demographics/ spending analysis

Transcription

CITY OF HARKER HEIGHTS, TEXAS
Date: May 15, 2008
DEMOGRAPHICS/
SPENDING ANALYSIS
Data Source: Geomarket Solutions, Inc.—May 15, 2008
Table of Contents
Welcome…………………………………… p 2
Understanding the Trade Area……………...p 3
- Demographics………………… p 4
- Income…………………………p 4
- Pop. By Age Group…………….p 5
- Race and Ethnicity…………….. p 5
City of Harker Heights……………………... p 6
- Demographics………………… p 6
- Income…………………………p 6
- Pop. By Age Group…………….p 7
- Race and Ethnicity…………….. p 7
- Spending Gap Analysis…………p 8
- Spending Potential Index………p 9
Thematic Maps………………………………p 11
Appendix…………………………………….p 22
Methodology………………………………...p 26
Page 2
WELCOME!
The City of Harker Heights is a dynamic, actively growing community located in what ranks as the fifth fastest growing Metropolitan Statistical Area in the State of Texas (source: POLICOM Crop study). Harker
Heights is located in Bell County, Texas. Heights has excellent access to U.S. Highway 190 which provides
direct access to Interstate 35.
Harker Heights has experienced historic residential growth over the past few years. In a response to these
rooftops, the City is currently undergoing record commercial growth. Businesses such as Super Wal-mart,
Target, Barnes and Noble, Cinemark, Furniture Row, and Cracker Barrell (just to name a few) have recently
located or are building in Heights.
These businesses didn’t arrive in Harker Heights by happenstance. They did their homework on the local
demographics and spending patterns—and they all came to the same conclusion—Harker Heights is the
place to be! Given the overwhelming vote of approval that these new businesses have given Harker Heights,
the question for you is why is your business not operating in Harker Heights. No matter what your trade—
residential building, service industry, or retail– Harker Heights provides fertile ground for your business.
Imagine your business being in the middle of a trade area (30 minute drive time) in excess of 295,000 people
with an average income level of over $58,000.00.
Direct Development recognized this potential and are now completing construction on an 80 acre development that will contain stores that will draw consumers regionally. The numbers all add up—a great development opportunity awaits your company in Harker Heights, Texas.
This document contains detailed demographics and spending analysis data for the City of Harker Heights and
its trade area. Source methodology data for the material provided herein is provided in the methodology
section of this report. This material is just the tip of the iceberg however when it comes to information
about our area. The City’s website, www.ci.harker-heights.tx.us, has an economic development webpage
which provides a wealth of information to prospective businesses. We are also here to serve you and to
answer any questions that you may have. Contact us at (254) 953-5643 or [email protected].
We look forward to having you in Harker Heights soon!
City of Harker Heights, 305 Miller’s Crossing, Harker Heights, TX 76548
254.953.5600
Page 3
UNDERSTANDING THE TRADE AREA (10, 20, AND 30 MINUTE DRIVE TIMES)
While consumers shop and spend within the municipality or jurisdiction in which they live, they also
spend monies in areas outside of this core area. In economic development terms, the area in which a
consumer or group of consumers normally travel, from their residence, to spend monies on goods and
services is called the “trade area”. Trade areas have traditionally been measured and displayed by concentric rings around a core point (i.e. the central point of a city). Modern technology, through geospatial software, allows us to view this concept through a whole new lens. With this technology, concentric rings have been replaced by “drive times”. Drive times indicate the amount of time that it will
take a consumer to reach a specific destination—generally a central point located in a municipality.
Generally, a trade area is now measured in ten, twenty, and thirty minute intervals. Geo-spatial software further allow us to query for very specific information about the population living within these intervals. The following map displays the ten, twenty and thirty minute drive time intervals or trade area
for the City of Harker Heights.
THE TRADE AREA
City of Harker Heights, Texas
Page 4
Within each drive time interval, specific demographic and economic information can be derived. The
following graph displays demographic information specific to the ten, twenty and thirty minute drive
times for the City of Harker Heights.
DEMOGRAPHICS
Population
Households
Families
Average Household Size
Owner-occupied HUs
Renter-occupied HUs
Median Age
10 minute
95,872
36,491
25,827
2.59
18,256
18,234
28.2
20 minute
201,854
67,195
49,857
2.81
32,688
34,507
26.6
30 minute
296,746
104,156
75,489
2.70
59,217
47,939
28.4
The following graph displays household income information specific to the ten, twenty and thirty minute
drive times for the City of Harker Heights.
INCOME
Households by Income
<$15,000
$15,000 - $24,999
$25,000 - $34,999
$35,000 - $49,999
$50,000 - $74,999
$75,000 - $99,999
$100,000 - $149,999
$150,000 - $199,999
$200,000 +
10 minute
Number Percent
3,573
3,563
5,695
7,418
7,845
3,817
3,216
817
544
9.8%
9.8%
15.6%
20.3%
21.5%
10.5%
8.8%
2.2%
1.5%
20 minute
Num- Percent
ber
6,311
9.4%
6,798
10.1%
10,944
16.3%
13,892
20.7%
15,206
22.6%
6,864
10.2%
5,183
7.7%
1,199
1.8%
798
1.2%
30 minute
Number Percent
10,728
10,476
15,693
20,434
23,107
10,983
8,683
2,095
1,959
10.3%
10.1%
15.1%
19.6%
22.2%
10.5%
8.3%
2.0%
1.9%
254.953.5600
Page 5
The following graph displays population by age information specific to the ten, twenty and thirty minute
POPULATION BY AGE GROUP
10 minute
Population by Age
Number Percent
0–4
5 – 14
15 - 19
20 - 24
25 - 34
35 - 44
9,721
13,869
6,532
10,018
20,573
12,543
10.1%
14.5%
6.8%
10.4%
21.5%
13.1%
45 - 54
55 - 64
65 – 74
75 – 84
85 +
10,560
6,273
3,456
1,861
470
11.0%
6.5%
3.6%
1.9%
0.5%
20 minute
30 minute
Number Percent NumPercent
ber
21,250
10.5%
28,747
9.7%
31,702
15.7%
44,641
15.0%
15,740
7.8%
22,356
7.5%
24,272
12.0%
31,285
10.5%
42,256
20.9%
55,691
18.8%
26,645
13.2%
39,341
13.3%
19,232
10,770
5,924
3,104
959
9.5%
5.3%
2.9%
1.5%
0.5%
32,122
20,480
11,634
7,539
2,911
10.8%
6.9%
3.9%
2.5%
1.0%
The following graph displays race and ethnicity information specific to the ten, twenty and thirty minute
RACE AND ETHNICITY
Race and Ethnicity
White Alone
Black Alone
American Indian Alone
Asian Alone
Pacific Islander Alone
Other Race Alone
Two or More Races
Hispanic
10 minute
20 minute
30 minute
Number Percent Number Percent Number Percent
47,526
47.2%
101,846
50.5%
168,958
56.9%
30,536
28.6%
56,129
27.8%
69,221
23.3%
908
0.9%
1,835
0.9%
2,503
0.8%
5,694
5.3%
7,865
3.9%
9,789
3.3%
983
0.9%
1,609
0.8%
1,840
0.6%
11,806
11.1%
21,361
10.6%
30,119
10.1%
6,332
5.9%
11,209
5.6%
14,316
4.8%
23,811
22.3%
42,573
21.1%
59,861
20.2%
Page 6
THE CITY OF HARKER HEIGHTS
The following information is specific to the city boundaries of the City of Harker Heights. The data in
this report is built from the 2000 Census, which itself painted a conservative picture of the population in
Harker Heights. As an example, this report indicates that the population of Harker Heights will reach
26,117 by 2012 while our internal methodologies indicate our current population to be around 27,600.
The data presented in this report should be viewed as a conservative picture of the City. The following
graph displays demographic data for the City. Note: green indicates growth, yellow indicates neutral,
and red indicates decline.
DEMOGRAPHICS
2000
2007
2012
Population
17,729
22,351
26,117
Ann. Growth
Rate
3.16%
Households
6,366
8,234
9,704
3.34%
Families
Average Household Size
Owner-occupied HUs
Renter-occupied HUs
Median Age
4,836
2.77
3,754
2,612
30.3
6,189
2.70
5,024
3,210
30.9
7,244
2.68
5,906
3,798
31.5
3.20%
The graph below presents income data for the City of Harker Heights.
INCOME
Households by Income
<$15,000
$15,000 - $24,999
$25,000 - $34,999
$35,000 - $49,999
$50,000 - $74,999
$75,000 - $99,999
$100,000 - $149,999
$150,000 - $199,999
$200,000 +
2000
Number Percent
691
813
1,069
1,082
1,245
818
465
144
84
10.8%
12.7%
16.7%
16.9%
19.4%
12.8%
7.3%
2.2%
1.3%
2007
Number Percent
574
666
1,021
1,375
1,522
1,119
1,264
393
301
7.0%
8.1%
12.4%
16.7%
18.5%
13.6%
15.3%
4.8%
3.7%
2012
Number Percent
529
542
864
1,502
1,713
1,454
1,787
651
663
5.5%
5.6%
8.9%
15.5%
17.7%
15.0%
18.4%
6.7%
6.8%
254.953.5600
Page 7
The graph below presents population by age data for the City of Harker Heights.
POPULATION BY AGE GROUP
2000
2007
2012
Population by Age
Number Percent Number Percent Number Percent
0–4
1,658
9.4%
2,124
9.5%
2,513
9.6%
5 – 14
2,969
16.7%
3,471
15.5%
4,198
16.1%
15 - 19
1,178
6.6%
1,674
7.5%
1,572
6.0%
20 - 24
1,426
8.0%
1,460
6.5%
2,022
7.7%
25 - 34
3,003
16.9%
3,894
17.4%
4,380
16.8%
35 - 44
3,141
17.7%
3,441
15.4%
3,473
13.3%
45 - 54
2,052
11.6%
2,877
12.9%
3,666
14.0%
55 - 64
1,171
6.6%
1,857
8.3%
2,363
9.0%
65 – 74
717
4.0%
983
4.4%
1,104
4.2%
75 – 84
327
1.8%
448
2.0%
646
2.5%
85 +
87
0.5%
124
0.6%
181
0.7%
The graph below presents race and ethnicity data for the City of Harker Heights.
