Model of Migration in the Ohio River Basin Energy Study Region

-
ORDES
PHASE II
OHIO RIVER DASIN ENERGY STUDY

-------
August 1980
A MODEL OF MIGRATION IN THE
OHIO RIVER BASIN ENERGY STUDY REGION
By
Steven I. Gordon
Christopher Badger
The Ohio State University
Columbus, Ohio ^+3210
Prepared for
Ohio River Basin Energy Study (ORBES)
Subcontract under Prime Contract EPA R8O5588
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D.C. 20^60

-------
CONTENTS
Acknowledgements 		ii
Tables 	ill
Figures		iv
Introduction and Purpose		1
Literature Review		2
Data Description		ij-
Derivation of Migration Models 		22
Migration Impacts of the ORBES Scenarios 		29
References		^3
Appendix	
i

-------
Acknowledgements
We must acknowledge the invaluable assistance of a number of
individuals. First, we thank Professor Oscar Fisch of the Department
of City and Regional Planning for his advice in the development and
execution of the project. We must also thank Dr. Jerome Pickard and
Mr. Joseph Cerniglia of the Appalachian Regional Commission for
providing us with migration data without which the project would not
have gone forward.
Finally, we must thank the ORBES core team and management team
for funding the project. Special thanks goes to Professor Gary Fowler
of the University of Illinois at Chicago Circle, Department of
Geography and Energy Resources Center, for his advice and counsel.
The authors bear all responsibility for any errors contained
herin.
ii

-------
TABLES
1.	Migration Data		10
2.	Net Migration For Each Region Within ORBES 1965-1970		n
3.	Migration From Each Sub-Region i to Each Sub-Region
within ORBES		12
Employment and Demographic Data	
5.	Characteristics of the Population Data		15
6.	Regional Summary Characteristics		15
7.	1970 Regional Employment by Sector	
8.	1967 Regional Production, Won-Energy Sectors 		19
9.	1967 Employment/Output Ratios		21
10.	General Linear Models Procedure		27
11.	Actual and Estimated Migration For 1970		28
12.	Percent Change In Employment Within Each Region From
1965 to 1970 For Various Sectors		32
13.	Sub-Region Wet Migration Impacts for Scenario 1		37
1^. Sub-Region Wet Migration Impacts for Scenarios 1, k, and 5....
iii

-------
FIGURES
1.	Migration Prediction Process		8
2.	AEC Migration Regions		9
3.	Manufacturing Shift		33
4.	Construction Shift		3^-
5.	Service Shift		35
6.	Financial Shift		36
7.	Scenario 1, Net Migration		38
8.	Scenario U, Net Migration		39
9.	Scenario 5, Net Migration		UO
iv

-------
1.0 Introduction and Purpose
The Ohio River Basin Energy Study (ORBES) has as its purpose
the analysis of the environmental, economic, and social impacts of
various scenarios of every facility development in the region. Much
of this analysis is dependent upon alternative economic growth
scenarios which show the energy and economic demand distribution
across industries in the region. Similarly, population forecasts
have focused on the region as a point and derived alternative
projections of population growth.
A remaining question in this type of analysis is the impact
of energy scenarios and policies on the internal movements of
population and industry within the ORBES region. For any particular
regional economic and population forecast, there is implied seme
level of regional growth. The location of this growth within the
region may be influenced by, or may induce, the location and move-
ment of population within the region. The regional forecasts thus
provide upper bounds on this growth while the present project focuses
on alternative ways in which this growth might be distributed.
Broader economic and energy issues within the U.S. undoubtedly are
having, and will have, an effect on migration between ORBES and the
rest of the nation. However, the ORBES study does not offer
projections for the nation as a whole which would enable us to
generate and use a model to predict such migration.
The report is divided into several chapters. Chapter 2 reviews
other attempts to derive empirical models of migration which relate
to the theory behind our model. Chapter 3 discusses the data which
was required to derive our migration models. Each data source is
described in terms of geography, time, and variable type. Chapter
describes the models which were derived and the pros and cons of
using each for simulating ORBES impacts. Finally, we use one of
these models in Chapter 5 to simulate the migration impacts of the
ORBES scenarios under alternative sets of assumptions.
1

-------
2.0 Literature Review
A number pf studies have been undertaken in order to delineate
those factors which affect migration. Many approaches have been
taken. Because of the nature of our work for ORBES, we will limit
our discussion to those models which relate migration to economic
growth.
Several major hypotheses have been tested with respect to the
relationships between economic growth and migration. These have
resulted in a "chicken and egg" controversy over causality. The
question remains whether economic growth (or decline) causes
migration or whether migration causes economic growth. Among the
first to investigate these matters, Borts and Stein (l) contended
that migration has tended to narrow wage and income differentials
over time and is thus a factor of economic growth and change. A
model derived by Muth (2) supported this argument on the basis that
total urban employment growth was affected more by migration than
migration was affected by employment growth.
On the other hand, several authors have derived models with
the tinderlying assumption that economic growth stimulates migration.
Lowry (3) found that economic conditions at the destination greatly
influenced the decision to migrate. Miller (4) contended that Lowry
and others had incorrectly controlled for conditions at the origin
of migration and thus had found those conditions not to be important.
Miller concludes that areas with high immigration rates also have
high out-migration rates because people who have migrated o nee will
have a higher probability of migrating again.
More recently, Santini (5) has utilized a reformulation of the
Lowry model for 1+9 SMSA's in the North Central U. S. His results
support the Lowry findings and contradict those of Muth.
Regardless of which arguments one supports or even whether there
is a simultaneous process, the models used to study migration have
much in common. Most have adopted a modified "gravity-type" or
spatial interaction model in which economic conditions, character-
istics of the population at the origin and destination, and distance
are the most important variables (11)." The general form of these
migration equations can be given as:
2

-------
M = f[E.)E,P,(l/d")]
-LJ	1 J 1	ij
where M.. = the migration from region i to region j
E j. = employment variables at i and j
i J
j = population variables at i and j
d = the distance between i and j
n = a constant.
Most researchers have employed various kinds of regression techniques
in an effort to define the relationship of specific variables to
migration rates. A number of authors have used employment variables
as a measure of employment opportunity that induces migration (3>^>6).
The variables used include employment differentials (total or by
sector), income differentials, unemployment rates in both the sending
and receiving region, and skills of the migrants.
The population variables used include age, sex, education, and
race (7,8,9,10).
In theory, migration frcm an area would be induced by high
unemployment and low income. Migrants would tend to be younger and
better educated and would be attracted to regions with employment
opportunities at higher incomes. All of these effects will be
mitigated by distance. The longer the distance, the lower the
propensity to migrate because of the associated costs and because
less information is available regarding employment opportunities.
Education helps overcome this distance factor in that employment for
better educated persons is more widely advertised.
The statistical methods used by migration researchers have
included ordinary as well as two stage and three stage least squares
regression (3,^,11,12). Logarithmic functions have been frequently
used to convert the non-linearities in the equations to linear
relationships.
Most of these studies focused entirely on migration between
meg or metropolitan regions or on interregional migration with a
division of the U. S. into 7-10 regions. The ORBES migration study-
is unique in that it was undertaken for 1+3 relatively amm i regions
within the ORBES area.
3

-------
3.0 Data Description
The ORBES migration model does not seek to advance the state-of-
the-art in migration analysis. Rather the developed theories and
procedures cited in section 2.0 were utilized to formulate an appro-
priate model based on migration, employment, distance and demographic
data specific to the ORBES region. Before reviewing this data and
how it was organized, it is helpful to understand the basic process
developed to predict internal migration.
3.1	Overview of Migration Prediction Process
The prediction process developed in this study followed three
basic steps:
1)	Data was assembled describing the number of persons
migrating from 1965 to 1970 between each of ^3 regions
within the ORBES geographic area. Employment and
demographic data describing each of these sub-regions,
as well as the distances between them, were also
compiled for the same time period. A regression analysis
was then conducted to define the relationship between
the rate of migration as the dependent variable and
the independent variable of employment, distance and
demographic characteristics. The result was an equation
of the form shown in section 2.0.
2)	Using the ORBES i/O Model to predict sector output levels
for a given scenario, and using existing data defining
the ratio of output to employment for each sector,
employment predictions for each scenario were made.
3)	Combined with distance and demographic data, these
employment predictions were then used as the independent
variables in the equation defined in step 1 to predict
new migration rates.
The fall migration prediction process is described in Figure 1 and
discussed in greater detail in section 4.3.
3.2	Migration Data
Ideally, migration data at the county level is necessary in order
to obtain the best picture of migration trends and determinants.
These data are available from the 1970 Census of Population only
through a special tabulation by The Bureau of the Census and was
beyond the financial resources and time frame of the present study.
b

-------
Consequently, a tabulation of the 1965-70 migration data was obtained
from The Appalachian Regional Commission (ARC). The ARC originally-
defined a set of regions which sub-divide the United States and which
follow county lines but occasionally cross state boundaries. The
Bureau of the Census then tabulated migration data for these ARC
regions.
The ARC defined these sub-regions so that each encompasses a
single economic area. In this manner localized population movements
having little regional significance tend to be ignored, while
migration from one economic area to another, usually involving a
change of both residence and place of work, is emphasized. These
sub-regions do not, however, make complete sense relative to ORBES
since some extend beyond the ORBES boundaries. In our analysis, if
a portion of an ARC sub-region fell within the ORBES boundary, the
entire sub-region was included in the analysis. Although this is not
an ideal solution, allocation of portions of regions would have
created equal or worse difficulties. Figure 2 shows these sub-regions
and Appendix A lists the ORBES counties within each sub-region.
For our analysis, then, only the tabulations pertaining to the
43 ARC sub-regions which fall partially or wholly within the ORBES
region were used. The general content of this tabulation is described
in Table 1; here one can see that the number of migrants from every
sub-region to every other sub-region, is given by age, sex and race.
From this data, net migration can be computed for the 43 sub-regions
as shown in Table 2. In this table, net migration for a given
sub-region is the result of migration between that sub-region and
every other sub-region in the U. S. From this data, it is clear the
ORBES region as a whole experienced a net loss of population frcm
1965 to 1970 and that only 15 of the 43 sub-regions showed an increase
in population.
Since our analysis is confined to modelling migration internal to
the ORBES region, only the tabulations of migration between the 43
sub-regions within ORBES are relevant. These are displayed in Table 3.
Note that these migration rates are not broken down by age, sex or
race, and that they represent the number of persons migrating from one
region to another rather than a net figure. It is this data which was
used as the dependent variable in our regression analysis.
A close look at Table 3 reveals several facts about migration
within the ORBES region from 1965 to 1970. Regions which included
large cities exchanged people with neighboring regions at a high rate.
Seme of these regions experienced a net loss in the exchange.
Region 1 (Pittsburgh) lost population to nearby regions 2 and 3, as
well as to region 75> the Canton-Akron area. Region 70, the area just
south of Cleveland, also lost people to region 75. Region 71,
5

