts of the city – a question on which the Council is unlikely to have the final
word – is a reflection of this temporary unity.
Yet variations across neighborhoods – and across race and class – are likely to support the
emergence of a sense of conflicting interests. In December 2005 conflict took the form of
opposition to proposals to locate FEMA trailers in public spaces within neighborhoods that
sustained less damage. In this case the interests of advantaged neighborhoods (advantaged by
protection from flooding and by having residents in place to express their views) were in conflict
with the interests of absent residents who have no place to return. Not surprisingly the City
Council gave its members veto power over new trailer parks in the areas that they represent.
In the longer term there is likely to be competition between damaged neighborhoods for the
supports that will be necessary for rebuilding. In mid-January 2006, the Bring New Orleans
Back Commission (a policy group formed by city government) released a planning report that
began to address this question. The Commission identified some parts of the city as “immediate
opportunity areas.” These areas are shaded in yellow on the map in Figure 4. Compared to the
damaged areas shown in Figure 2 above, it is clear that the Commission proposes rebuilding in
some areas that were flooded. They include areas along the river near the central business
district, Holy Cross in the Lower Ninth Ward Planning District, and Lakewood/West End and
Lakeshore/Lake Vista bordering Lake Pontchartrain. Generally these are zones where sustained
flooding was under about four feet.
The Commission also recommended the designation of large portions of the city as
“neighborhood planning” areas. Within these areas the Commission recommends a temporary
15
halt on issuance of building permits. Only where there is evidence within the next four months
that residents are committed to returning to each neighborhood in large numbers would
rebuilding be approved.
Figure 4. Recommended areas of rebuilding in the City of New Orleans
Source: Bring Back New Orleans Commission, Action Plan for New Orleans: The New
American City. January 11, 2006. (http://bringneworleansback.org)
The precise criteria underlying these designations have not been announced, and the
Commission’s recommendations have not been formally adopted. This is a temporary solution
to the political problem that few public officials are willing to state openly that some areas will
not be permitted to be resettled. On its face it leaves the decision to local residents. Clearly
some areas are at risk of being closed to reconstruction. In January 2006, the full-time
population of the city has been estimated at only 150,000. The analysis in this report suggests
that if the future city were limited to the population previously living in zones undamaged by
Katrina it would risk losing about 50% of its white residents but more than 80% of its black
population. This is why the continuing question about the hurricane is this: whose city will be
rebuilt?
16
Hoffman Triangle in City Business!