RACE AND ETHNICITY
Race and Ethnicity
White Alone
Black Alone
American Indian
Alone
Asian Alone
Pacific Islander
Alone
Other Race Alone
Two or More Races
Hispanic
2000
2007
2012
Number
Percent Number Percent Number Percent
12,611
71.1%
15,046
67.3%
16,847
64.5%
2,618
14.8%
3,484
15.6%
4,206
16.1%
136
0.8%
185
0.8%
227
0.9%
634
70
3.6%
0.4%
970
103
4.3%
0.5%
1,286
132
4.9%
0.5%
1,018
641
2,194
5.7%
3.6%
12.4%
1,648
915
3,538
7.4%
4.1%
15.8%
2,257
1,163
4,829
8.6%
4.5%
18.5%
Page 8
SPENDING GAP ANALYSIS FOR HARKER HEIGHTS
The table below and on the following page displays spending information for Harker Heights for a variety of
economic sectors. Given the demographic and economic data for the City, those sectors shown in red represent sales that are taking place outside of Harker Heights (leaking) due to greater demand than supply
within the City. Those areas shown in green represent sectors that have a greater supply than the demand
that exists within the City and are termed as “surplus”.
Retail MarketPlace Profile
Prepared by: GeoMarket Solutions, Inc.
Harker Heights, TX
Summary Demographics
2007 Population
2007 Households
2007 Median Disposable Income
2007 Per Capita Income
Industry Summary
21,947
8,069
$46,858
$27,713
Total Retail Trade and Food & Drink (NAICS 44-45, 722)
Total Retail Trade (NAICS 44-45)
Total Food & Drink (NAICS 722)
Demand
(Retail Potential)
$243,385,922
$211,869,360
$31,516,562
Supply
Retail Gap Leakage/Surplus
(Retail Sales) (Demand - Supply)
Factor
$173,262,831
-$70,123,091
-16.8
-$67,209,299
-18.9
$144,660,061
$28,602,770
-$2,913,792
-4.8
Number of
Businesses
108
75
33
Industry Group
Motor Vehicle & Parts Dealers (NAICS 441)
Automobile Dealers (NAICS 4411)
Other Motor Vehicle Dealers (NAICS 4412)
Auto Parts, Accessories, and Tire Stores (NAICS 4413)
Demand
(Retail Potential)
$59,187,292
$50,189,799
$4,846,255
$4,151,238
Supply
(Retail Sales)
$15,125,554
$10,842,516
$2,132,259
$2,150,779
Retail Gap
-$44,061,738
-$39,347,283
-$2,713,996
-$2,000,459
Leakage/Surplus
Factor
-59.3
-64.5
-38.9
-31.7
Number of
Businesses
9
3
3
3
Furniture & Home Furnishings Stores (NAICS 442)
Furniture Stores (NAICS 4421)
Home Furnishings Stores (NAICS 4422)
$8,167,591
$6,056,337
$2,111,254
$3,804,238
$1,174,034
$2,630,204
-$4,363,353
-$4,882,303
$518,950
-36.4
-67.5
10.9
5
1
4
Electronics & Appliance Stores (NAICS 443/NAICS 4431)
$4,581,940
$2,624,779
-$1,957,161
-27.2
6
Bldg Materials, Garden Equip. & Supply Stores (NAICS 444)
Building Material and Supplies Dealers (NAICS 4441)
Lawn and Garden Equipment and Supplies Stores (NAICS 4442)
$7,517,366
$7,123,210
$394,156
$1,987,240
$1,914,246
$72,994
-$5,530,126
-$5,208,964
-$321,162
-58.2
-57.6
-68.7
5
4
1
$46,400,456
$44,957,019
$748,811
$694,626
$39,748,351
$35,541,054
$581,585
$3,625,712
-$6,652,105
-$9,415,965
-$167,226
$2,931,086
-7.7
-11.7
-12.6
67.8
14
8
2
4
$6,719,935
$969,412
-$5,750,523
-74.8
2
Gasoline Stations (NAICS 447/NAICS 4471)
$29,236,024
$20,280,732
-$8,955,292
-18.1
5
Clothing and Clothing Accessories Stores (NAICS 448)
Clothing Stores (NAICS 4481)
Shoe Stores (NAICS 4482)
Jewelry, Luggage, and Leather Goods Stores (NAICS 4483)
$14,206,540
$11,595,123
$1,612,690
$998,727
$1,468,204
$998,013
$314,968
$155,223
-$12,738,336
-$10,597,110
-$1,297,722
-$843,504
-81.3
-84.1
-67.3
-73.1
5
3
1
1
$3,854,291
$2,413,916
$1,440,375
$591,760
$505,598
$86,162
-$3,262,531
-$1,908,318
-$1,354,213
-73.4
-65.4
-88.7
3
2
1
Food & Beverage Stores (NAICS 445)
Grocery Stores (NAICS 4451)
Specialty Food Stores (NAICS 4452)
Beer, Wine, and Liquor Stores (NAICS 4453)
Health & Personal Care Stores (NAICS 446/NAICS 4461)
Sporting Goods, Hobby, Book, and Music Stores (NAICS 451)
Sporting Goods/Hobby/Musical Instrument Stores (NAICS 4511)
Book, Periodical, and Music Stores (NAICS 4512)
Page 9
SPENDING ANALYSIS—CITY OF HARKER HEIGHTS (cont)
Retail MarketPlace Profile
Prepared by: GeoMarket Solutions, Inc.
Harker Heights, TX
Demand
(Retail Potential)
$24,217,427
$14,876,479
$9,340,948
Supply
(Retail Sales)
$54,922,065
$50,283,286
$4,638,779
Retail Gap
$30,704,638
$35,406,807
-$4,702,169
Leakage/Surplus
Factor
38.8
54.3
-33.6
Number of
Businesses
2
1
1
Miscellaneous Store Retailers (NAICS453)
Florists (NAICS4531)
Office Supplies, Stationery, and Gift Stores (NAICS4532)
Used Merchandise Stores (NAICS4533)
Other Miscellaneous Store Retailers (NAICS4539)
$4,209,273
$442,314
$2,385,859
$501,865
$879,235
$2,985,408
$174,652
$460,560
$319,267
$2,030,929
-$1,223,865
-$267,662
-$1,925,299
-$182,598
$1,151,694
-17.0
-43.4
-67.6
-22.2
39.6
18
2
5
3
8
Nonstore Retailers (NAICS 454)
Electronic Shopping and Mail-Order Houses (NAICS4541)
Vending Machine Operators (NAICS 4542)
Direct Selling Establishments (NAICS 4543)
$3,571,225
$2,052,498
$495,338
$1,023,389
$152,318
$0
$0
$152,318
-$3,418,907
-$2,052,498
-$495,338
-$871,071
-91.8
-100.0
-100.0
-74.1
1
0
0
1
$31,516,562
$12,291,234
$16,712,414
$1,353,949
$1,158,965
$28,602,770
$19,053,392
$6,748,509
$0
$2,800,869
-$2,913,792
$6,762,158
-$9,963,905
-$1,353,949
$1,641,904
-4.8
21.6
-42.5
-100.0
41.5
33
2
24
0
7
Industry Group
General Merchandise Stores (NAICS452)
Department Stores Excluding Leased Depts. (NAICS 4521)
Other General Merchandise Stores (NAICS4529)
Food Services & Drinking Places (NAICS 722)
Full-Service Restaurants (NAICS7221)
Limited-Service Eating Places (NAICS 7222)
Special Food Services (NAICS 7223)
Drinking Places - Alcoholic Beverages (NAICS7224)
Page 10
SPENDING POTENTIAL INDEX—CITY OF HARKER HEIGHTS
As indicated in the table below, when compared to national averages for spending, the Harker Heights’
citizen is slightly more affluent than his national counterpart. Any Spending Potential Index greater than
100 indicates that spending exceeds the national average.
Harker Heights, Texas
Spending
Average
Potential
Amount
Total
Index
Spent
Apparel and Services
Men's
Women's
Children's
Footwear
Watches & Jewelry
Apparel Products and Services
92
95
85
101
80
107
111
$2,518.98
$469.28
$820.35
$445.39
$408.74
$210.67
$164.55
$20,673,266
$3,851,358
$6,732,599
$3,655,324
$3,354,511
$1,728,996
$1,350,477
Computer
Computers and Hardware for Home Use
Software and Accessories for Home Use
105
111
$230.41
$33.18
$1,890,958
$272,316
Entertainment & Recreation
Fees and Admissions
Membership Fees for Clubs
Fees for Participant Sports, excl. Trips
Admission to Movie/Theatre/Opera/Ballet
Admission to Sporting Events, excl. Trips
Fees for Recreational Lessons
TV/Video/Sound Equipment
Community Antenna or Cable Television
Color Televisions
VCRs, Video Cameras, and DVD Players
Video Cassettes and DVDs
Video Game Hardware and Software
Satellite Dishes
Rental of Video Cassettes and DVDs
Sound Equipment
Rental and Repair of TV/Sound Equipment
Pets
Toys and Games
Recreational Vehicles and Fees
Sports/Recreation/Exercise Equipment
Photo Equipment and Supplies
Reading
103
105
102
107
103
111
107
101
98
108
105
105
107
106
110
104
95
102
105
108
97
106
97
$3,526.95
$640.97
$162.49
$120.84
$154.72
$63.63
$139.28
$1,180.64
$654.39
$149.18
$40.60
$64.14
$35.19
$1.63
$65.61
$164.41
$5.48
$448.22
$191.29
$491.22
$223.06
$145.05
$206.50
$28,945,694
$5,260,439
$1,333,546
$991,743
$1,269,820
$522,219
$1,143,111
$9,689,472
$5,370,546
$1,224,361
$333,237
$526,379
$288,770
$13,363
$538,471
$1,349,348
$44,997
$3,678,511
$1,569,941
$4,031,441
$1,830,692
$1,190,420
$1,694,777
Food
Food at Home
Bakery and Cereal Products
Meats, Poultry, Fish, and Eggs
Dairy Products
Fruits and Vegetables
Snacks and Other Food at Home
100
$8,432.06
$69,201,951
99
99
99
99
99
101
$4,992.68
$710.90
$1,296.81
$542.78
$866.11
$1,576.07
$40,974,904
$5,834,393
$10,642,949
$4,454,601
$7,108,170
$12,934,791
THEMATIC
MAPS
Page 12
Page 13
Page 14
Page 15
Page 16
Page 17
Page 18
Page 19
Page 20
Page 21
APPENDIX
Page 23
Businesses – Greater Then $2,500,000 in Sales
Company
SIC
NAICS_EXT
SALES_VOL
NUMBER_EMP
WAL-MART SUPERCENTER
H-E-B FOODS
722101
541105
54192110
44511003
$
$
50,000,000
37,050,000
300
150
HEIGHTS HOME HEALTH
808201
62161001
$
24,300,000
450
WARD GROUP
641112
52421001
$
18,620,000
95
L-3 COMMUNICATIONS ILEX SYSTS
179977
23899018
$
16,800,000
100
AYCOCK CONSTRUCTION INC
152103
23611505
$
12,728,000
37
INTERSTATE MERCHANT SVC L
504428
42342018
$
12,210,000
30
BELL COUNTY HARLEY-DAVIDSON
557106
44122108
$
12,100,000
25
ASPEN AIR INC
171117
23822002
$
11,656,000
62
FAMILY CARE HOME HEALTH
808201
62161001
$
11,340,000
210
R K BASS ELECTRIC INC
173101
23821007
$
9,840,000
60
COOPER & BRIGHT PLUMBING
171105
23822025
$
9,400,000
50
INDIAN OAKS
805101
62311016
$
8,750,000
125
TIME WARNER CABLE
484101
51521001
$
6,895,000
-
WASTE MANAGEMENT INC
495302
56211901
$
6,681,000
-
RELOCATION DEPARTMENT
653118
53121003
$
6,566,000
49
CENTEX HOMES
152112
23611506
$
6,192,000
18
TEAM WHEATLEY
653118
53121003
$
5,762,000
43
CYBER-NOSTICS
508205
42381037
$
4,890,000
-
HEIGHTS LUMBER & SUPPLY INC
525104
44413005
$
4,425,000
25
RE/MAX PLATINUM
653118
53121003
$
4,020,000
30
FIRST COMMUNITY MORTGAGE
616201
52229202
$
3,980,000
20
CENTEX WASTE MANAGEMENT
495302
56211901
$
3,930,000
30
FIRST NATIONAL BANK
602101
52211002
$
3,913,000
13
KWICK STOP
541103
44512001
$
3,808,000
17
TEXAS BOAT WORLD
555104
44122204
$
3,768,000
8
BEDROOM EXPRESSIONS
571216
44211012
$
3,672,000
12
ROSEWOOD NURSING & REHAB CTR
805101
62311016
$
3,500,000
50
MC DOWELL ENTERPRISES INC
171105