-------
(Cincinnati) lost population to surrounding regions, particularly to
region 72 which includes Dayton. Those regions which experienced a
net gain in population did so as a result of migration from southern
rural counties. Region 73 (Columbus) and 72 (Dayton area) drew people
from southeastern Ohio (regions 13,li,15)j region 152 (Lexington) drew
from southeastern Kentucky (regions 22,23,2U,28 as well as surrounding
regions 151,153). In general there was a net movement of people away
frcm rural regions of Kentucky, West Virginia and southeastern Ohio
to major cities. Region 18, for example, experienced a net loss to
the regions near Pittsburgh, Wheeling, and the Akron-Canton area.
3. 3 Employment, Demographic and Distance Data
The employment and demographic data corresponding to the 1965-
1970 migration period were obtained from an array of census materials
at the county level and then summed to get totals for the U3 ABC
sub-regions. Tables ^ and 5 show what data were collected and their
source(s). They included county population, characteristics of the
population such as median age and years of schooling completed, female
to male ratios, employment by sector and percent unemployment. The
actual data is summarized in Tables 6 and 7. The distances between each
of the 1+3 ARC sub-regions was established by measuring frcm centroid to
ccntroid on a scaled map.
3.U Data Structure
The data in Table 3, 6 and 7 (with the exception of the net
migration figure in Table 6) provided the data base for establishing
the relationship between migration on the one hand, and employment,
demographic and distance data on the other. A regression analysis was
conducted with this data in which gross migration rates were treated as
the dependent variable and selected employment, demographic and distance
variable were treated as the independent variable. In depth discussion
of the regression analysis is given in sections b.0-k.2.
This regression data was organized in the following manner: For
each sub-region i there are k2 data records describing migration from
sub-region i to each of the other sub-regions j. The basic structure
of any one record, then, is as follows:
Region i data: demographic and employment data describing
sub-region i
Region j data: demographic and employment data describing
sub-region j
Region i - Region j data: distance between i and j
^ migrants from i to j
6

-------
Since there are ^3 sub-regions a maximum of ^3 times k2 (=1806)
data records describing all sub-regions of the ORBES area and all
migration between these sub-regions from 1965 to 1970 were possible.
The Census Bureau was unable to tabulate migration for some sub-region
pairs and thus the actual number of complete records is 1757.
3.5 Employment/Output Ratios
In order to predict migration patterns for each scenario a
prediction of employment levels in each sub-region for the major
economic sectors is necessary. These predictions were obtained
through the use of employment/output ratios which establish the
relationship between output and employment levels in a given sector.
Such ratios were calculated for a variety of employment sectors for
the year 1967.
1967 output levels were provided by the ORBES I/O Model as shown
in Table 8 (13). Standard Industrial Codes (SIC) were assigned to each
sector in Table 8 and the employment corresponding to these sectors for
all counties in the ORBES region was obtained from the 1967 County
Business Patterns. The employment for each sector in Table 8 forthe
entire ORBES region was then obtained by summing across all U23 counties
for each SIC code. The employment and output data were then further
aggregated into more broadly defined sectors as shown in Table 9- The
employment/output ratios were then calculated from these figures and
are also shown in Table 9.
When multiplied by the output level projected by the ORBES I/O
Model for a given scenario, these employment/output ratios yield
the corresponding predicted sector employment for that scenario. Since
the result is a sector employment figure for the ORBES region as a
whole, it is necessary to break that total down and allocate it to
each sub-region. 2Ms was done by using the data in Table 7 to
calculate the percent of a given sector's total employment contributed
by each sub-region in 1970. These same percentages were then used to
calculate a sub-region's share in the predicted ORBES total employment
for a given sector for the year 2000. The predicted sector employment
by sub-region can then be used, along with demographic and distance
data, as independent variables in the migration prediction equations.
7

-------
Figure 1 : Migration Prediction Process
Scenario
Step 2
Production Output Totals From
Regional Growth Model	 	
Step 1
Energy and Fuel
Demand Model
Coal Supply and
Allocation Model
Siting
Model
Estimated Mining
Employment by
County
	1
Estimated Construction
& Operation Employment
by County
Aggregate Across
Sub-Regions
19 70 Employment by
Industry for ORBES
Region
19 70 Employment by
Industry for ORBES
Sub-Regions
Employment/Output
Ratio
Year 2000 Employment
by Industry for
ORBES Region

A Employment by
Industry for ORBES
Region
1970 Sub-Regions share
of ORBES Employment by
Industry (Percent)
A Employment by
Industry for each
Sub-Region
Year 2000 Employment by
Industry for Each
Sub-Region
(r
Regression Model
Distances Between
Sub-Regions
Sub-Regions Demo-
graphic Data
Estimated Migration for
Scenario
8

-------
0RBES REGION
MGll RE 2
ARC MIGRATION REGIONS
75
80
99
77
78
83
88
73
72
82
84
87
22
152
89
153
28
24
154
155
25
[«p
H OUTSIDE ORBES REGION

-------
Table 1: Migration Data
Item:
Total number of persons
migrating from sub-region i to (
sub-region j from 1965 to 1970 <
Number of persons migrating from
sub-region i to sub-region j from
1965 to 1970, broken down by:
Source:
Appalachian Regional Commission
data derived from U.S. Bureau
of the Census migration data
same
Age: 5-17
18-24
25-34
35-55
over 55
Sex: male
female
Race: white
other
10

-------
i
2
3
7
8
11
12
13
14
IS
16
17
IB
19
21
22
23
24
25
28
70
71
72
73
75
77
78
79
80
61
82
83
84
85
97
88
89
99
.51
152
53
54
iTHER
2734
-608
-156
206
-332
-889
746
-434
-100
—56 3
-466
9
-22 5
1069
2327
-162
-157
-28 e
-485
¦>168 8
6020
1161
2554
3446
2390
1072
119
-29 3
935
1668
3246
513
975
-1034
915
3056
538
4102
63 7
113
54 0
-359
-1850
-227 3
NET MIGRATION FOR EACH REGION WITHIN ORBES: 1965 TO 1970
9Y AGE	BY SEX
NET NUMBER
MIGRANTS
AGE 5-17
AGE 18-2*
AGE 25-3*
AGE 35-5*
AGE 55
FEMALES
MALES
-57971
-90*1
-26357
-3560
-8211
-10802
-291**
-28827
1001
11*2
3188
-2909
-386
-3*
763
238
-7095
118*
-*707
-1768
-791
-1013
-3*2*
-3671
-2301
-509
3803
-*122
-921
-552
-2015
-286
-1*987
-1200
-10189
-1702
-956
-9*0
-7955
-7032
-13793
-125*
-7807
-1633
-1668
-1*31
-6679
-711*
-*072
-BOO
-2603
-812
-25 2
395
-3126
-9*6
-481
1*
-2659
1563
191
*10
-109
-372
-103
-339
3091
-29**
—*2
131
-1303
120C
-6172
-803
-5122
25
-756
*8*
-2912
-3260
-8779
-1895
-305*
-137*
-1790
—666
-*23*
-*5*5
-1602
199
-167*
17
-179
35
-553
-10*9
-7120
-15*7
-2578
-2270
-*18
-307
-3515
-3605
-29826
-5*07
-13951
-366 *
-**2*
-2380
-15*62
-1*36*
-1*974
-21*8
-8868
-1237
-1*7*
-12*7
-8071
-6903
-6153
-827
-3556
-505
—956
-*09
-2897
-3256
13*7
835
2222
-165*
*
-60
950
397
-597*
-1068
-3061
-506
-791
-5*8
-31*0
-283*
-2133
8*
-12**
-1373
339
61
-1255
-878
-*12*9
-7309
-20537
-5999
-*626
-2778
-21108
-201*1
-33577
-6193
-1*2*7
3082
-6*30
-9789
-13753
-1982*
-11296
-3589
1020
-2286
-2829
-3612
-*691
-6605
7096
1091
1397
6007
96
-1*95
3761
3335
16*06
-922
2*315
-*882
-9*9
-1156
9*16
6990
2*09
3210
-3292
5715
351
-3575
2791
-382
-1139
1101
-2**2
785
-81
-502
-981
-158
-*278
331
-3832
-2*
-223
-530
-1589
-2689
3256
*10
2355
-261
10*3
-291
176*
1*92
37**
1728
-1995
2*08
1*82
121
1882
1862
-5165
182*
-6737
1007
69
-1328
-18*5
-3320
3*87
1386
-613
*909
-161
-203*
22*9
1238
—31V*
-37 89
11053
-6 357
-250*
-1597
-166*
-1530
*76*
501
87*6
-3115
-825
-5*3
19*2
292 2
-10275
-1052
-7259
-238
-1**5
-281
-*89*
-5381
-5902
1171
-7030
173
-57 5
359
-220*
-3698
11875
-1725
22598
-7*1*
-1320
-26*
7809
*066
6796
1*87
532*
-1718
9**
759
2801
3995
6937
*298
-1766
3*53
1038
-86
30*1
3896
-6*
6*3
-2959
2605
615
-968
973
-1037
10633
32
9*15
-226
706
706
5807
*82 6
3676
-*31
8151
-3621
-787
36*
-3118
679*
1753
831
7*2
-700
5*2
338
1086
667
-2716
-75*
2368
-3689
-1966
*25
-2213
-503
-1200
8*3
-3138
91
61 1
393
-827
-373