23822025
$
3,384,000
18
ROYCE ELECTRIC INC
173101
23821007
$
3,280,000
20
DIAMOND SHAMROCK
554101
44719005
$
3,248,000
7
FOLKERSON COMMUNICATIONS LTD
599904
44311223
$
3,220,000
10
SKIN DEEP-A CLINICAL SKIN CARE
801104
62149301
$
3,050,000
10
ABSOLUTE HEALTH PROFESSIONALS
808201
62161001
$
2,916,000
54
MICKEY'S CONVENIENCE FOOD STRS
541103
44512001
$
2,912,000
13
ACE FENCE CO
503903
42339013
$
2,880,000
8
OMALLY'S DECOR
519904
42499054
$
2,872,000
-
KIIZ
483201
51511203
$
2,816,000
16
CINGULAR WIRELESS
481207
51721201
$
2,805,000
-
MURPHY OIL USA INC
554101
44719005
$
2,784,000
6
Page 24
Businesses – Greater Then 25 Employees
Company
HEIGHTS HOME HEALTH
WAL-MART SUPERCENTER
HARKER HEIGHTS HIGH SCHOOL
FAMILY CARE HOME HEALTH
H-E-B FOODS
INDIAN OAKS
L-3 COMMUNICATIONS ILEX SYSTS
EASTERN HILLS MIDDLE SCHOOL
WARD GROUP
HARKER HEIGHTS ELEMENTARY SCHL
UNION GROVE MIDDLE SCHOOL
MOUNTAIN VIEW ELEMENTARY SCHL
ASPEN AIR INC
R K BASS ELECTRIC INC
MC DONALD'S
GOODWILL INDUSTRIES
ABSOLUTE HEALTH PROFESSIONALS
COOPER & BRIGHT PLUMBING
ROSEWOOD NURSING & REHAB CTR
RELOCATION DEPARTMENT
TEAM WHEATLEY
AYCOCK CONSTRUCTION INC
HARKER HEIGHTS FIRE DEPT
INTERSTATE MERCHANT SVC L
RE/MAX PLATINUM
CENTEX WASTE MANAGEMENT
WILD COUNTRY
LEARNING ZONE CHILD CARE CTR
PIZZA HUT
HARKER HEIGHTS POLICE DEPT
PAPA JOHN'S PIZZA
BELL COUNTY HARLEY-DAVIDSON
HEIGHTS LUMBER & SUPPLY INC
SIC
808201
722101
821103
808201
541105
805101
179977
821103
641112
821103
821103
821103
171117
173101
581208
832218
808201
171105
805101
653118
653118
152103
922404
504428
653118
495302
581304
835101
581208
922104
581222
557106
525104
NAICS_EXT
62161001
54192110
61111007
62161001
44511003
62311016
23899018
61111007
52421001
61111007
61111007
61111007
23822002
23821007
72221105
62419012
62161001
23822025
62311016
53121003
53121003
23611505
92216003
42342018
53121003
56211901
72241006
62441003
72221105
92212003
72211016
44122108
44413005
SALES_VOL
$ 24,300,000
$ 50,000,000
$
$ 11,340,000
$ 37,050,000
$
8,750,000
$ 16,800,000
$
$ 18,620,000
$
$
$
$ 11,656,000
$
9,840,000
$
2,400,000
$
$
2,916,000
$
9,400,000
$
3,500,000
$
6,566,000
$
5,762,000
$ 12,728,000
$
$ 12,210,000
$
4,020,000
$
3,930,000
$
2,010,000
$
1,260,000
$
1,200,000
$
$
1,456,000
$ 12,100,000
$
4,425,000
NUMBER_EMP
450
300
250
210
150
125
100
100
95
89
80
75
62
60
60
60
54
50
50
49
43
37
31
30
30
30
30
30
30
28
26
25
25
Page 25
Page 26
METHODOLOGY
An ESRI ® White Paper • June 2007
ESRI® Demographic Update
Methodology: 2007/2012
ESRI 380 New York St., Redlands, CA 92373-8100 USA
TEL 909-793-2853 • FAX 909-793-5953 • E-MAIL [email protected] • WEB www.esri.com
Copyright © 2007 ESRI
All rights reserved.
Printed in the United States of America.
The information contained in this document is the exclusive property of ESRI. This work is protected under United States
copyright law and other international copyright treaties and conventions. No part of this work may be reproduced or
transmitted in any form or by any means, electronic or mechanical, including photocopying and recording, or by any
information storage or retrieval system, except as expressly permitted in writing by ESRI. All requests should be sent to
Attention: Contracts and Legal Services Manager, ESRI, 380 New York Street, Redlands, CA 92373-8100 USA.
The information contained in this document is subject to change without notice.
ESRI, the ESRI globe logo, ArcGIS, www.esri.com, and @esri.com are trademarks, registered trademarks, or service marks
of ESRI in the United States, the European Community, or certain other jurisdictions. Other companies and products
mentioned herein may be trademarks or registered trademarks of their respective trademark owners.
J-9663
ESRI Demographic Update
An ESRI White Paper
Contents
Page
What's Hot............................................................................................. 1
What's New in 2007........................................................................ 5
Geography Changes ........................................................................ 5
2007 Demographic Update Methodology.............................................
County Totals..................................................................................
Block Group Totals.........................................................................
Population and Household Characteristics .....................................
Housing ...........................................................................................
6
6
6
9
11
Labor Force Update Methodology........................................................
Data Sources ...................................................................................
Methods...........................................................................................
Concepts..........................................................................................
Dissimilarities in Sources of Labor Force Information ..................
Impact of the Gulf Coast Region on the U.S. Labor Force.............
12
12
13
13
14
14
Income Update Methodology ...............................................................
Data Sources ...................................................................................
Income Methods..............................................................................
Disposable Income..........................................................................
Net Worth........................................................................................
Use of Projections ...........................................................................
15
15
16
17
18
19
ZIP Code Update Methodology ............................................................ 19
Data Source for Boundaries ............................................................ 19
Comparisons over Time.................................................................. 20
Conclusion ............................................................................................ 20
ESRI White Paper
i
J-9663
ESRI Demographic Update
What's Hot
If you want to follow the latest trends, find the trendsetters. To "follow the
money," you may opt for areas with either the highest household (HH)
income or the most rapidly growing income. To find both high income and
rapid growth in income, check out the areas in table 1:
Table 1
Metropolitan Statistical Area
2007 Median
Household Income
2000-2007 Change:
Median HH Income
Bridgeport-Stamford-Norwalk, CT
$84,800
31%
Boston-Cambridge-Quincy, MA-NH
$73,000
34%
Minneapolis-St. Paul-Bloomington, MN-WI
$71,400
32%
Washington-Arlington-Alexandria, DCVA-MD-WV
$80,100
27%
Denver-Aurora, CO
$67,500
31%
Each of these metropolitan areas, except for Bridgeport-Stamford-Norwalk, CT, has more
than three million residents.1 If you prefer high income in a smaller residential area, look
at a micropolitan statistical area such as Los Alamos, New Mexico, that has a median
income of $101,200 in 2007, up 30 percent since 2000, or a high-income county such as
those in table 2:
Table 2
County/Metropolitan
Statistical Area
Douglas County, CO
Denver-Aurora, CO
Loudoun County, VA
Washington-ArlingtonAlexandria, DC-VA-MD-WV
Fairfax County, VA
Washington-ArlingtonAlexandria, DC-VA-MD-WV
Hunterdon County, NJ
New York-Northern New JerseyLong Island, NY-NJ-PA
Somerset County, NJ
New York-Northern New JerseyLong Island, NY-NJ-PA
1
2007 Median
Household Income
2000-2007 Change: Median
HH Income
$111,300
36%
$108,100
35%
$104,500
30%
$103,800
31%
$100,300
31%
Bridgeport-Stamford-Norwalk, CT, metropolitan statistical area has a 2007 population of 918,315.
ESRI White Paper
ESRI Demographic Update Methodology: 2007/2012
J-9663
To "keep up with the Joneses," you want to find the areas with the best housing. Using
median home value as the yardstick, a median of $400,000 or more and double-digit
appreciation every year since 2000 qualifies the country's top metropolitan areas (found
primarily in California or Hawaii). Geographic exceptions to this pattern are the
metropolitan statistical areas of New York-Northern New Jersey-Long Island, NY-NJPA, and Washington-Arlington-Alexandria, DC-VA-MD-WV.
Requirements at the county level are more stringent—a median home value of $600,000
or more is necessary to rank among the best county markets. Double-digit appreciation in
home value, 2000–2007, is typical, but not universal. The top counties are also common
to California and Hawaii; notable exceptions are counties that specialize in seasonal
attractions such as Nantucket and Dukes (Martha's Vineyard) in Massachusetts and Pitkin
in Colorado, better known as the site of Aspen.
Not interested in tracking the money or keeping up with the Joneses? You can simply "go
with the flow" by following the population movement to the South and West. There are
few changes in the list of hot spots in 2007. The fastest-growing metropolitan areas are in
Florida and the Southwest. The fastest-growing counties still represent the suburban
sprawl of growing metropolitan areas—Flagler County, Florida; Loudoun County,
Virginia (Washington, D.C., metropolitan area); Kendall County, Illinois (Chicago
metropolitan area); Rockwall County, Texas (Dallas metropolitan area); Douglas County,
Colorado (Denver metropolitan area); plus several counties in the Atlanta, Georgia,
metropolitan area.