-------
Table 3 Migration Fran Each Sub-Region i to Each Sub-Region (j) within ORBES
ft
E





R
R
R
R
R
R
R
R
R
R
R
R
R
R
R
R
G
R
ft
R
R
R
E
6
E
E
E
E
E
F
E
E
E
E
E
E
E
E
I
E
t
E
e
E
G
G
G
G
G
G
G
G
G
G
G
G
G
G
G
G
0
G
G
G
G
G
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
7
N
1
2
3
7
8
1
2
3
4
5
6
7
8
9
1
2
3
4
5
8
0
I
•
15982
7596
2862
4942
2552
708
38
170
364
23 5
221
1003
729
109
130
11
12
13
107
4498
2
2032B
•
3900
1892
2261
1962
1202
•
160
136
290
756
2833
1996
879
34
•
6
•
162
1830
3
116*9
4181
•
2792
1397
284
120
71
90
173
62
24
239
125
6
•
6
•
17
31
1716
7
3926
1819
2412
.
2767
91
208
•
82
11
37
•
32
83
28
•
•
•
5
.
402
8
3231
1573
736
1897
•
13b
744
•
13
77
7
30
113
89
9
•
6
•
•
46
821
11
2553
986
239
70
199
.
265
46
325
1862
328
879
1418
855
286
50
13
T
22
145
1072
12
614
859
82
151
1011,
353
•
46
23
32
76
153
1044
388
183
25
.
•
•
27
238
13
34
•
16
11
•
43
7
•
809
106
268
54
20
92
55
172
189
29
•
299
202
14
6 59
73
120
24
57
494
89
913
•
1077
2372
879
110
587
118
196
64
•
8
209
2360
IS
511
146
149
37
36
1988
31
58
1270
•
248
654
382
312
162
46
26
•
6
459
1507
16
467
222
18
10
17
512
84
288
1474
255
•
504
510
4237
1347
1558
97
25
•
1938
1157
17
455
308
63
6
23
993
177
49
650
435
529
•
1258
2067
270
14
29
27
•
156
325
18
1364
1741
84
42
26
1060
956
•
13
38
455
1300
•
2712
636
13
20
•
•
151
735
19
581
847
86
58
63
853
631
10
417
88
2590
1159
2018
•
2981
181
38
•
26
959
1781
21
108
140
62
•
52
196
223
•
29
24
426
254
383
3323
•
67
35
•
6
1641
927
22
122
50
16
•
39
80
66
105
87
17
1438
14
95
317
102
•
745
138
53
1351
186
23
59
.
•
10
.
30
~
353
76
76
123
18
8
90
60
1523
•
8 78
77
3960
103
2*
•
•
•
7
.
.
•
5
•
•
31
14
•
•
•
46
591
.
221
37 8
13
25
7
16
•
•
•
•
7
•
6
8
7
m
•
34
•
•
19
317
•
141
168
28
68
42
16
•
•
119
107
462
85
69
641
49
181
537
1449
655
1415
295
95
•
741
70
5779
2578
2613
720
1258
1646
325
141
1684
1376
1058
399
1282
3207
1404
148
158
60
169
1351
•
71
1704
332
4^2
14 9
153
625
108
10431
1201
739
1338
318
92
1154
286
1082
2751
1046
94
6577
5396
72
1645
278
480
347
83
493
44
1757
2076
698
1354
182
222
644
375
1084
1852
257
357
3814
3631
73
23 92
467
476
378
255
2617
161
1774
8169
6477
4689
1180
475
1913
925
1517
165
31
7
3451
9332
75
6819
2398
5063
926
1023
5288
411
286
1354
6623
1072
933
2960
2849
547
262
98
8
176
830
25435
77
657
142
308
52
20
504
100
156
722
1361
820
162
246
409
337
870
107
50
15
815
2570
78
272
55
101
59
62
231
35
75
388
268
222
47
14
111
49
190
194
53
•
662
1233
80
456
160
173
119
144
92
39
139
145
153
171
54
•
222
152
126
253
21
25
2291
909
81
595
121
151
70
94
67
50
16
114
39
43
42
48
549
37
114
60
58
10
68 3
629
82
910
85
242
68
132
237
31
92
94
215
25 3
75
26
334
218
201
343
1187
666
2227
1172
83
592
142
215
58
49
74
101
138
95
144
25 2
27
147
286
259
25 5
435
646
660
1482
775
84
J46
115
175
79
40
46
93
193
50
140
43
50
12
64
31
85
463
209
6T
93 5
389
85
161
12
45
•
10
46
7
10
11
65
93
•
48
77
•
•
54
74
32
254
46
87
214
34
34
12
17
6
13
14
53
104
71
28
•
17
32
12
19
•
•
81
151
80
561
89
302
121
162
117
107
20
118
211
107
82
45
150
33
35
3B
125
155
180
752
89
6V
•
•
28
16
22
7
7
28
39
9
•
•
15
88
•
•
•
11
107
62
99
447
146
74
47
5
21
180
46
122
197
106
91
30
167
33
31
45
7
80
357
701
191
415
121
96
78
37
18
52
189
93
116
298
146
36
308
87
821
1059
1246
441
2258
632
152
308
32
33
17
61
34
63
243
52
30
28 5
19
30
135
65
1312
5614
942
196
3735
306
155
293
131
93
48
54
111
212
100
124
103
176
28
lie
239
100
390
1041
1200
279
1955
490
154
111
11
55
25

54
•
15
28
34
26
»
6
105
»
152
254
5 30
7 58
1177
87
153
64
13
15
20
•
24
62
13
.
74
50
•
13
56
50
87
142
152
139
206
371
160
12
.
13
58
29
.
.
.
14
14
13
13
20
6
.
31
16
•
419
36
81
~ Example 15982 people migrated froo sub-region 2 (REG2) to sub-region 1
(under REGION); 20328 people migrated frcm sub-region 1 (REGL)
to cub-region 2 (under REGION).
Periods indicate missing data

-------
Table If; Employment and Demographic Data
Item
Source
Total County Population
County Population Density
Female/Male Ratio, County
Median Age, County
Median Years of Schooling
completed, County (persons
25 yrs. and older)
Median Family Income, County
Percent Unemployment, County
Total Employment, County, 1965
and 1970
County Employment by Sector:
1965, 1970
agricultural services
mining
contract construction
manufacturing
transportation and other
public utilities
wholesale trade
retail trade
finance, insurance,
real estate
services
unclassified establishments
1970 Characteristics of the
Population, U.S. Bureau of the
Census, for Ohio, Indiana,
Illinois, Kentucky, Pennsylvania,
West Virginia (see footnote*)
1970 Characteristics of the
Population*
1970 Characteristics of the
Population*
1970 U.S. Bureau of the Census
Summary Topes, Fourth Count
1970 U.S. Bureau of the Census
Summary Topes, Fourth Count
19 70 Characteristics of the
Population*
1970 Characteristics of the
Population*
County Business Pattern:
1965, 1970
County Business Patterns:
1965, 1970
Distance in miles from centroid
of sub-region i to centroid of
sub-region j
Measured from U.S. Geological
Survey Maps.
* See Table 7 for specific references.
Ik

-------
Table 5 : Characteristics of the Population Data
Item	Table Number
Total Population, County
Table
9
Population Density, County
Table
9
Female/Male Ratio, County
Table
34
Median Family Income, County
Table
124
Percent Unemployment, County
Table
121
State	Volume No.	Part No.
Ohio
Indiana
Illinois
Kentucky
West Virginia
Pennsylvania
All references from U.S. Bureau of the Census, 1970 Census of
Population and Housing, Characteristics of the Population.
16
15
19
50
40
15

-------
Table 6 : Regional Summary Characteristics
1970	1965-/0	1970 Total
Region Density Population Net Migration Employment
1
821.48
2401362
-17178
749856
2
120.58
562908
7218
107677
3
129.55
340292
1967
86973
7
59.68
156084
405
39185
8
169.45
262822
-5675
62488
11
209.96
411334
-7688
111330
12
18.04
8607
-1543
1829
13
86. 84
169960
802
16126
14
59.76
238026
1566
38193
15
69. 39
321028
-2 728
71845
16
134.50
321536
-2035
67933
17
74.48
172597
-155
43631
18
68.09
245445
-4161
53269
19
56. 34
461921
-12696
104372
21
80.52
238532
-4435
44050
22
87.93
151971
-4557
27526
23
47.67
204987
2671
2 7676
24
38.58
96361
-5093
11268
25
33. 32
26666
-1275
2657
28
56.43
475565
-28284
65765
70
195.09
82717
-7747
14624
71
1031.26
1646764
4349
504250
72
348.58
1089278
10079
325439
73
267.13
1174893
27248
352021
75
528.82
1784006
15003
542330
77
119.45
373357
1174
110637
78
113.37
288937
-1616
80499
80
126.64
469021
5347
160413
81
70.39
S4266
-686
9588
82
319.76
1138284
8192
365443
83
114.47
961573
4198
258039
84
79. 31
758484
6355
179998
85
86.46
510761
- 7705
134667
87
76.82
902941
-4894
180154
88
66.81
1672517
7426
430367
89
64.64
217115
3071
40713
99
60.01
423475
-1293
98171
151
808.12
826553
1785
286078
152
117.37
325239
11815
88418
153
54.16
449389
-680
62397
154
54.13
302612
2592
68916
155
49.10
311100
-304
59192
160
49.67
10183
-834
2236
16

-------
Table 6 : (Cont'd)
Region Median
Median Years of
Schooling
Female/Male
Ratio
Median Family
Income
Percent
Unemployment
1
27
11
107.07
9368.8
4.2848
2
32
10
105.36
7507.4
4.9762
3
32
9
104.75
8425.0
4.2307
7
42
11
106.96
7826.7
6.1391
8
22
11
106.44
8098.5
4.9058
11
32
11
106.12
8273.8
4.1372
12
27
9
101.66
- 5320.0
5.6891
13
27
11
103.05
7658.8
4.3549
14
22
11
101.72
7114.4
6.8158
15
27
11
105.32
7890.9
4.9886
16
27
10
105.77
7037.7
6.1722
17
27
9
103.30
6907.0
4.2621
18
32
8
105.82
6102.0
2.1767
19
32
9
103.46
6360.9
5.6962
21
32
10
108.21
6246.7
6.0027
22
27
8
102.36
5682.0
5.3180
23
27
10
101.54
5306.6
5.3316
24
27
8
103.48
4742.5
4.6612
25
27
7
101.64
3942.0
4.8306
28
27
7
103.05
4138.6
7.1904
70
27
8
102.20
11178.0
3.2336
71
54
19
211.58
17924.4
7.9252
72
27
11
104.97
10204.6
5.8109
73
27
11
103.83
9183.5
4.8904
75
27
11
105.22
10313.9
4.8317
77
27
11
103.80
9191.9
3.7125
78
27
11
103.81
9250.0
3.5079
80
27
11
105.54
9938.4
2.9154
81
27
10
104.01
8738.5
3.6634
82
27
11
103.03
10421.7
3.8318
83
27
10
104.43
9310.0
4.6136
84
27
11
106.20
8416.0
4.9628
85
27
10
104.73
7701.3
5.1787
87
32
10
106.16
7958.5
12.2903
88
32
10
106.43
8825.3
9.3776
89
2 7
11
102.85
7559.0
9.4622
99
32
10
103.57
9577.8
6.7640
151
32
10
107.97
9766.7
4.0303
152
32
10
105.41
7535.4
3.4861
153
27
9
101.71
6998.0
5.1107
154
27
8
104.13
6215.8
4.4667
155
32
8
102.63
6184.7
6.5633
160
27
8
115.46
5546.0
6.4542
17

-------
Table 7 ¦ 1970 Regional employment by Sector
iGION
AGftl
MINING
CON
NFC
UTL
WHL
R71
FIN
SER
UNCL
1
94 2
BB76
37808
290851
46457
49825
136471
39888
137593
1145
2
213
10738
6966
38721
7238
4533
21494
3462
13865
334
3
133
1276
292 7
42856
5241
3137
14657
3169
13123
155
7
80
179 3
1354
19938
26 78
1555
5784
872
4877
63
8
57
4640
2609
24710
4507
2792
10835
1917
10255
166
11
144
5677
*219
45781
6214
4153
18822
3806
15354
88
12
0
0
0
398
250
60
243
35
84
0
13
<~7
112
1295
4476
1095
582
5386
882
2078
69
14
65
567
1693
14900
2632
1551
9238
1618
565*
42
15
56
3624
26 5 7
32986
4167
3207
12 76 1
2303
9262
¦95
16
260
497
40 24
26735
4591
406 3
13977
3223
10056
151
17
43
901
3028
18379
2625
2265
7331
1443
6072
118
IB
28
6242
2789
16479
5101
2519
9391
1773
8476
39
19
7T
10U21
72 69
244 94
9991
7066
18 894
5095
16559
202
21
14
15088
1628
4830
2401
2"*05
8251
1864
7114
60
22
0
441
2136
11353
1549
1417
4671
1120
4242
127
23
31
213
961
9144
1490
1386
6775
902
4057
7
24
24
124
623
4628
646
560
2633
482
1251
22
25
19
10
101
627
60
143
644
82
370
-1
28
46
20100
2808
7451
3230
3424
13291
2338
11572
311
70
76
60
679
6166
4 95
68 2
3208
984
2252
22
71
67 6
369
24037
206028
30823
3B321
88538
29202
84414
803
72
457
feci
1 *637
154256
14735
15632
61875
12431
52182
471
73
76/
840
17335
120174
21012
22282
75806
27124
65800
736
75
890
1639
22682
2 70943
29602
2681 1
95 973
19138
74362
833
77
133
797
3659
59056
5669
3906
19613
3737
13619
174
78
240
93
3987
40114
3320
4586
14701
3084
10107
3b
80
31T
225
6682
74735
8924
11320
29187
830B
202 72
260
B1
16
0
416
4051
494
444
2 279
373
1475
0
82
609
630
20500
132942
23318
31875
69808
28800
56095
607
83
440
234
9263
138476
9743
8961
51096
10171
28071
274
84
297
24 90
9015
81121
9129
8699
38550
74 34
21741
308
85
115
598B
6085
52561
7033
7614
25384
5078
21206
204
87
409
4V86
V676
67821
11765
7770
37676
8578
30705
3*8
88
891
163$
12417
151434
25493
26762
94245
26881
77084
565
89
30
2772
1917
14058
2687
1756
9250
1772
5682
42
99
307
738
5521
39601
5225
4877
22 915
3303
13920
6 1
151
478
467
13660
121092
17160
20866
49637
16269
45910
*71
152
836
141
6558
29366
5041
5250
17635
5376
16824
166
153
99
263
3014
21223
3495
3560
14672
2565
8291
68
154
13*
3181
4626
28079
3219
3412
13899
2695
8 794
103
155
246
V79
3387
20896
3709
4135
12718
3015
8990
1*4
160
0
0
61
753
73
353
592
95
304
5
Key
AGRI
MINING.
CON-
MFC
UTI
KM L.:
RTL.
Agriculture
Mining
Construction
Manufacturing
Utllitles
Wholesale
Retail
FIN Tinances and Real Estate
SER Services
UNCL Unclassified

-------
Table 8 : 1967 Regional Production, Non-Energy Sectors (Gilmore, 1979)
SIC CODE
ASSIGNMENT
SECTOR DESCRIPTION
TOTAL
(millions of
OUTPUT
'67 dollars)
H
VO
15,16,17
20-29
201
20 excl. 201
23	excl. 239
239
24
261,264,266,27
262
263
265
281
287,289
282
295
30
32
331
332
33	excl
336
34
35
36
371
37	excl. 371
39
41,42,44,45,48
41,48
42 ,421 ,4211
331, 332, 336
CONSTRUCTION
MANUFACTURING
meat products
food excluding meat products
apparel and misc. textile products
misc. fabricated textile products
logging and misc. wood products
misc. paper products and publications
paper mills
paperboard mills
paperboard containers
industrial organic/inorganic chemicals
agricultural and misc. chemicals
plastic and synthetic resins
paving and asphalts
rubber and misc. plastic products
glass, stone, and clay products
blast and basic steel products
iron and steel foundries and forging
other primary metal manufacturing
nonferrous forge, cost, and rolling
metal containers
and farm machinery
equipment and components
and auto manufacturing
transportation equipment
manufacturing
fabricated
industrial
electrical
truck, bus
misc,
misc.
TOTAL MANUFACTURING OUTPUT
TRANSPORTATION
misc. transportation and communication
motor freight transportation
8714
1795
8741
769
1508
1601
2828
342
173
685
2153
2403
1119
134
2492
2933
8820
1398
516
2595
4954
7815
7800
3930
2609
1568
71688
2766
2039

-------
Table 8 : (cont'd)
SIC CODE
ASSIGNMENT
SECTOR DESCRIPTION
TOTAL OUTPUT
(millions of '67 dollars)
44
45
50-51
52-59
60-67
60-64,6,67
65
70,72,73,75,80,82,86
70,79
72,73 excl. 731
731
75
80 ,82
86
water transportation
air transportation
TOTAL TRANSPORTATION OUTPUT
WHOLESALE TRADE
RETAIL TRADE
FINANCE
finance and insurance
real estate
TOTAL FINANCE OUTPUT
SERVICES
hotels, lodging and amusements
misc. business and personal services
advertising
auto repair
medical and educational services
nonprofit organizations
236
152
5193
6050
10289
4132
4132
7828
11960
2388
3063
980
1325
3638
1389
TOTAL SERVICES OUTPUT
12783

-------
Table 9 : 1967 Employment/Output Ratios
(No. of Employees/Millions of 1967
SIC CODE	TOTAL ORBES
RANGE	SECTOR CATEGORY	EMPLOYMENT
15,17	CONSTRUCTION	319,205
20-39	MANUFACTURING	2,2 70,69 3
41-48	TRANSPORTATION	246,917
ro
H 50-51	WHOLESALE TRADE	337,979
52-59	RETAIL TRADE	1,059,762
60-68	FINANCE	273,924
70-86	SERVICES	749,594
Dollars)
TOTAL ORBES OUTPUT
(millions of 1967 dollars)
8714
71688
5193
6050
10289
11960
12783
EMPLOYMENT/
OUTPUT RATIO
36.631282
31.674659
47.548045
55.864297
102.99951
22.903344
58.639912

-------
4.0	Derivation of Migration Models
Given the basic data and theoretical approach described in
sections 2.0 and 3.0jtwo migration models were derived. The first
utilized employment by sector along with the other population and
distance variables as the independent explanatory variables for
migration. In using this model for future projections, some major
problems arose because of the nature of the data base. Thus, a second
model was derived which substituted unemployment estimates for the
employment variables. Each of these models is discussed in turn below.
4.1	Employment Based Migration Model
The first model tested in this study utilized a migration
equation with the following general form:
E<1. 1
My = f (TE^ TEj, Eli - E10i, El.. - ElOj, D , A ,
ED., ED., FMR., FMR.)
i j	y
where: M. . = total number migrants from i to j
i J
TE. ("EE.) = total employment in sub-region i (j)
J
El. - E10 (El. - E10.) = employment in sub-region i (j)
i	i J	J
for each of ten sectors.
El:
agricultural services
E2:
mining
E3:
contract construction
Ek:
manufacturing
E5:
transportation
E6:
wholesale trade
E7:
retail trade
E8:
finance, insurance, real estate
E9:
services
E10: unclassified establishments
D.. = distance in miles between centroids of i and j
A. (A.) = median age of i (j)
J
ED. (ED.) = median years of schooling completed in i (j)
FMR. (FMR.) = female to male ratio in i (j)
J
22