Sustained population growth in two micropolitan areas—Palm Coast, Florida (Flagler
County), and Lake Havasu City-Kingman, Arizona (Mohave County)—resulted in their
revised classification to metropolitan areas this year. However, without the economic
draw of a metropolitan central city, most micropolitan areas do not grow as quickly as
metropolitan centers. Counties that are neither metropolitan nor micropolitan are least
likely to experience rapid population growth and are most likely to lose population.
Table 3
Metropolitan
Counties
Micropolitan
Counties
Nonmetropolitan
Counties
High Growth: ≥ 2%
25%
6%
3%
Moderate Growth: 1–2%
29%
19%
10%
Minimal Growth: < 1%
36%
48%
44%
Loss
10%
27%
43%
Total Counties
1,092
692
1,357
Average Annual Rate
1.4%
0.6%
0.2%
2000–2007 Annual Rate
Small, subcounty areas are the exception to this pattern. Because they represent a fraction
of a county's population, subcounty areas can experience trends that are diametrically
different from population change at the county level. For example, the population of
June 2007
2
J-9663
Washington, D.C., has increased by only 3.4 percent since 2000, but Washington ZIP
Code 20004 has grown by 80 percent. Although ZIP Codes can pose a challenge to time
series analysis due to administrative changes by the U.S. Postal Service in their
territories, ZIP Codes remain popular geographic areas for subcounty analysis. The
average annual rate of change in populated ZIP Codes from 2000 through 2007 is
1.2 percent—comparable to metropolitan counties.
ESRI® 2007 demographic data updates reflect current events such as rising inflation or
interest rates and regional distinctions like the availability of jobs or affordable housing.
The housing market remains central in the discussion of current trends. In 2006, the
housing market was slowing after more than a decade of upsurge. The appreciation of
home value was decelerating in most markets, and sales of existing homes were
declining. Today's U.S. housing market faces various challenges. Sixty-eight percent of
U.S. householders own their homes in 2007, but this represents a leveling in the rate of
homeownership, which started to climb in the mid-1990s. Inventories of new and existing
homes have increased; sellers are lowering their asking prices, receiving fewer offers, and
experiencing longer selling periods. Home builders are reducing sales forecasts, cutting
back on staff, offering buyer incentives, and experiencing rising cancellation rates. Home
prices are actually declining in some markets, and new home sales are down. Existing
home sales dropped 8.4 percent in March 2007, which is the largest monthly drop since
2
1989, according to the National Association of Realtors.
What's happening? Demographics played a key role in shaping the housing market boom
that began in the mid-1990s. During that period, the baby boomers, born between 1946
and 1964, reached their peak earning years, with many in their 30s and 40s. At that stage
in their lives, they became first-time homebuyers or traded up to meet their changing
housing needs. This contributed to a surge in housing demand that boosted the housing
market.3 Demographic change, including immigration and population growth, was not the
only factor behind increasing demand. The transitory surge in demand can also be
attributed to the Federal Reserve, mortgage lenders, and Wall Street investment firms.
The Federal Reserve's easy money stance contributed to the demand growth. In early
2001, the Federal Reserve began its series of rate cuts when it reduced its short-term
target from 6.5 to 6 percent. The cuts continued through mid-2003 when rates plummeted
to 1 percent. The Federal Reserve finally reversed its policy in mid-2004. Surprisingly, a
Federal Reserve district bank president conceded that bad preliminary data on inflation
4
led the bank's committee members to reduce short-term rates too low for far too long. As
mortgage rates fell to historic lows, lenders introduced creative mortgages and loosened
credit lending standards, which benefited borrowers in the subprime market but fueled
speculative behavior. Finally, Wall Street investment firms saw an opportunity for high
returns by extending credit to subprime mortgage lenders to distribute to borrowers with
the greatest credit risk. All these factors unleashed the demand for housing.
2
Haggerty, James R., "House Prices Slide as Property Glut Grows," The Wall Street Journal, April 25, 2007,
D1, http://online.wsj.com/article/SB117745915366081228.html.
3
Dowell Myers and Lonnie Vidaurri's article, "Real Demographics of Housing Demand in the United States,"
The Lusk Review, 1996, pages 55–61, provides a detailed discussion on how the size of an age cohort affects
housing demand.
4
Ip, Greg, "Fed Official Says Bad Data Helped Fuel Rate Cuts, Housing Speculation," The Wall Street Journal,
November 3, 2006, A6.
ESRI White Paper
3
J-9663
Although more householders could afford to buy homes, mortgage payments on
adjustable rate loans were poised to increase. When monetary policy reversed course and
the Federal Reserve began to raise short-term rates, teaser rates on risky loans increased
and strained many household budgets. Borrowers who were unable to refinance into more
conventional mortgages were left with few alternatives. Investment properties flooded the
resale market, placing additional downward pressure on prices and eroding equity. The
result has been rising delinquencies, defaults, and foreclosures. The Federal Reserve
believed that defaults would not spread to the broader segment of borrowers. But it is
becoming evident that borrowers who fall in the middle between subprime and prime
mortgages, called Alt-A, are starting to succumb to the same problems as subprime
borrowers.
That explains how we got to this point. Now what? Demographic change continues to
affect demand. Both the baby boomers and Generation Y can have a significant impact
on the housing market in the next decade. As baby boomers enter their retirement years,
their housing needs will change. Some will downsize to smaller homes with less
maintenance, while others will seek second or vacation homes. The Generation Y
cohorts, born between 1978 and 1997, are now forming new households or entering
adulthood. In 2007, the median income for households headed by 25–29-year-olds is
$46,600; the median income for householders younger than 25 years old is $29,800.
While still in their 20s and with home prices still high, Generation Y is an important
factor in the rental housing market. But as they enter their 30s in the next few years, they
will be potential homebuyers.
The demand for affordable housing has not changed. In fact, housing affordability
remains a key issue, particularly in large metropolitan areas such as Los Angeles or San
Francisco where median home values are more than nine times median household
incomes. However, subprime mortgages with adjustable interest rates were not the best
response. If the mortgage industry, or Congress, tightens credit lending practices, the
market will certainly experience a drop in demand that can prolong or worsen the slump
in the short term. In the long run, reducing the volume of subprime mortgages can also
restore the market by bringing supply and demand back to a sustainable balance.
Since a market correction appears inevitable, what is the effect on the economy? Some
analysts estimated the impact the housing contraction had on gross domestic product
5
(GDP) growth was approximately one percentage point during the last half of 2006.
Other pressures such as recent productivity trends may indicate higher inflation in the
future. Labor productivity, which measures the output per hour worked, has begun to
weaken, which can elevate prices for goods and services. GDP is expected to grow, but at
a slower pace than in 2006, and employment continues to expand. A strong labor force
may serve as one explanation for the lower productivity rates. As the business cycle
expansion matures and technological gains dissipate, firms may find it necessary to
increase labor to expand output.
Inflation, as measured by the core personal consumption expenditure index, has risen
above the Federal Reserve's upper range. Meanwhile, it has held short-term interest rates
steady at 5.25 percent since mid-2006 and assumed a cautious stance on future policy
5
Wheeler, David C., "Housing Slump Could Lean Heavily on Economy," The Regional Economist, April 2007.
June 2007
4
J-9663
adjustments. For now, the Federal Reserve believes the overall economy will exhibit
"moderate growth" in 2007.
What's New in 2007
ESRI's 2007/2012 demographic data updates include more new developments uncovered
by the innovative combination of spatial and demographic analysis that ESRI introduced
in 2006. Collectively known as Address-Based Allocation, the techniques are designed to
capture change in the distribution of household population. To date, these techniques
have uncovered not only changes in the settlement of established neighborhoods but also
new housing in previously unpopulated areas. This year, ESRI could also apply these
new methods to the Gulf Coast communities impacted by the 2005 hurricane season.
Gauging the effects of Hurricanes Katrina, Rita, and Wilma was complicated in 2006 by
the lack of information from ESRI's usual data sources. Because the situation was too
"fluid" in the impacted areas, databases that are normally updated continually were not
revised right away to incorporate the loss of population and businesses. Measuring the
demographic and economic consequences in 2006 proved to be a singular challenge that
6
required the development of new methods. Building from this work and incorporating
data released later in 2006, ESRI was able to integrate past and current changes in the
distribution and characteristics of the population along the Gulf Coast.
Geography Changes
Change is inevitable with any geographic area—political or statistical. Identifying the
changes in the areas for which data is tabulated and reported is critical to the analysis of
trends. In the past year, there have been minor changes to metropolitan areas by the
Office of Management and Budget, boundary revisions for Designated Market Areas
(DMAs) by Nielsen Media Research, and changes to the boundaries of congressional
districts in Georgia and Texas in addition to the usual adjustment of ZIP Codes by the
U.S. Postal Service.
Metropolitan changes include the latest revisions to Core Based Statistical Areas
(CBSAs), released in January 2007. Changes include one new micropolitan statistical
area: Fredericksburg, Texas (Gillespie County), and two revisions from micropolitan to
metropolitan areas: Palm Coast, Florida (Flagler County), and Lake Havasu CityKingman, Arizona (Mohave County). There are now 939 CBSAs, 363 metropolitan
areas, and 576 micropolitan areas.
DMAs represent the 2006–2007 markets defined by Nielsen Media Research. Most
DMAs correspond to whole counties, but there are a few exceptions where counties are
split into different DMAs. There are no code or name changes to DMAs; however,
several counties were assigned to different DMAs. Data for congressional districts was
updated to represent the 110th Congress, including boundary revisions to congressional
districts in Georgia and Texas. Finally, ZIP Codes, which are defined by the U.S. Postal
Service, are updated to reflect its November 2006 inventory.
ESRI presents its 2007/2012 demographic forecasts, including population, age by sex,
race by Hispanic origin, age by sex by race and by Hispanic origin, households and
families, housing by occupancy, tenure and home value, labor force and employment by
industry and occupation, and income—including household and family income
6
The data sources, assumptions, and methods used to estimate the change along the Gulf Coast in 2006 are
provided in a separate ESRI white paper, Gulf Coast Update Methodology, which is available at
http://www.esri.com/data/community_data/demographic/methodology.html.
ESRI White Paper
5
J-9663
distributions, household income by age of householder, and per capita income.7 Updates
of household income are also extended to provide after-tax (disposable) income and a
measure of household wealth: net worth. Changes in the update base from the Census
Bureau's Count Question Resolution (CQR) revisions, updated boundaries, and
improvements to forecasting techniques may preclude comparison to 2006 or earlier
updates.