-------
The actual equations were derived using linear regression
analysis, assuming a non-linear relationship between the dependent
variable, migration, and all independent variables. A logarithmic
transformation of the ARC migration data and the employment/
demographic/distance data gave excellent empirical results, yielding
a migration equation with a coefficient of determination (R^) of
0.968. This means that almost 97$ of the migration changes were
explained by the independent variables chosen. The actual equation
derived was:
Eq. 2
log10 M.d = 0.U7205030 log10 TE. + 0.59835792 log10 TE.
-0.6525517 log1Q E3. - 0.17851250 log 1Q E4.
-0.31055967 log1Q E5. - 0.92875838 log1Q E6.
-0.40063660 log 1Q E7i + 0.66509703 log1Q E10i
-0.089^3625 log10 E3j - 0.16063^57 log10 Ek.
-1.10771851 lob1Q E6. + 1.02603079 log1Q EQ.
+0.43^20710 log E10. - 1.93003811 log D..
-Lv/	J	-lAJ 1J
+0.84364045 log1Q A. - 1.43443765 log1Q Ed.
+2.37172640 log10 FMR. + 1.58257969 log1Q A.
-1.03546675 log1Q ED^ - 2.21746006 log1Q FMR^.
and M.. = anti-log (log.,- M. .).
10 ij'
In this model, then, estimates of future migration between sub-regions
were obtained by estimating total employment, TE, for each sub-region
as well as employment for individual sectors, E1-E10.
It is unclear what effect, if any, such phenomena as the post
World War II baby boom will have on future levels of employment; it
has not been studied in the migration literature and no data are
currently available with which to model "its impact on employment
in the future. Thus demographic characteristics such as median age,
years of schooling completed, and female/male ratio were assumed
constant overtime and taken from the original 1970 data file.
23

-------
As we discuss in detail in section k.3, this model did not work
well in projecting migration to the year 2000, The reasons for this
appear to be related to the state of the economy in 1970, particularly
in the mining sector. Thus, we derived an alternative empirical
migration model.
k.2 Unemployment Based Migration Model
The second migration model shows the relationship between
unemployment rates, the population and distance variables, and
migration. The general equation is as follows:
3
M. . = f(UNEMP., UNEMP., TE , TE., MFI., MFI., D..)
ij	i	J i J i j' ij'
where: M.. = total number migrants from sub-region i to j
UNEMP. (UNEMP.) = percent unemployment in i (j)
TE. (TE.) = total employment in i (j)
J
MFI. (MFI.) = median family income in i (j)
1 J
D.. = distance in miles between centroids of i and j
The specific regression results are shown as Table 10. Here, we
see that again we have obtained a very high indicating good
empirical results. The major problem posed by this equation is one
of estimating future unemployment rates. Although we cannot do so
with a high degree of accuracy, we found that the 0RBES scenarios
gave us enough information to make estimates suitable for comparing
scenario by scenario differences in migration impacts.
^.3 Application of the Migration Equations to 0RBES
Our application of the first migration equation, that based on
employment, involved a number of steps. These are illustrated in
Figure 1. We began with output estimates taken from the regional
1/0 model for the year 2000. Our employment/output ratios were then
used to derive the year 2000 QRBES employment estimates by industry.
This can be thought of as the first constraint on our model.
Constraints on minimum county level employment in mining, construction
and utilities were also derived using information from the coal supply
and allocation model and the power plant siting model.
2k

-------
The next steps involved, slimming the constraints to sub-regional
and. regional levels. The base year (1970) employment totals were used
to calculate the change in employment by sector (A employment) for
the region. For those industries which were not constrained In their
subregional location by the ORBES scenarios (i.e. everything but
mining, construction, and utilities) we allocated the new employment
in proportion to the share of employment present in 1970. Thus, we
implicitly assumed no shift occured in the location of employment
in these industries.
For mining, construction, and utilities, we allocated employment
to the sub-regions based on the shares as constrained by the
aggregated sub-regional totals. This completed the employment data
needed for the year 2000 in order to run the regression model.
Next, we read in the remaining regression model data on
distances and demographic characteristics and obtained an estimate
of migration between 1970 and 2000.
Using this process, we estimated migration in ORBES for each
scenario. Unfortunately, we found a grave inconsistency in model
results. In those regions with high projected increases in mining
employment, the mod.21 was predicting heavy out-migration. This is
the opposite of what we would expect. In fact, we would expect
those sub-regions to have lower out-migration, net migration near
zero, or net in-migration relative to the base period. Looking
back at our original data we discovered that the reason for the
discrepancy was the unemployment level in mining in 1970. At that
time, a large number of miners were out of work because of depressed
market conditions. Thus, those regions with high mining employment
also had a large number of unemployed which in turn brought about
net out-migration.
The model we derived based on employment thus predicted that
as the proportion of mining employment increased, so did net out-
migration. However, the market conditions assumed by our scenarios
in the year 2000 are not those that existed in 1970. As a result,
we were forced to derive a second migration model linked to unemploy-
ment rates in order to avoid this problem. Although this unemployment
based model is not as well linked to the ORBES I/O work it was the
only way we could see to get around the empirical problems caused by
the employment based model.
The unemployment based equation was found to be a much more
consistent estimator of migration trends in the region. Table 11
shows the actual vs. estimated net migration for the region in 1970.
Here, one can see that the model projects the direction and order of
25

-------
magnitude of migration correctly for most of the migration regions.
However, there are a significant number of regions for which an
incorrect estimate is made. The reason for this discrepancy is
related to the non-linear form of the equation. Although the linearly
transformed equation gives a regression estimate with an R2 of .95,
the logarithmic form of the equation means the transformation back to
non-linear form expands the errors of the model exponentially. This
is not a problem unique to our migration model but instead is one
common to all similar models reported in the literature.
Another problem with the unemployment model lies in the
prediction of unemployment and income for each subregion. Here, we had
to make estimates of changes based on projected mining and other
employment changes. These estimates axe made rather arbitrarily since
we have no regional unemployment and income model. However, they
should still be accurate enough to allow the comparison of migration
impacts of various scenarios.
Given these problems, we do not have very great confidence in the
numeric predictions from our migration model. However, we do believe
that the general direction and magnitude of migration predicted is
adequate for a comparison of the migration impacts of various economic
and other conditions ^elated to the QRBES scenarios. Our final
chapter makes these comparisons.
26

-------
Table 10
GENERAL LINEAR HOUELS PROCEDURE
DfcPeNDLNT VARIABLES	IQtALHlG
iUUHCt	OH
MUUEL	7
fcRCOK	1750
um.UKKEC.1ED TOTAL	1757
SUJHCE
PCUNtMPl
PCUNfcHPJ
KFI1
flFU
IU1EI
lUltJ
UlSlANCt
OF
1
1
1
1
1
1
1
SUM OF SUUAKES
9699.08351808
418*07903251
10117.16315059
TYPE 1 SS
8754.76280087
154.75174642
303.87398867
18.66008841
57.49502624
40.702 56952
368.83f29795
MEAN
sguAKt
F VALUE
PR > F
R-SQUAKt
L. V.
1385.58335973
5799.78
0.0001
0.95867b
21.3893
0.2389O265

STD OEV
TUTALHJb HFAn



0.<>8877668

2.28514984
F VALUE
PR > F
OF
TYPE IV SS
F VALUE
PR > F
36645.73
0.0001
1
0.75009576
3.14
0.0766
647.76
O.OOOl
1
0.74962585
3.14
0.076/
1271.96
O.OoOl
1
0.6429506 3
2.69
0. 1011
78.11
0.0001
1
If.38007199
72.75
0.0301
240.66
O.OOOl
1
18.24870064
76. 3V
0.0001
170.37
0.0001
1
15.08975823
63.16
0.0001
1543.88
0.0001
1
368.83729795
154^.i>8
0.0001
ro
-o
PARAMETER
PCUNcMPl
»»CUNfcMPj
HFll
MF1J
10TE1
TOTfcJ
Ulif ANCL
ESTIMATE
0.15719813
0.15821303
O. 18323816
0.95011305
0.21218852
0.1V240299
-1.85650032
T FOR H0<
PARAHET ER-0
1.77
1.77
1.64
8.53
8.74
7.95
-39.29
PR > IT I
0.0766
0.0767
0.1011
0.0001
0.0001
0.0001
0.0001
STO ERROR OF
ESTIMATE
0.08871550
0.08935012
0.11169599
0.11139353
0.02427B19
0.02420924
0.04724851

-------
Table 11: Actual and Estimated Migration For 1970
KEGiUN
AC TU4L
ESriMA TE
1
-1 71 /a
2078
2
7^itJ
-1129
3
1^07
-162
7
40:>
-44 3
8
-56 75
-9 7
li
- 76 88
161
12
-15*3
-732
13
802
-422
14
1566
-962
13
— ll 28
-813
16
-20 35
-549
17
-155
-1072
Id
-4lol
-2118
19
-12696
-1703
21
— 44 35
-1148
22
-45 5/
-2 56 2
23
2671
-4451
24
-5093
-1880
25
-1275
-870
28
-2 8284
-4557
70
- 77h7
3028
71
4349
3469
7
-7705
-213
8 7
-4894
-387
88
74^o
5 7
89
3070
44
9V
-1293
726
151
1785
4247
152
11815
1787
153
-OtfO
-1274
154
25^2
-136V
155
-304
-134a
lbO
-834
-251
28

-------
5.0 Migration Impacts of the ORBES Scenarios
Given our migration model, we proceeded to simulate the migration
impacts of various changes or shifts in regional employment and the
impact of scenario differences. The regional employment shifts were
based on the 1965-70 rates shown in Table 12.
In order to estimate unemployment and median family income, we
utilized changes in employment as an indicator. We simulated
migration using several rate differences and decision criteria but
only report same representative ones here.
In the first simulation we used 1965-70 shifts in manufacturing
employment as our indicator for change. Scenario 1 (Business As Usual
or BAU) was used as our backdrop relative to total regional employment
calculated using our employment/output ratios. Then we calculated
the total sub-region employment in the year 2000 if manufacturing
continued to shift from one part of the region to another at 1965-70
rates. The manufacturing employment in the year 2000 was calculated
and used as an indicator of unemployment changes. Unemployment levels
were initially set to 5.5$. Arbitrarily, we said that if manufacturing
employment increased by over 1000 then unemployment would be reduced
to 2.5% and median family income would increase to $10,000 (1970"
dollars). If manufacturing employment decreased, then unemployment
shifted to 6.5$ and median family income dropped to $9,000. All other
sectors were assumed stationary for this simulation run. The results
of the simulation are shown in Figure 3.
A shift in manufacturing employment at the 1965-70 rate appears
to result in a shift of population away from most of the major
population areas to smaller urban areas and to rural regions. The
exceptions to this are the Indianapolis, Indiana and Lexington,
Kentucky regions which are still forecast to have net inmigrants.
This finding seems consistent with recent urban-rural migration trends,
reports of older industries in urban areas closing and of new
industries in less populated areas opening. Examples include the
closing of Youngstown Sheet and Tube and U.S. Steel in Youngstown,
the building of a new Volkswagen assembly plant in New Stanton,
Pennsylvania and the plans for a major steel facility in Conneaut,
Ohio. Should this trend continue, the implication for ORBES is that
changes in population related to energy growth will be reinforced by
changes in the location of manufacturing concerns. Thus the combined
impacts may in fact be larger than anticipated. However, these
impacts may be more easily ameliorated than otherwise might be the
case because growth in some areas will be stable.
29