2007 Demographic
Update Methodology
County Totals
Forecasts are prepared initially for counties and block groups (BGs). From the county
database, forecasts are aggregated to CBSAs, states, or higher levels. From the block
group database, forecasts can be retrieved for census tracts; places; county subdivisions;
ZIP Codes; Congressional Districts for the 110th Congress; DMAs; or any user-defined
site, circle, or polygon.
The change in total population is a function of changes in household population and the
population in group quarters (GQ), which are subject to different trends. The addition of a
prison, for example, produces a sudden increase in the group quarters population that is
unlikely to yield an attendant change in the household population or the projected
population growth of a county. A military base closing effects an immediate decrease in
the household population with the reduction not only of military personnel but also their
families and civilian personnel; however, this drop is unlikely to continue. The disparity
of trends in household versus group quarters population is accommodated by separate
projections. The group quarters population is the Census 2000 count of group quarters,
with CQR revisions and updates culled from a variety of federal, state, and local sources.
Forecasting change in the size and distribution of the household population begins at the
county level with several sources of data. ESRI begins with a time series from the U.S.
8
Census Bureau that includes county estimates through 2005. Because testing has
revealed improvement in accuracy by using a variety of different sources to track county
population trends, ESRI also employs a time series of building permits and housing
starts, plus residential postal delivery counts. Finally, local data sources that tested well
against Census 2000 are reviewed.
Block Group Totals
Measuring the change in population or households at the county level is facilitated by the
array of data reported for counties. Unfortunately, there is no current data reported
specifically for block groups. Past trends can be calculated from previous census counts,
but there is nothing current. To measure current population change by block group, ESRI
models the change in households from three primary sources—InfoBase database from
Acxiom Corporation, residential delivery statistics from the U.S. Postal Service, and
residential construction data from Hanley Wood Market Intelligence—in addition to
several ancillary sources.
The U.S. Postal Service (USPS) publishes monthly counts of residential deliveries for
every U.S. postal carrier route. This represents the most comprehensive and current
information available for small, subcounty geographic areas. USPS establishes carrier
routes to enable efficient mail delivery. Carrier routes are a fluid geographic construct
that is redefined continually to incorporate real changes in the housing inventory and
7
8
Forecasts represent the midyear population on July 1, unless otherwise specified.
U.S. Bureau of the Census, Population Division, Table CO-EST2005-ALLDATA.
June 2007
6
J-9663
occupancy plus administrative changes in staffing and budgets of local post offices.
These frequent changes in the carrier routes are not the only difficulty.
Converting delivery statistics from postal carrier routes to census block groups is a
complex challenge. Carrier routes are defined to deliver the mail, while block groups are
constructed to collect and report census data. Comparing two different areas that are
defined for wholly different purposes provides a significant conversion issue. Carrier
routes commonly overlap multiple block groups. In many cases, a carrier route
encompasses disjointed areas that can be distant from each other, but block groups are
rarely divided into multiple polygons. These overlaps require an effective method of
allocating the postal delivery counts across multiple block groups.
One way to distribute delivery statistics among component block groups is to create a
correspondence using boundary files. Changes in postal carrier routes can be tracked
through quarterly updates of carrier route boundaries, and delivery statistics can be
assigned to block groups with 2000 census block data. Another way also employs
boundary files but assumes a uniform distribution of households within the area. Using
standard geodemographic tools, it is possible to estimate the change in households from
carrier route delivery statistics and to apply that change to any block group(s) in the area.
But the estimated change is simply being redistributed from one summary area to
another.
ESRI has developed another way to link a carrier route to the correct block groups—
using the actual locations of mail deliveries. Its proprietary Address-Based Allocation
(ABA) solves the complex challenge of converting delivery counts from carrier routes to
block groups. This allocation method uses the addresses from Acxiom's InfoBase
household database. Addresses in the database are geocoded with carrier route and block
group codes, using an enhanced geocoding technique and database, and serve as the
foundation for the conversion. This approach is unbounded by geographic borders or
arbitrary assumptions about the distribution of households or postal deliveries.
ABA results have been tested extensively. The tests include benchmarking against the
2000 Census. Manual reviews confirm the capability of the method to identify areas with
high growth. The ABA allocation method reveals sprawls and new developments across
the country since Census 2000. Assessments based on other data sources verify the
efficacy and precision of ABA. For the small portion of block groups where addresses are
not available from the InfoBase database, delivery statistics are allocated from a
correspondence file. The correspondence between census block groups and postal carrier
routes is developed using quarterly updated data from Tele Atlas.
ESRI White Paper
7
J-9663
The effectiveness of the ABA methodology relies on the precision of block group
assignment to InfoBase addresses. ESRI improved the geocoding accuracy of the
InfoBase file by applying ArcGIS® 9.2 with the Dynamap/Address Points database from
Tele Atlas, which provides coordinates that are accurate to the building. It offers a new
development in large-scale geographic databases where addresses are represented as
points rather than approximations estimated from address ranges or street segments. The
database currently covers the most densely populated areas in the United States, with
continuously increasing geographic coverage. Addresses that fall outside the coverage
were geocoded with the conventional approach, based on address ranges.
Post office delivery counts or address counts provide less coverage in rural areas.
Sparsely populated areas tend to have post office box ZIP Codes because there are few
rural addressing systems and little comparability to urban, street delivery. The same
problems characterize rural addresses in the InfoBase database. To track new housing
developments, especially in previously unpopulated areas, ESRI licensed a new data
source from Hanley Wood Market Intelligence—new and planned residential
construction in the top 75 U.S. housing markets including 7 new markets added in 2006.
The new residential construction database from Hanley Wood Market Intelligence adds a
unique component to ESRI's strategy for producing accurate demographic forecasts. This
database identifies individual construction projects—for instance, a complex of singlefamily homes or townhomes or a condominium building—with their exact locations by
latitude and longitude. It also pinpoints conversions of apartments into condominiums.
The construction information includes

Total number of units planned
Inventory of units under construction, sold, and/or closed
Type of housing—Detached homes, townhomes, condominiums, and so forth
Target markets—Families, seniors, empty nesters, and so forth
The use of this type of information in demographic forecasts has traditionally been
confined to small-scale implementation such as producing forecasts for a specific county.
ESRI partners with Hanley Wood Market Intelligence to introduce this information in a
large-scale forecasting effort. The new construction database complements and
corroborates the postal delivery statistics. More important, it tabulates planned
construction to be completed in upcoming years. This information is incorporated in
ESRI's five-year forecasts. Tracking residential development since 2000 with enhanced
demographic and spatial analysis tools provides better information for the five-year
forecasts than past trends.
A revised housing unit methodology applies the change in households estimated from
address counts, delivery counts, and new housing construction to update household
population by block group. The best techniques are derived from a combination of
models and data sources. Discrepant trends are checked extensively against independent
sources. Finally, totals for block groups are controlled to the county totals. The
integration of demographic and spatial analysis and the addition of the Hanley Wood data
about residential development represent a break from past methods and preclude
comparisons to earlier updates.
June 2007
8
J-9663
Population and
Household
Characteristics
ESRI's population and household characteristics include the population by sex and age,
race and Hispanic origin, sex by age by race and Hispanic origin, and household type.
Population by sex and age include estimates by five-year age groups and by single years
from less than 1 year to 84 years.
The population by age and sex is projected via a cohort survival model that calculates the
components of population change separately, by age and sex. Applying survival rates
specific to the cohort carries the 2000 population forward. Changes in the population by
age and sex diverge at the household level. For example, an area that is losing population
can age more rapidly with the loss of population in prime migrant ages, 20–34 years—
unless there is a college nearby. An influx of college students can offset the loss of
youthful out-migrants.
To capture these variations, ESRI's model first separated the group quarters population
from the household population and, second, keyed the calculations to the size and
characteristics of the population. This stratification identified several different patterns of
change by age and sex that were applied in the cohort survival model. Births were
projected from area-specific, child-woman ratios. Migration was computed as a residual,
the difference between the surviving population and independent projections of the total
population.
Accurate allocation of funds to minority groups and tracking of immigration to the
United States are two important reasons to accurately measure the growth of population
by race and Hispanic origin. ESRI's database is supplemented with the Diversity Index, a
measure that summarizes racial and ethnic diversity. The index shows the likelihood that
two persons, chosen at random from the same area, belong to different races or ethnic
groups. The index ranges from 0 (no diversity) to 100 (complete diversity).
The U.S. Diversity Index currently stands at 59, an increase of 1 percent annually since
2000. Led primarily by Hispanic diversity, California, New Mexico, and Texas are the
most diverse mainland states, with diversity indexes higher than 70. The process of
diversification in these states is advanced; therefore, these areas are among the states with
slow rates of change in diversification. Although immigration is still rising in these states,
it has a smaller impact on the diversity level. Traditionally nondiverse states, such as
Maine, Vermont, and Connecticut, are experiencing some of the highest rates of
diversification. Pockets of diversity are common in less diverse states. For example, the
Liberal and Garden City micropolitan areas in Kansas have diversity indexes of more
than 75.
The Hispanic population now stands at 46 million, or 15 percent of the total U.S.
population. The influence of this ethnic group in American culture is on the rise, due to
growth rates of 3.7 percent a year since 2000 and a projected total of 54.7 million by
2012 (approximately 17 percent of the U.S. population). Although they are smaller
population groups, Asian and non-Hispanic multiracial populations are following
Hispanic trends closely, with growth rates of 3.8 percent and 3.1 percent, respectively.
Historical trends in race and Hispanic origin play an important role in the analysis and
forecasting process. Tracking intercensal population change by race was encumbered by
the new reporting method in Census 2000. Race was reported as a multiple-choice item,
not "one person—one race," as reported in past censuses or estimates. The Census 2000
ESRI White Paper
9
J-9663
data is not directly comparable to 1990 Census data or to any earlier estimates or
projections.
Comparisons made between single-race reporters in 2000 and 1990 underestimate the
change by race. Excluding the rapid growth of the multiracial population minimizes the
change by race from 1990 to 2000. Alternatively, combining single-race reporters with
races reported in any combination can cut down the 63 racial groups reported in Census
2000. For example, a person who reports "White and Asian" is counted as both white and
Asian. This combination of single-race and multiracial reporters overcounts multiracial
reporters and overestimates the change by race from 1990 to 2000. To achieve a true
picture of population change by race, it is important to account for the change in
multiracial reporting.
ESRI takes an innovative approach in analyzing this data to make effective use of the
9
additional information from Census 2000. The Census Bureau released most race-related
data by six single-race groups and one multiple-race group. ESRI's data preserves this
format and enables a comparison of 1990 and 2000 data for six single races and one
multiracial group. Assuming that the probability of reporting more than one race varies
by race group and geographic area as shown in Census 2000, ESRI estimates the number
of likely multiple-race reporters from 1990 Census data. The same approach is adopted
for the population of Hispanic origin by race.
The most current data sources by race and Hispanic origin are 2005 data available by
county and state from the Census Bureau's estimates or its American Community Survey
(ACS). Survey data is analyzed in conjunction with ESRI's estimate of change from 1990
to 2000 by race and Hispanic origin to establish county population by race and Hispanic
origin. Forecasts by block group combine local changes in the distributions by race and
projected change for counties. The last step controls block group distributions to county
projections.
The composition of the American household continues the slow change from marriedcouple families to nontraditional families and single-person households. Between 1990
and 2000, the dominant share of households remained married-couple families in most
states but decreased from 55 percent of all households to 52 percent in 2000. Increased
shares of single-parent and single-person units comprise the difference. The attendant
change in average household size is the decline from 2.63 in 1990 to 2.59 in 2000.
Through 2007, these changes continue, but even more gradually than in the 1990s.
The gradual change in household size makes it uniquely suitable to forecasting the
change in households from the change in household population. Average household size
is one of the most stable and predictable components of the forecasts. Household
forecasts are predicated on local patterns of change, which are controlled to the more
constant trends for states and counties. Nationally, household change stabilized in the
1990s and remains at 2.59 in 2007.
9
A more detailed discussion of ESRI's 1990–2000 race analysis is available from Sangita Vashi's paper, Trends
in the U.S. Multiracial Population 1990–2000, presented at the 2001 Southern Demographic Association
Annual Meeting.
June 2007
10
J-9663
Local change, however, is affected more by the singular composition of the population,
and trends often vary from the national norm. Nationally, average household size
decreased by less than 0.4 percent annually from 1990 to 2000. By county, the change
varied from a low of –2.1 percent to a high of 1.3 percent. An increase in household size
can result from higher rates of fertility locally or from an increase in multigenerational
households. Census 2000 has documented the increase in multigenerational households in
areas where there is high immigration or areas with housing shortages and higher costs.
From 2000 through 2007, change in household size by county ranges from –0.9 percent
to a high of 0.6 percent.
Few block groups represent a cross section of U.S. households. In areas that gained
population from immigration in the 1990s, the trend in average household size actually
reversed and increased. To distinguish local variation, ESRI's model is keyed to the
characteristics of households at the block group level. This stratification identifies several
different patterns of change by household type that are applied to forecast trends in the
characteristics of households—both family composition and tenure. Local change is
emphasized in the 2007/2012 forecasts of households and families for counties and block
groups. National and state trends are monitored by sources such as the Current Population
Survey (CPS) and American Community Survey from the Census Bureau, then applied as
controls.
Housing
ESRI's housing updates include total housing units, occupancy, tenure, and home value.
As the supply of housing units surged and investor activities scaled back, the housing
market has cooled. The homeownership rate remains at 68 percent in 2007. The sharp
growth in homeownership since 1995 has leveled off. Home value remains high, with a
median of $192,300 in 2007, compared to $111,800 in 2000. The drastic changes in home
prices from 2000 to 2007 vary by geography. Residents in metropolitan areas are
benefiting from a 73 percent increase in median home value, while their nonmetropolitan
counterparts are dealing with only a 58 percent increase. Some of the key housing
concerns include affordability in areas where home prices have outpaced the growth of
income and high incidences of foreclosures and mortgage delinquencies.
Current data on change in the housing inventory encumbers the application of past trends.
From 1990 to 2000, the housing stock increased by less than 1.4 million annually. From
2000 to 2007, the annual increment has grown to more than 1.7 million units. Total
housing units are updated from the Census 2000 base by recorded changes in the housing
inventory and estimated changes in occupancy rates since April 2000. Recorded change
in the housing inventory is culled from several data sources including ESRI's latest
addition, construction data from Hanley Wood Market Intelligence; building permits for
permit-issuing places and counties; data for new and demolished public housing from the
Department of Housing and Urban Development; and data for new manufactured homes
placed by state from the Census Bureau. Dozens of independent sources were consulted
to retrieve detailed information on housing development data where no building permits
existed. Fewer than half the counties have complete coverage with building permits.
Independent estimates of change in occupancy were calculated from U.S. Postal Service
residential lists, the Current Population Survey, and the Housing Vacancy Survey from
the Census Bureau.
ESRI White Paper
11
J-9663
The data for tenure represents owner- and renter-occupied housing units. Together, the
two components sum to total households, or total occupied housing units. A time series
model based on data from the Housing Vacancy Survey, combined with changes in the
Current Population Survey and the latest census data, guide tenure forecasts. With a
blend of top-down and bottom-up techniques, the forecasts take advantage of the latest
information from survey data at higher levels of geography while employing local
characteristics at the lower levels. The data from the lower levels of geography are
controlled to the higher levels to produce the tenure updates. Changes in owner-versusrenter occupancy are forecasted independently and controlled to the total households.
ESRI tracks the change in home value using the House Price Index (HPI) from the Office
of Federal Housing Enterprise Oversight (OFHEO). The HPI is designed to monitor
change in the loan-to-value ratio of mortgages held or guaranteed by Fannie Mae or
Freddie Mac. OFHEO affirms the "significant advantages" of the HPI to Commerce
Department surveys. ESRI has evaluated the accuracy of the HPI in estimating change in
home value through the past decade.
HPI data is released quarterly for states and metropolitan areas, with county or county
group data for larger metropolitan areas. ESRI has applied time series analysis to
extrapolate both short-term (2007) and long-range (2012) trends in home value from
states and metropolitan areas to block groups. Local estimates of home value incorporate
supply-demand characteristics, the socioeconomic traits of householders in the area, and
HPI trends assessed for larger markets.
Labor Force Update
Methodology
ESRI forecasts the civilian labor force and employed population by industry and
occupation for 2007 and 2012. While GDP, productivity, and spending softens, the U.S.
labor force is still growing at a solid pace. As of July 2007, the job market includes an
additional 2.4 million people, an increase of 1.7 percent from a year ago. Much of the
growth occurred in the South, primarily in Florida, Georgia, and Texas. Over the same
period, the U.S. unemployment rate improved by 0.3 percentage points to 6.6 percent.
While the five-year forecast of employment anticipates slower growth at 1.4 percent per
annum, unemployment will improve to 6.1 percent of the total civilian labor force.
Since 2006, many industries have added to their workforce, which is to be expected in an
economy that has grown by 3 percent for the past three years. A number of industries
expanded more than others. Some of the largest employment gains were in construction
and real estate, with more than 730,000 new jobs. However, housing-related sectors
benefited from the housing boom. The hiring increase in these sectors will temper as
housing demand drops.
Data Sources
Estimates of the civilian labor force integrate recent change in the supply and demand for
labor from the Local Area Unemployment Statistics (LAUS), Employment Projections
programs from the Bureau of Labor Statistics (BLS), and the American Community
Survey and Current Population Survey from the U.S. Census Bureau. Federal statistical
surveys are the principal sources of information about labor force trends. The LAUS
program is the premier resource for current and local economic conditions utilized by
state and local governments, media outlets, the private sector, and academic researchers.
June 2007
12
J-9663
Methods
Employment and unemployment forecasts are developed from the Census 2000 base.10
Trends are adapted from an LAUS monthly time series, projected to July 2007. LAUS
state estimates are based on the concepts and definitions from the program's main input
source, the monthly Current Population Survey, as well as the Current Employment
Statistics program from BLS and state unemployment insurance systems. Additionally,
LAUS substate models incorporate data from the decennial census. ESRI's labor force
methodology retains the strategic improvement introduced in 2004 to enhance the
accuracy of the July 1 estimate of employment status. Change between Census 2000 and
ESRI's labor force estimates is more closely tied to historical and seasonal patterns in the
LAUS state and county monthly series.
ESRI's industry and occupation updates capture temporal change from three federal
statistical sources: the ACS and CPS from the Census Bureau and the Employment
Projections program from the BLS. From the Census 2000 base, national industry and
occupation distributions are updated with trends from all three sources, and state trends
from the ACS are applied. These targets, with total employment, are used to model
substate areas.
Concepts
The civilian labor force includes the population aged 16 years and older, who are
classified as either employed or unemployed, and excludes active-duty Armed Forces
personnel. The employed population includes persons who were either

Working during the reference week as a paid employee, self-employed, working on a
farm, or working as unpaid workers for 15 hours or more on a family farm or
business

Temporarily absent from their job due to vacation, illness, bad weather, labor
disputes, or other personal reasons, excluding layoffs
Total employment excludes volunteer workers and caretakers of home or family. The
unemployed population includes persons who were

10
Neither at work nor temporarily absent from a job
Seeking employment during the last four weeks
Available to accept employment
Waiting to return from a layoff
In July 2002, the Census Bureau reported a processing error affecting its 2000 labor force estimates for areas
surrounding college towns. The error apparently overstated the number of unemployed persons and the
unemployment rate while underestimating the employed population and persons classified as not in the labor
force. Further research by the Census Bureau uncovered a response pattern to the employment questions that
extends beyond the population living in college towns. The Census Bureau estimates employment responses
for roughly 15 percent (or 500,000 people) of the working-age, civilian noninstitutional GQ population were
affected. Furthermore, it surmises the positive bias in the number of unemployed appeared to artificially
increase the 2000 U.S. unemployment rate of 5.8 percent by 0.4 percentage points. ESRI addressed the
apparent bias at the block group level and realigned the affected Census 2000 labor force estimates before
any forecasts were calculated. For more information, refer to appendix 3 in the U.S. Census Bureau Housing
and Household Economic Statistics Division report Comparing Employment, Income, and Poverty: Census
2000 and the Current Population Survey, September 2003, found at
http://www.census.gov/hhes/www/laborfor/final2_b8_nov6.pdf.
ESRI White Paper
13
J-9663
Dissimilarities in
Sources of Labor
Force Information
It is important for data users to recognize differences that exist across surveys of labor
markets. To illustrate: The U.S. unemployment rate reported in the 2000 decennial census
is 5.8 percent, while the CPS estimate for the same time period is 3.7 percent (seasonally
unadjusted). This gap stems from differences in survey methodology. Census 2000 labor
force data consists of sample estimates produced from responses reported in the longform questionnaire mailed to roughly 17 percent of all households. The CPS produces
more timely monthly data, but from a much smaller sample size. Definitions of
employment status are similar, but methods of data collection are not. The decennial
census is self-reported, while for the CPS, the Census Bureau employs experienced
interviewers to ask more probing questions to minimize survey nonresponse or data
misclassification. Due to the differences between the decennial census and the CPS, ESRI
focuses on rates of change to capture current trends and seasonal patterns to produce
accurate civilian labor force forecasts.