-------
A second, simulation, using the same methods for determining
unemployment and income levels, was conducted for the construction
industry. Figure 1+ shows the migration forecast. There are several
differences from the manufacturing simulation. The Cincinnati area
is expected to have net inmigration rather than net outmigration.
Similarly Portsmouth, Ohio, Central Illinois, Northwestern Pennsylvania,
Southern West Virginia, and the area south of South Bend, Indiana,
will all have a reversal in migration. This implies that, historically,
construction unrelated to manufacturing has been occuring in these
areas and has induced inmigration.
Figures 5 and 6 show similar distributions using services and
finance sectors respectively as the forecasting variables. Here
again, there are minor differences but no major changes. Table 13
summarizes the differences between figures 3j^»5 and 6.
What these results indicate is that a general shift of
population away from major metropolitan areas to rural areas has been
occurring in the recent past in conjunction with shifts in employment.
If these shifts continue this same pattern of internal migration will
result in the future and may have some affect on the direct population
impacts of coal mines and power plants.
Another set of simulations was made to compare the net migration
impacts of various scenarios. Although the model predicts gross
migration, net figures are reported in order to simplify the results
for discussion. Scenarios 1, U, and 5 were selected because of their
differences in terms of economic assumptions. Essentially these
model runs assumed that all sectors except mining would shift at the
1965-70 rate. Mining employment was projected based on scenario
projections of the amount and location of new mining employment by
county. For these runs, unemployment rates were initially set high
at 7.5$ to be consistent with higher levels in mining areas. These
were then summed to the subregional level. If mining employment
increased by over 1000, unemployment levels dropped to 2.5 and income
increased to $10,000. The result of these simulations show the
impacts of scenario-based mining projections on subregional migration.
The results of these runs are given as Figures 7, 8, and 9, and
Table lU summarizes their differences. Here, one can see that the
migration model is insensitive to scenario differences in mining at
this geographic scale with only region 8 in Pennsylvania falling into
a different net migration class in scenario 5 vs. 1 and if. The reasons
for this are because mining is concentrated in the same portions of
the 0RBES region regardless of scenario and all scenarios produce
significant increases in mining employment. Comparing these figures
to Figure 3, one can see a significant shift in population toward
mining subregions brought about by these assumptions. This is
consistent with what one would expect to occur when these relatively
rural, low population areas require major increases in labor force.
30

-------
Although there are no major scenario differences in migration
impacts according to our model, there are significant shifts in
population predicted as a result of potential shifts in employment in
the major sectors such as manufacturing and mining. These shifts, in
turn, imply changes in the distribution of point and non-point sources
of air and water pollution. Current ORBES efforts do not allow the
estimation of the impacts of these shifts on pollutant levels hut
future research efforts could do so.
In summary, the migration model which we operationalized
demonstrates several trends that are of relevance to the ORBES
assessment. First, the continuation of internal migration trends
into the future may result in a shift of pollutant sources away from
many major metropolitan areas to more rural areas. These trends would
particularly exacerbate the air pollution problems associated with the
location of power plants in the study region. It is important to note
that in many cases population is predicted to shift to the same areas
where utilities have scheduled plant additions and where the ORBES
siting model has allocated "conjured" plants.
The second result of importance to the ORBES analysis is the
predicted shift of population toward the mining regions of ORBES.
Although the predicted net migration at the scale of our ^3 regions
is low, net migration at the county or community scale can be expected
to be quite high. This implies in turn, potential problems associated
with the provision of public and health services.
Finally, the combination of these trends points to the potential
for synergistic impacts on the physical, social, and economic
environment as people, power plants, mines, and industries concentrate
in previously low population areas of the region. Our results thus
point to a number of secondary impact analyses which could be
performed in order to assess the magnitude of those synergistic effects.
31

-------
Table 12
HtRLtNT CHANGt IN EHHLOYMtHI rfllHIN fcACH KLulON FHOM 1965 TO 1970 FOR VARIOUS SEL1UKS
CUMblNtb
HHOLfcSALt
Rb1A1l
KEG1UI
NININO
LUN^IRULTIQN
MANUFACIUHINb
WnOLESALE
RETAIL
FINANCIAL
FlNANL 1AL
bE KV1LE 5
1
0.1
—0.2
-1.3
-0.1
-0.1
-0.5
-0.7
-o.a
2
-0.2
0.8
-0.1
-0.0
-0.1
-0.2
-0.2
-0.0
3
-0.1
0.1
-0.2
-0.1
-0.1
-0.0
-o. z
0.0
I
-0.5
-0.1
0.1
-0.0
-0.1
-0.1
-0.2
-0.0
8
0.0
0.2
-0.1
-0.0
-0.1
-0.1
-0. 2
-0.0
11
1.0
-1.2
-0.3
-0.2
-0.0
-0.1
-0.3
-0.1
12
-0.3
0.0
-0.0
-0.0
-0.0
0.0
-O.b
-0.0
13
0.1
0.1
0.0
-0.0
0.0
0.0
0. 1
0.0
14
-0.3
0.0
0.1
0.0
-o.o
0.1
0.0
0.0
15
0.7
0.1
0.1
-0.0
-0.1
0.1
o.c
0.0
16
-0.3
-0.1
-0.1
-0.2
-0.1
-0.1
-0.4
0.0
IT
-0.1
0.3
0.0
0.0
-0.1
0.0
-o.o
0.0
18
—0.8
0.3
-0.0
-0.1
-0.1
-0.0
-0. 3
-0.1
19
-0.1
0.8
-0.3
0.0
-0.2
-0.0
-0. 1
-0.1
21
2.6
0.0
0.0
-0.1
-0.1
-0.0
-0.2
-o.o
22
-0.1
-0.0
0.0
0.1
-0.0
0.0
0. 1
0.1
23
-0.1
-0.1
0.1
0.1
0.1
0.0
0. 1
-0.0
24
-0.0
0.0
0.1
-o.o
0.0
0.0
0.0
-0.0
25
-0.0
0.0
-0.0
0.0
-0.0
0.0
0.0
0.0
28
-0.2
-0.0
0.1
0.1
0.0
0.0
0.1
0.2
70
-0.0
0.0
0.1
0.1
0.0
0.0
0. 1
-0.0
71
0.0
0.6
0.2
0.7
-0.0
-0.2
0.5
0.2
T2
0.1
-0.8
0.3
-0.3
0.4
0.2
0.3
0.1
T3
-0.1
-0.2
0.1
-o.o
0.6
0.6
1.2
0.5
79
-0.2
-0.1
-0.1
0.1
0.2
0.1
0.4
-0 .i
77
-0.0
0.0
0.1
-0.0
0.1
-0.0
0.0
-o.o
78
-0.0
0.2
0.2
0.1
-0.1
-0.0
-o.o
-0.0
80
0.1
0.1
0.2
0.0
0.1
0.1
0.2
0.1
81
-0.0
0.0
0.0
0.0
-0.0
0.0
0. 0
-0.0
82
0.1
0.*
—0 .1
0.4
0.3
0.1
0.7
0.2
83
-0.0
—0.3
-0.3
-0.3
-0.1
-0.2
-0.6
-0.1
80
-0.1
0.5
-0.1
-0.1
-0.1
0.0
-0. 1
0.1
83
-2.1
-0.7
0.1
-0.0
-0.1
-0. 1
-O.i
0.1
87
0.8
-0.5
-0.2
-0.3
-0.2
-0.1
-0.6
0.3
88
—0.3
-0.3
-0.0
-0.3
-0.3
0.1
-0.6
-0.1
69
0.1
-0.1
0.1
-0.0
0.0
0.0
0.0
-0.0
99
-0.1
0.5
-0.0
-0.0
0.1
0.0
0. 1
0.0
151
0.0
-0.8
0.7
0.1
0.2
0.0
0.3
-0.1
152
0.0
-0.0
0.3
0.1
-0.0
0.1
0.2
-O.I
153
0.0
0.1
0.2
0.1
0.0
0.0
0.2
0.0
15*
0.6
0.3
0.2
0.2
0.0
0.1
0.3
O.i
155
-0.0
-0.1
0.1
0.1
-0.1
0.0
-0.0
0.1
160
0.0
-0.0
0.0
-0.0
0.0
-0.0
-0.0
-0.0

-------
uo
IjO
OUTSIDE 0R3E
LESS THAN -1000.0
FIGURE 3
0RBES REGION
SCENARIO NO. 1 : MANUFACTURING SHIFT
-1000.0 THRU -1.0
0.0 THRU 750.0
CHESTER THAN 750.0

-------
FIGURE 4
QRBES RrGION
SCENARIO NO. 1 : CONSTRUCTION SHIFT
00
¦t=-
OUTSIDE ORBES REGION
LESS THfiN -1000.0
-1000.0 THRU -1.0
0.0 THRU 750.0
GREATER THAN 750.0

-------
ORBhS
SCENARIO NO. 1
JRE 5
REGION
: SERVICE SHIFT
u>
vn
OUTSIDE ORBES REGION
LESS THAN -1000.0
-1000.0 THRU -1.0
0.0 THRU 750.0
GREfiTER THAN 750.0