Impact of the Gulf
Coast Region on the
U.S. Labor Force
In an effort to quantify the number of Gulf Coast evacuees as well as their displacement
and employment status after Hurricane Katrina, the Census Bureau added hurricanerelated questions to its monthly CPS questionnaire in October 2005. Because the CPS is a
household survey, the posthurricane estimates did not represent the total evacuee
population. People living in hotels, churches, shelters, or on cruise ships were, by
definition, outside the scope of the CPS sample. One year later, the Census Bureau
discontinued collection of this data with the CPS due to survey-related deficiencies. The
small sample size, respondent confusion, and the diminished impact of the storms with
the passage of time were the primary factors that led the Census Bureau to drop these
questions from the survey in October 2006.
From the final month of evacuee estimates in this series, the CPS identified
approximately 1.1 million people who evacuated from the Gulf Coast due to the
hurricanes. Approximately 38 percent did not return home. Almost 63 percent
participated in the civilian labor force, but unemployment hit 11 percent. A large gap
exists between the unemployment of evacuees who returned versus those who did not.
Those who returned home had an unemployment rate of 7 percent; displaced residents are
experiencing rates of joblessness close to 18 percent.
For the second year, ESRI's civilian labor force estimates reflect adjustments to the
affected region's population base as well as their employment status. The large losses to
the civilian labor force base due to the population shifts are not as pronounced this year.
ESRI's 2007 employment estimates reflect the remaining effects of the storms as
measured by the LAUS program's hurricane modifications to their state and county
models. The LAUS program continued to include model adjustments to reflect the
poststorm demographic changes, although it has yet to publish monthly employment
estimates for the seven parishes that comprise the New Orleans metro area. However, the
LAUS program partner at the Louisiana Department of Labor continues to fulfill its
mandate to provide parish-level detail for the allocation of federal funds. ESRI indirectly
incorporated these estimates in its time series forecasts but realigned the 2007 labor force
distributions to the more stable metro forecast.
Louisiana was the hardest-hit state in the Gulf Coast after the 2005 hurricanes. The state
lost more than 178,000 jobs between 2005 and 2006; unemployment climbed from
8.4 percent to 9.8 percent. One year later, the state is still undergoing recovery efforts, but
June 2007
14
J-9663
its civilian labor force has improved. More than 117,000 jobs have been added since
2006, while the state's unemployment rate dropped to 7.7 percent.
The hardest-hit areas in the state were in the New Orleans metropolitan area. Before the
hurricanes, local businesses within the metro area employed nearly 580,000 people with
unemployment at 8.1 percent—1.2 percentage points higher than the national rate. By
July 2006, only 60 percent of the area's employment remained, and the rate of joblessness
jumped to almost 10 percent. One year later, the economy is slowly recovering, and job
losses are beginning to reverse. As of July 2007, the metro economy employs almost
485,000 people, or roughly 83 percent of its 2005 employment. Unemployment has
markedly improved to 6.7 percent, which is only a tenth of a percentage point higher than
the U.S. rate.
Income Update
Methodology
Supported by favorable labor market conditions, median household income has
maintained a growth of 3.2 percent since Census 2000. Median household income for
2007 is $53,150. Growing at a slightly faster pace of 3.4 percent a year, average
household income and per capita income reached $73,100 and $27,900, respectively.
Driven by job opportunities and income potential, U.S. population growth in metropolitan
areas is three times the rate of growth in nonmetropolitan areas. Today, median
household income in metropolitan areas is almost $17,000 higher than the median income
of $37,740 in nonmetropolitan areas. Ninety-five percent of U.S. aggregate personal
income is earned in metropolitan areas.
Douglas County, Colorado, continues to grow in population and prosperity. With median
household income now growing at an annual rate of 5.5 percent, this area has the nation's
highest county median household income, at more than $111,000. In the last year,
another four counties have passed the $100,000 mark for median household income. In
addition to Douglas County, Loudoun and Fairfax Counties, Virginia, and Hunterdon
County, New Jersey, households in Somerset and Morris Counties in New Jersey, as well
as Los Alamos, New Mexico, and Falls Church City, Virginia, now have median incomes
of more than $100,000.
Median disposable income is $41,640 in 2007; average disposable income stands at
$55,000. On average, a household's disposable income is approximately 75 percent of its
pretax income. Householders younger than 25 years or older than 65 years have
disposable incomes of more than 80 percent of their income. With a median household
income of $67,000, householders in the 45–54 age group represent peak earning years
and pay the highest taxes. They earn a median disposable income of $51,360 and pay
23 percent of their income in taxes. This demographic group, near the end of the baby
boomer cohort, has accrued a median net worth of $155,000 and an average net worth of
$625,000.
Data Sources
ESRI's projection base is the income that was reported in Census 2000. Technically, 2000
income data represents income from 1999 because the Census Bureau tabulates income
received in the "last year" before the decennial census. Similarly, ESRI's 2007 income
updates represent income received in 2006, expressed in 2006 dollars. Projections for
2012 are shown in 2011 dollars, assuming a continuation of the current rate of inflation.
ESRI White Paper
15
J-9663
ESRI uses the definition of money income used by the Census Bureau, which enables the
direct comparison of income updates and decennial census data. For each person 15 years
old and older, money income received in the preceding calendar year is tallied from each
of the following sources: earnings, unemployment compensation, Social Security,
Supplemental Security Income, public assistance, veterans' payments, survivor benefits,
disability benefits, pension or retirement income, interest, dividends, rent, royalties,
estates and trusts, educational assistance, alimony, child support, financial assistance
from outside the household, and other income.
Data for consumer income collected by the Census Bureau covers money income
received (exclusive of certain money receipts such as capital gains) before payments for
personal income taxes, Social Security, union dues, Medicare deductions, and so forth.
Therefore, money income does not reflect the fact that some families receive part of their
income in the form of noncash benefits such as food stamps, health benefits, rent-free
housing, or goods produced and consumed on a farm. In addition, money income does
not include noncash benefits such as the use of business transportation and facilities and
full or partial payments by business for retirement, medical, and educational expenses,
and so forth.
Income Methods
To estimate income for all households and family households, ESRI evaluated several
federal data sources including the Current Population Survey and American Community
Survey from the Census Bureau plus the personal and per capita income data and the
Census of Employment and Wages from the Bureau of Labor Statistics.
After Census 2000, ESRI conducted a detailed evaluation of data sources employed in
past income forecasts and analysis of more recent data from the Supplementary and
American Community surveys. Data for 2000 from each source varied from the income
that was reported in Census 2000. It was concluded that one point in time is just not a
good measure of a data series. For any given year, any estimate of income is likely to
vary from the true population value. However, the sources that ESRI employed
throughout the 1990s proved to be effective measures of change in income. Testing
revealed the power of time series data in tracking income. ESRI's postcensal updates
emphasize the use of time series data from household surveys to establish a base trend
line. Annual updates evaluate current trends in wage inflation and other economic shocks
that impact income growth.
After forecasting the state income distributions, household income is estimated for
counties, tracts, and block groups. ESRI's income forecasts are uniquely designed to
distinguish local variation, changes in income inequality, and urbanicity as differentiators
of income growth. The model correlates the characteristics of households at the block
group geography level with changes in income. This stratification identifies several
different patterns of change by household type that are applied to forecast trends in
income. The annual change in income is derived from national surveys. Modeling links
the current income change to all households with similar socioeconomic characteristics.
Separate forecasts of the change in income by strata are aggregated to comprise the
income distributions.
Once the base 2000 income tabulations are updated, the distributions are extended to
provide additional data for the wealthiest households. The Pareto function is employed to
extend the upper interval of the income distributions from $200,000 or more to include
June 2007
16
J-9663
the intervals $200,000–$249,999, $250,000–$499,999, and $500,000 or more. Finally,
the models are calibrated to distinguish the change in average household income, for
example, from the change in median income.
Average and median income for 2007 and 2012 are calculated in the same way that the
Census 2000 average and median income are computed. Medians are calculated from the
distributions using linear or Pareto interpolation; averages, from aggregate household
income.11 Differences arise from the distributions. The 2000 income base from the
Census Bureau is different from the income tables that they report to the public. ESRI's
2007/2012 income base is also different from the Census 2000 reported tables. Medians
and averages for 2007/2012 represent the extended income distributions, to $500,000 or
more. It is the extended income distributions that provide the base for updating aggregate
income. Using the midpoints of income intervals in the extended distribution, aggregate
household income is calculated to be consistent with the distribution of household
income—and the aggregate incomes that are estimated for the extended distributions of
income by age of householder.
Household income reported by age of householder is updated to be consistent with the
2007/2012 distributions of household income and age of householder. To update the age
distribution of householders, the ratio of householders by age to population by age in
2000 is extrapolated to 2007/2012 and applied to the current age distributions. After the
targets are set, the 2000 distributions of household income by age of householder by
block group are fitted to current distributions of households by income and by age of
householder.
Disposable Income
Disposable income represents an estimate of a household's purchasing power, or after-tax
income. The proportion of household income left after taxes is estimated from special
studies conducted by the Census Bureau to simulate household taxes. With the release of
the 2004 Annual Social and Economic Supplement (ASEC) to the Current Population
Survey, a new tax model was implemented. The new model performs a statistical match
of tax variables not collected in the ASEC with the 2000 Statistics of Income (SOI) file
from the Internal Revenue Service. The most recent release of tax data in the 2005 ASEC
implements the 2001 SOI file.12 These changes impact the time series of tax variables
available and are reflected in this release of ESRI's disposable income. Four types of
taxes are deducted: federal individual income taxes, state individual income taxes,
Federal Insurance Contributions Act (FICA) (or Social Security) and federal retirement
payroll taxes, and property taxes for owner-occupied housing.
Sophisticated modeling techniques are employed to improve the handling of top-coded
earnings and tax data from the CPS. Internal Revenue Service tax rates are used as
guidelines for model testing. ESRI then applied the proportions of after-tax earnings to
income intervals that were cross-tabulated by age of householder for each state. Statespecific proportions account for the variation in taxes by state. The proportions, or
11
For further information on calculations used with Census 2000 data, see Census 2000 Summary File 3
Technical Documentation prepared by the U.S. Bureau of the Census, 2002.
12
Further information on changes to tax variables in the latest Current Population Survey is available at
http://www.census.gov/hhes/www/income/cpsasec2005taxmodeldoc.pdf. A detailed review of the tax model
is available at http://www.census.gov/hhes/www/income/oharataxmodel.pdf and http://www.irs.gov/pub/
irs-soi/06ohara.pdf.
ESRI White Paper
17
J-9663
multipliers, were then applied to the age by income forecasts for block groups and
counties to calculate disposable income.
Net Worth
Net worth is estimated from data on household wealth that is collected from the Surveys
of Consumer Finance (SCF) from the Federal Reserve Board from 1992 through 2004.
From 2001 to 2004, inflation-adjusted average and median net worth grew annually at
2 percent and 0.5 percent, respectively. This growth rate is somewhat slower than the
growth reported in previous surveys. Most of the recent growth is attributable to the
appreciation in residential real estate, a rise in the number of new and second
homeowners, and the growth in speculative investment properties. However, household
debt also increased as a result of growing residential real estate portfolios. The amount of
secured debt serviced as a share of household income also rose despite the decline in
interest rates over this period. And although the stock market has rebounded since the last
recession, the amount of corporate equities held by households has declined since 2001.