-------
FIGURE 6
0RBES REGION
SCENARIO NO. 1 : FINANCIAL SHIFT
u>
(T\
OUTSIDE ORBES REGION
LESS THBN -1000.0
-1000.0 THRU -1.0
0.0 THRU 750.0
GRERTER THAN 750.0

-------
Table 13: Sub-Region Net Migration Impacts for Scenario 1
A Summary of Figures 3,4,5, and 6
Shifts Due To:
OBS
REGION
Manufacturi ng
Cons truction
Servi ces
Fi nances
1
1
2
2
2
2
2
2
3
3
3
3
3
3
2
3
3
3
4
7
3
2
3
2
5
8
2
3
3
2
6
11
2
2
2
3
7
12
2
2
2
3
8
13
3
3
3
3
9
14
3
3
3
3
10
15
3
3
3
3
11
16
1
3
3
2
12
17
3
3
3
3
13
18
3
3
3
2
14
19
2
3
1
3
15
21
3
3
3
2
16
22
4
3
3
3
17
23
3
2
3
3
18
24
3
3
3
3
19
25
3
2
3
2
20
28
3
3
2
3
21
70
3
3
3
3
22
71
2
3
2
2
23
72
2
2
2
2
24
73
2
2
2
2
25
75
2
2
2
2
26
77
3
3
3
3
27
78
3
3
3
3
28
80
3
3
2
2
29
81
2
3
2
3
30
82
3
3
2
2
31
83
2
2
2
2
32
84
3
3
3
3
33
85
2
2
2
2
34
87
2
2
2
2
35
88
2
3
2
2
36
89
3
3
3
3
37
99
3
3
3
3
38
151
2
2
2
2
39
152
3
3
3
3
40
153
3
3
3
3
41
154
3
3
3
3
42
1 55
3
3
3
3
43
160
2
3
3
2
Code:
Net Migrants



1:
more than
¦1000 3:
0 thru 750


2:
-lOOO thru
-1
greater than 750





37

-------
F IGUR E 7
ORBES REGION
SCENARIO NO. 1 : NET MIGRATION
-------
F I G U R E 8
ORBES REGION
SCENARIO NO. 4 : NET MIGRATION
UJ
VO
OUTSIDE ORBES REGION
LESS THAN -1000.0
-1000.0 THRU -1.0
0.0 THRU 750.0
GREATER THAN 750.0
-------
F IGUR E 9
0RBES REGION
SCENARIO NO. 5 : NET MIGRATION
-p-
o
OUTSIDE ORBES REGION
LESS THRN -1000.0
-1000.0 THRU -1.0
0.0 THRU 750.0
GREATER THAN 750.0
-------
Table 14: Sub-Region Net Migration Impacts for Scenarios 1, 4, and 5
A Summary of Figures 7, 8, and 9
Scenari o
REGION
ONE
FOUR
FIVE
1
2
2
3
2
3
3
3
3
3
3
3
7
3
2
3
8
3
3
2
11
3
3
3
12
2
2
3
13
2
2
2
14
4
4
4
15
3
3
3
16
1
1
1
17
1
1
1
18
3
3
3
19
3
3
3
21
3
3
3
22
1
1
1
23
1
1
1
24
2
2
2
25
2
2
2
28
4
4
4
70
4
4
3
71
3
3
3
72
4
4
4
73
2
2
2
75
3
3
3
77
2
2
2
78
3
3
3
80
3
3
3
81
3
3
3
82
4
4
4
83
2
2
2
84
3
3
3
85
3
3
3
87
3
3
3
88
3
3
3
89
3
3
3
99
3
3
3
151
4
4
4
152
3
3
3
153
2
2
2
154
3
3
3
155
2
2
2
160
2
2
2
Code:	Net Migrants
1:	more than -1000
2:	-1000 thru -1
3:	0 thru 750
4:	greater than 750
-------
References
1)	Borts, G.H. and Stein, J.L., Economic Growth in a Free
Market, (New York: Columbia University i-ress, 1964).
2)	Muth, R.F., "Differential Growth Among Large U.S. Cities,"
in Papers in Quantitative Economics, edited by J.P. Quirk
and A.M. Zarley (Lawrence: The University Press of Kansas,
1968) pp. 311-355. Also see: "Migration: Chicken or Egg,"
Southern Econ. J., Jan. 1971, 37(3), pp. 295-306.
3)	Lowry, I.S., Migration and Metropolitan Growth: Two Analy-
tical Models (San Francisco: Chandler Publishing Company,
1960).	
4)	Miller, E., "Is Out-Migration Affected by Economic Conditions?"
Southern Econ. J., Jan. 1973, 39(3), pp. 396-405.
5)	Santini, Danilo J., "Lowry versus Muth: The Chicken or Egg
Question Reexamined." Mimeographed.
6)	Perloff, H.S., Dunn, E.S. Jr., Lampard, E.E. and Muth, R.F.,
Regions, Resources, and Economic Growth (Baltimore: Johns
Hopkins Press, 1960).
7)	Gallaway, L.E., "Age and Labor Mobility in The United States,
1957-1960", Social Security Administration Office, Office of
Research and Statistics Research Report No. 28., Washington,
D.C.: U.S.G.P.O., 1969.
8)	Schwartz, A., "Interpreting the Effect of Distance on
Migration", J. Polit. Econ., Sept./Oct., 1973, 81(5), pp. 1153-1169
9)	Suval, E.M. and Hamilton, C.H., "Some New Evidence on
Educational Selectivity in Migration to and from the South",
Social Forces, May 1965, 43(4), pp. 536-547.
10)	Kain, J.F. and Persky, J.J., "The North's Stake in Southern
Rural Poverty", in J.F. Kain and J.R. Meyer (eds.). Essays
in Regional Economics (Cambridge: Harvard University Press,
1971), pP7 243-278.—
11)	Greenwood, Michael J., "Research on Internal Migration in
The United States: A Survey", Journal of Economic Literature,
13: pp. 397-433.
12)	. "A Simultaneous - Equations Model of Urban
Growth and Migration", Journal of the American Statistical
Association, Dec. 1975, No. 352, pp. 79 7-810.
13. Gilmore, Douglas, Input/Output Model for ORBES, Dept. of
Economics, University of Illinois at Urbana-Champaign, 1979.
43
-------
Appendix A
*
COuNlY NAMES AND FIPS CODES FOR EACH MIGRATION RfcGION
REGION
FlPi
COUNTY
1
42003
ALLEGHENY
1
42007
tilt AVER
1
421*15
WASHINGTON
1
42129
WESTMORELAND
2
42005
ARMb1KONG
2
420A9
SUTLER
z
420 bl
FAYETTE
2
42059
GREEN
2
420t>3
INDIANA
2
54061
MONONGALIA
2
54077
PRESTON
3
42031
CLARION
3
420 53
FOREST
3
42073
LANKfcNCE
3
42085
MfckCER
3
42121
VENANGO
7
420 33
CLEARFIELD
7
420 47
ELK
7
42065
JEFFERSON
8
420*1
CAM6RIA
B
42111
SOMERSET
11
39013
bELMONT
11
39067
HARRISON
11
39061
JEFFERSON
11
39111
MONROE
11
5)4009
BRGOKk
11
54029
HANCOCK
11
54051
MARSHALL
il
5406V
OHIO
11
540 95
TYLER
11
54103
WETZEL
12
54023
GRANT
13
39001
ADAMS
13
39015
BROWN
13
390 25
CLERMONT
13
39071
HIGHLAND
14
39009
ATHENS
14
39053
GALLIA
It
39073
HOCKING
14
39079
JACKbON
14
39105
MEIGS
14
3V131
PIKE
14
391*1
ROSS
14
39163
VINTON
15
3V019
CARROLL
15
39031
COSHOCTON
15
390 59
GUERNSEY
15
39075
HOLMES
15
39115
MORGAN
15
39119
MuSKINGUM
15
39121
NOtJLfc
15
39127
PERRY
15
39157
TUSCARAWAS
16
39087
LAWRENCE

U4

-------
Appendix (Con't)
COUNTY NAMES AND FIPS COOES FOR EACH MIGRATION REGION
REGION
FIPS
COUNTY
16
391*5
SCIOTO
lt>
5*011
CABELL
16
5*0*3
LINCOLN
16
5*05 3
MASON
16
54099
WAYNE
11
39167
WASHINGTON
17
5*013
CALHOUN
17
5*073
PLEASANTS
17
5*085
RITCHIE
17
54105
WIRT
17
54107
WOOD
18
54001
BARBOUR
lb
54017
DODDIDGE
16
54021
GlLMMER
ie
54 033
HARRISON
le
54041
LtWlS
18
54049
MARION
ie
5408 3
RANDOLPH
18
54 091
TAYLOR
18
54093
TUCKEK
18
5409?
UPSHUR
19
54 005
BOONE
19
54007
BRAXTON
19
54015
CLAY
19
54019
FAYETTE
19
54025
GREENbRIER
19
54035
JACKSON
19
54039
KANAWHA
19
54067
NICHOLAS
19
54075
POCAHONTAS
19
54079
PUTNAM
19
54 087
ROANE
19
54101
WtBSTER
21
54047
MCDOWtLL
21
54055
MERCER
21
54063
MGNRE
21
54081
RALEIGh
21
54089
SUMMERS
21
54109
WYOMING
22
21019
BOYD
22
21043
CARTER
22
21063
Elliott
22
21089
GKEENUP
22
21115
JOHNSON
22
21127
LAWRENCE
22
21135
LEWIS
23
21011
BATH
23
21049
CLARK
23
21065
ESTILL
23
2iC69
FLEMING
23
2107V
GARRARD
23
21109
JACKSON
23
21129
LEE
23
21137
LINCOLN