Between 2006 and 2007, the U.S. median net worth grew by 2.6 percent annually to
nearly $106,000. Average net worth rose 6.9 percent to more than $517,000 during the
same period. Some of the largest gains were in the 55-to-64-year-old cohort. This group's
average appreciated nearly 7 percent to more than $947,000.
The size of the triennial surveys used in estimating net worth is approximately 25,000
households. The major strengths of the SCF surveys lie in their enhanced representation
of wealthy households and in the comprehensive measurement of net worth components.
By definition, net worth equals total household assets less any debts, secured or
unsecured. Assets include own home, rental property, own business, individual retirement
accounts (IRAs) and Keogh accounts, pension plans, stocks, mutual funds, and motor
vehicles. Examples of secured debt include home mortgages and vehicle loans; unsecured
debt includes credit card and other bills or certain bank loans.
The first step in calculating net worth is to measure the relationship of net worth to
household income by age of householder. The relationship is further differentiated by
tenure since homeownership represents a major factor in household wealth. The next step
is to model the relationship statistically to enhance the reliability of the estimates. This
effort represents a model introduced in 2004 to reflect the recent trends in the housing
markets and their impact on net worth. As interest rates have risen from historic lows and
more homes flood the resale market, the value of residential real estate will not be as
significant a driver of household net worth growth as it was during the market's peak.
The extension of the 2000 household income distribution from an upper interval of
$200,000 or more to $500,000 or more also enhances the calculation of net worth for the
wealthiest households. The 2007 estimates of net worth reflect current income and
homeownership with adjustments for inflation and updates based on economic growth
since the 2004 SCF survey.
ESRI again paid special attention to the adjustment of the net worth-income relationships
for homeowners along the Gulf Coast to account for changes in the posthurricane value
of residential housing in the hardest-hit areas. In 2007, as the recovery efforts continue in
the Gulf Coast, fewer areas necessitate home value adjustments for the estimates of net
worth.
June 2007
18
J-9663
Use of Projections
Projections are necessarily derived from current events and past trends. The past and the
present are known; the future must be extrapolated from this knowledge base. Even
though projections represent the unknown, they are not uninformed. Guidelines for the
development of projections also inform the use of those projections:

The recent past provides a reasonable clue to the course of future events, especially if
that information is tempered with a historical perspective.

A stable rate of growth is easier to anticipate than rapid growth or decline.

The risk inherent in projections is inversely related to the size of an area: the smaller
the area, the greater the risk.

The risk increases with the length of the projection interval. Any deviation of the
projected trends from actual events is amplified over time.
ESRI revises its projections annually to draw on the most recent estimates and
projections of local trends. However, this data can be complemented with personal
knowledge of an area to provide the qualitative, anecdotal detail that is not captured in a
national database. It is incumbent upon data users and producers to incorporate as much
information as possible when assessing local trends, especially for areas that are subject
to "boom-bust" cycles.
ZIP Code Update
Methodology
Data for residential ZIP Codes is estimated by ESRI; Census 2000 geographic areas are
the building blocks for the estimates. Because ZIP Code boundaries change frequently,
census geography provides a comparatively stable base for the development of ZIP Code
data. ZIP Code data has been estimated from block groups, which are assigned to
residential ZIP Codes by overlaying the centroids of component blocks onto ZIP Code
boundaries. Expressed as latitude-longitude coordinates, centroids approximate the
geographic centers of blocks. If the centroid of a block falls within a ZIP Code, it is
included in the residential inventory; otherwise, it is classified as nonresidential. Block
data is then aggregated, and the ratio of block totals to block group data is used to
apportion demographic characteristics to a ZIP Code.
The 2007/2012 updates include data for 30,006 residential ZIP Codes. This
geodemographic method does not provide data for ZIP Codes with no assigned boundary.
If a polygon is not defined for a ZIP Code, or no blocks are assigned to a ZIP Code
polygon, data cannot be retrieved. In most cases, information about post office box ZIP
Codes or single address ZIP Codes is incorporated with the data for the enclosing,
residential ZIP Code.
Data Source for
Boundaries
Tele Atlas creates boundary files for ZIP Codes. The complete ZIP Code inventory
includes both point and boundary ZIP Codes. ZIP Code boundaries are current as of
November 2006.
ESRI White Paper
19
J-9663
Comparisons
over Time
ZIP Codes are not amenable to time series analysis, thereby preventing a direct
comparison with ZIP Codes from previous updates. Changes typically include new
residential ZIP Codes (65 in 2007), deleted ZIP Codes (26 in 2007), and boundary
revisions. The 2007 inventory of residential ZIP Codes includes 8,985 ZIP Codes that
have the same geocode as the 2006 inventory but a different population base as a result of
boundary changes or slightly different block allocations. These changes reflect revisions
of ZIP Codes by the U.S. Postal Service in addition to any changes in the techniques used
by Tele Atlas to define ZIP Code boundaries.
About ESRI's Data
Development Team
Led by chief demographer Lynn Wombold, ESRI's data development team has a history
of more than 30 years of excellence in market intelligence. The combined expertise of the
team's economists, statisticians, demographers, geographers, and analysts totals nearly a
century of data and segmentation development experience. The team has crafted data
methodologies such as the demographic update, segmentation, the diversity index, and
the Retail MarketPlace that are now industry benchmarks. Authors of white papers such
as Evaluating Population Projections: The Importance of Accurate Forecasting and
Trends in the U.S. Multiracial Population from 1990–2000, the team frequently presents
sessions and papers to industry and professional organizations.
June 2007
20
Analysis Summarization Statistical Formulas
Formulas defined below are used in Business Analyst for summarizing values during
many of the analyses, spatial overlay, and find similar and during report generation.
Consider BlockGroup (Bg) as a feature from a polygonal data layer. Instead of
BlockGroup features from the Census Tract layer, ZIP Code layer, County layer, or State
layer can be used.
Consider geometry “A” as a boundary geometry. For example, it could be a boundary
feature from analysis results (i.e., drivetime, simple ring, other trade area).
Below is a discussion about computing the value for one field. The same field is used in
“Bg” and (possibly) one additional weight field in “Bg” (only if WeightField is defined in
metadata).
Terminology definition:
1. Value(Bg) is the value of the field in the Block group “Bg”.
2. Value(A) is the summarized value of the corresponding field for specified
geometry “A”. This is an output value, and it will be computed using the
formulas below.
3. Ratio(Bg,A) is the weight of the part of block group “Bg” that is inside geometry
“A”. The weight variable is determined by the user’s choice during Analysis
Layer Setup and is stored in the layer’s metadata.
For arbitrary geometry “C” (“C” can represent either “Bg” or “A”) the following
values will be determined:
a. POP(C) is the number of population in geometry “C”.
b. HHD(C) is the number of households in geometry “C”.
c. HU(C) is the number of housing units in geometry “C”.
So Ratio of (Bg,A) could be:
Area( Bg I A)
• Ratio( Bg , A) =
Area( Bg )
•
Ratio( Bg , A) =
POP( Bg I A)
•
Ratio( Bg , A) =
HHD( Bg I A)
•
Ratio( Bg , A) =
HU ( Bg I A)
POP( Bg )
HHD( Bg )
HU ( Bg )
Ratio( Bg , A) = 1 if Bg is fully inside A
•
4. WeightFieldValue(Bg) is a value of the additional weight field of block group
“Bg”.
5. Weight(Bg,A) is a weight of the block group as the degree of influence of the
value of this particular block group on the value calculated for the geometry A. It
is not the same as the value of weight field for the block group. It depends on the
basic layer’s metadata and may be different for various fields.
It could be equal to:
• Weight ( Bg , A) = WeightFieldValue( Bg ) * Ratio( Bg , A) (if weight field is
used)
•
Weight ( Bg , A) = Ratio( Bg , A) (if weight field isn’t used)
So the term Weight just takes into account both Ratio and value of weight field.
Now determine how to calculate Value(A). There are several ways to calculate it:
summarization, summarization with weight, average, and average with weight.
Sum:
Value( A) =
∑ Ratio( Bg , A) * Value( Bg )
∀Bg , Bg
I A≠ 0
Weighted sum:
Value( A) =
∑
∀Bg , Bg
Ratio( Bg , A) *Value( Bg ) *Weight ( Bg , A)
I A≠ 0
Average:
∑ Ratio( Bg , A) * Value( Bg )
Value( A) =
∀Bg , Bg
I A≠ 0
∑1
∀Bg , Bg
I A≠ 0
Average with weight:
∑
Value( A) =
∀Bg , Bg
I
Ratio( Bg , A) *Value( Bg ) *Weight ( Bg , A)
A≠ 0
∑
∀Bg , Bg
Weight ( Bg , A)
I A≠ 0
The median is the number in the middle of a set of numbers; that is, half the numbers
have values that are greater than the median, and half have values that are less.
Variance and standard deviation.
X is an equiprobable sample.
x is an element of X
n is a count of v in X,
n = ∑1
x∈ X
Variance:


n∑ x −  ∑ x
 x∈X 
Value( A) = x∈X
n(n − 1)
2
2
Standard deviation:


n∑ x −  ∑ x
x∈ X
 x∈X 
n(n − 1)
2
Value( A) =
2
Density
Density is the most difficult type of field processing in Business Analyst. For each
demography the demography area—for example, in square miles—needs to be
calculated. To achieve precision it is required to create a spatial reference for each
demography and then project it. Then, there are a number of methods that can be used to
calculate the final value of Density depending on the case:
 Area( Bg ∩ A)


* Value( Bg ) 
(Ratio( Bg , A) * Value( Bg ) ) ≈
 Area( Bg )
⇔
Density ( A) = ∑
∑
1.
∩
∩
Area
Bg
A
Area
Bg
A
(
)
(
)
Bg ∩ A
Bg ∩ A
⇔
Value( Bg )
Bg ∩ A Area ( Bg )
∑
2. Density ( A) =
∑
Bg ∩ A
(Ratio( Bg , A) * Value( Bg ) )
Area( Bg ∩ A)
∑ Area( Bg ∩ A)
Bg ∩ A
3. Density ( A) =
∑ (Ratio( Bg , A) * Value( Bg ))
Bg ∩ A
∑ Area( Bg ∩ A)
Bg ∩ A
Value( Bg )
Bg ∩ A Area ( Bg )
≈ ( see1)
∑ Area( Bg ∩ A)
∑
Bg ∩ A
CITY OF HARKER HEIGHTS, TEXAS
305 Millers Crossing
Harker Heights, Texas 76548
Phone: 254-953-5600
Fax: 254-953-5666
E-mail: [email protected]