-------
ADDendix (Con't)
COUNTY NAMES and FIPS CODti FOK
REGION FIP S
23
21151
23
21165
23
21173
23
2 1197
23
21203
23
212 05
23
21237
24
21001
24
21045
24
210 67
24
2U99
24
212 07
24
212 31
25
21053
25
210 57
25
21171
28
21013
28
21025
28
^1051
28
21071
28
21095
28
21119
28
21121
28
211 25
28
21131
28
21133
28
21147
28
21153
28
21159
28
21175
28
2118V
26
21193
28
21195
26
212 35
28
5*045
26
54059
70
39103
71
16029
71
18047
71
18115
71
16137
71
18155
71
21015
71
21023
71
210 37
71
210 77
71
21081
71
llll?
71
21161
71
21191
71
39017
71
39027
71
39061
71
3 9165

EACH MIGRATION REGION
COUNTY
MAOISON
MENIFEE
MONTGOMERY
PGnEIL
ROCKCASTLE
ROWAN
WOLFE
AOAlK
CASEY
GREEN
PULASKI
RUSSELL
WAYNE
Clinton
CUMBERLAND
MONROE
BELL
BREATHITT
CLAY
FLOYD
HARLAN
KNOTT
KNOX
LAUREL
LESLIE
LETCHER
MCCREARY
MAGOFFIN
MARTIN
MORGAN
OWSLEY
PERKY
PIKE
WHITLEY
LOGAN
MINGO
MEDINA
DEARBORN
FRANKLIN
OHIO
KIPLfcY
SWITZERLAND
BOONE
BRACKEN
CAMPBELL
GALLLATIN
GRANT
KENTON
MA SUN
PENDLETON
BOTLtR
CLINTON
HAMILTON
WARREN
-------
ADDendix (Con't)
COUNT V NAMES AND FIPS CODES FOR EACH MIGRATION REGION
REGION
FIPS
COUNTY
It
39021
CHAMPAIGN
11
39023
CLARK
12
39037
DARKE
11
39 0 5 7
GREEN*
11
39109
MIAMI
It
39113
MONTGOMERY
It
39135
PREBLE
73
39041
DELAWARE
73
390*5
FAIRFIELD
73
390*7
fayette
73
39049
FRANKLIN
73
3908V
LICKING
73
39097
MADISON
73
39129
PICKAWAY
73
39159
UNION
75
39029
COLUMBIANA
75
39099
MAHONING
75
39133
PORTAGE
75
39151
STARK
75
39153
SUMMIT
75
39155
TRUMBULL
75
39169
WAYNE
77
39005
ASHLAND
77
39033
CRAWFORO
77
39083
KNOX
77
39102
MARION
77
39117
MOKROW
77
39139
RICHLAND
77
39175
WYANDOT
78
39003
ALLEN
78
39011
AUGLAIZE
78
3906 5
HARDIN
78
39091
LOGAN
78
39107
MERCER
78
39149
SHELBY
80
18001
ADAMS
80
18003
ALLEN
80
18069
HUN1INGTON
80
18085
KOSCIUSKO
80
18 113
NOBLE
80
18179
WELLS
80
18183
WhITLEY
81
18099
MARSHALL
81
18149
STARKE
82
18011
BOONE
82
18057
HAMILTON
82
18 059
HANCOCK
a ^
18 063
HENDRICKS
be
16O8I
JunNSUft

180V7
MARION
az
io iOV
MLiHbAiX
8c
i8 133
PUTNAM
at
16145
SHELBY
83
18007
B EnT ON

^7

-------
Appendix (Con't)
COUNTY NArtfci AMD FIPS CODES FOk fcACH MIGRATION REGION
REGIUN FIP S	COUNTY
83
1800V
BLACKFORD
83
lb015
CAKHOlL
83
16017
CaSS
33
1 80 d'i
CL1N10N
83
180 3*
Ok LAMARt
63
ie04v
FULTON
83
18053
GRANT
83
18065
nE NRY
83
18067
HOWARD
83
18073
jASPtR
63
180 75
JAY
83
18095
MADISON
83
18103
MIAMI
83
18107
MONTGOMERY
83
18131
PULASKI
83
18135
KANDULPH
83
18157
TIPPECANOE
83
18159
TIPTON
83
16169
WABASH
83
181 61
WHITE
64
170 23
CLAR*
84
170 33
CRAWFORD
84
17101
LAwRENCt
84
18005
BARTHOLOMEW
84
laO 13
BKOWN
84
180^1
CLAY
8*
1B027
DAVIESS
84
180 31
DECATUR
84
18041
FAYETTE
84
18055
GREENE
84
180 71
JACKSON
84
1807V
JFNNlNGS
84
18083
KNOX
b4
180V3
LAWRENCE
84
18101
MARTIN
84
18105
MONROE
84
18119
OWEN
64
ism
PARKE
84
1813V
RUSH
84
18153
SULLIVAN
84
18161
UNION
84
18165
VERMILLION
84
18167
VIGO
84
181 rt
WAYNE
b5
170 47
EDWARDS
85
17059
GALLATIN
65
17065
HAMILTON
85
17165
SALINE
85
17185
WABASH
85
17193
Wn 11 E
85
18037
DUBOIS
85
18051
GIBSUN
85
18W3
PERRY
85
181^5
PI Kt

^8

-------
Appendix (Con't)
COUNT > NAMES AND FIPS CODES FOR EACH MIGRATION REGION
REGION
FIPS
COUNTY
85
18129
POSEY
85
18147
SPENCER
85
18163
VANDFRBURGh
65
16173
WARRICK
85
21101
HfcNDfcRSON
85
21107
HOPKINS
85
21225
UNION
85
212 33
WEBSTER
87
17005
BOND
87
17013
CALHOUN
87
170 25
CLAY
87
17027
CLIN10N
87
17049
EFFINGHAM
87
17051
FAYETTE
87
17061
GREENE
67
170 79
JASPER
87
17081
JEFFERSON
87
17083
JERSEY
87
17117
MACOUPIN
87
17119
MADISON
87
17121
MARION
87
17133
MONROE
87
17135
MONTGOMERY
87
171 59
RICHLAND
87
17163
ST CLAIR
87
17189
WASHINGTON
87
17191
WAYNE
8b
170 01
ADAMS
88
17009
BROWN
68
17017
CASS
88
17019
CHAMPAIGN
88
17021
CHRISTIAN
68
17029
COLES
88
17035
CUMBERLAND
88
17039
Dfc WITT
88
17041
DOUGLAS
88
17045
EDGAR
88
17053
FORU
88
170 57
FULlON
68
17067
HANCCOK
88
17071
HENDERSON
88
170 95
KNOX
88
17107
LOGAN
88
17109
MCDONOUGH
88
17113
MCLEAN
88
17115
MACON
88
17123
MARSHALL
88
17125
MASON
88
17129
MENARD
88
17137
MORGAN
88
17139
MOULTRIE
88
17143
PEORIA
88
17147
PIATT
88
17149
PIKt
1+9
-------
Appendix (Co n't)
COUN1Y NAMES AND FIPS CODES FOR
REGION FIPS
88
1 /167
88
17169
88
17171
88
17173
b8
17175
88
17179
88
17183
88
17187
88
172 03
88
18045
88
lbl 71
89
17055
89
17077
89
17087
89
17145
89
17157
89
171 bl
89
17199
99
17011
99
17063
99
17073
99
17075
99
17091
99
i 7099
99
171 C 5
99
17131
99
17155
151
18019
151
16043
151
21111
152
21005
152
2 1017
152
21067
152
21073
152
21097
152
21113
152
21167
152
21181
152
21201
152
21209
152
212 39
153
18025
153
18061
153
18077
153
16117
153
18143
153
161 75
153
21021
153
21027
153
21029
i.53
21041
153
21085
153
210 93
153
21099
EACH MIGRATION REG1 UN
COUNTY
SANGAMON
SCHUYLER
SCOTT
SHELBY
ST AKK
TAZEWELL
VERMILION
WAKREN
WOODFORD
FOUNTAIN
WARRfcN
FRANKLIN
JACKSON
JOHNSON
PERRY
RANDOLPH
UNION
WILLIAMSON
bUkfcAU
GftUNUY
fiENRY
IROQUOIS
KANKAKEE
LA SALLc
LIVINGSTON
MEKCER
PUTNAM
CLARK
FLOYD
JEFFERSON
ANDERSON
BOUK6UN
FAYETTE
FRANKLIN
HARRISON
JESSAMINE
MERCtR
NICHOLAS
ROBERTSON
SCOIT
WOODFORD
CRAWFORD
HARRISON
JtFFtRSON
ORANGE
SCOT!
WASHINGTON
BOYLt
SRbCKlNRlDGc
BULLITT
CAKROlL
GRAYSON
HAROlN
HART
50
-------
Appendix (Con't)
COUNTY NAME S AND FIPS COOES FOR EACH MIGRATION KEGlLN
REGION
FIPS
COUNTY
153
21103
HENRY
153
21123
LARUE
153
21155
MARION
153
21163
MEAD?
153
21279
NELSON
153
21185
OLDHAM
153
21187
0 w£ N
153
21211
SHELBY
153
21215
SPENCER
153
21217
TAYLOR
153
21223
TRIMBLE
153
21229
WASHINGTON
154
21003
ALLEN
154
21009
BARREN
154
21031
BUTLER
154
21059
DAVIESS
154
21061
EDMONSON
154
21091
HANCOCK
154
21141
LOGAN
154
21149
MCLEAN
154
2116V
METCALFE
15*
21177
MUHLENBERG
154
21 lb3
OHIO
154
21213
SIMPSON
154
21227
WARREN
155
17003
ALEXANDER
155
17069
HARDIN
155
17127
MASSAC
155
17151
POPE
155
17153
PULASKI
155
21007
BALLARD
155
21033
CALDWELL
155
21035
CALLOWAY
155
21C39
CAKL1SLE
155
21047
chkistian
155
21055
CRITTENDEN
155
^108 3
GRAVES
155
21105
HICKMAN
155
2113V
LIVINGSTON
155
21143
LYON
155
2114*
MCCRACKEN
155
21157
MARSHALL
155
21219
TODD
155
21221
TRIGG
160
21075
FULTON
*FIPS codes uniquely identify counties; The first two
digits are associated with the state and the last three
identify particular counties within that state. The
FIPS code prefixes for states included in the ORBES
region are: Illinois - 17; Indiana - 18; Kentucky - 21;
Ohio - 39; Pennsylvania - 42; West Virginia - 54.
51
-------