ED 353 102

AUTHORTITLE

INSTITUTION

PUB DATENOTEAVAILABLE FROM

PUB TYPE

EDRS PRICEDESCRIPTORS

IDENTIFIERS

ABSTRACT

DOCUMENT RESUME

RC 018 903

Redman, John M.Metro/Nonmetro Funding Allocation Under Title II-A,Job Training Partnership Act.Economic Research Service (DOA), Washington, DC.Agriculture and Rural Economy Div.Nov 9042p.

ERS-NASS, P.O. Box 1608, Rockville, MD 20849-1608(AGES 9071, $8).Reports Evaluative/Feasibility (142)

MF01/PCO2 Plus Postage.*Economically Disadvantaged; *Employment Programs;Federal Programs; Financial Support; *Job Training;Poverty; *Resource Allocation; *Rural Economics;*Rural Urban Differences; Unemployment*Job Training Partnership Act 1982 Title IIA

The Job Training Partnership Act (JTPA) Title II-Aprogram is the main federal effort to enhance the employability ofeconomically disadvantaged youths and adults. The amount of fundsallotted to each state is determined by a formula that allocatestwo-thirds of available funds on the basis of relative unemploymentlevels and one-third on the basis of the number of disadvantagedpersons in each state. This allocation formula produced a rise inexpenditures per member of the civilian labor force and expendituresper disadvantaged person in Service Delivery Areas (SDA's) in ruralareas between July 1987 and June 1988. This was due to the unusuallyhigh unemployment rates experienced by many nonmetro areas at thetime. The current regional distribution of JTPA funds was comparedwith distributions obtained from application of two alternativeformulas using measures of economic disadvantage (unweighted andweighted by the poverty rates). Application of either alternativeproduced a substantial reallocation of program activity amongregions. Under either alternative, however, the New York/New Jerseyregion would receive substantially larger allocations, and theindustrial Midwest would experience a substantial reduction. If oneused the number of disadvantaged living in the SDA as the soleallocation criterion, metro SDA's enjoyed a measurable increase inJTPA funds. Metro/nonmetro differences in unemployment rates havebeen steadily diminishing. Therefore, nonmetro areas may be betterserved during the 1990's by a new funding formula emphasizing povertyrates more than unemployment rates. The report contains numeroustables. (KS)

***********************************************************************

Reproductions supplied by EDRS are the best that can be madefrom the original document.

***********************************************************************

United StatesDepartment ofAgriculture

EconomicResearchService

Agricultureand RuralEconomyDivision

C\/

Metro/Nonmetro FundingAllocation Under Title II-AlJob Training Partnership ActJohn M. Redman

BEST CrrY

-PERMISSION TO REPRODUCE THISMATERIAL HAS B GRANTED BYE

TO THE EDUCATIONAL RESOURCESINFORMATION CENTER (ERIC).-

U.S. DEPARTMENT OF EDUCATIONOffice of Educational Research and Improvement

EDUCATIONAL RESOURCES INFORMATIONCENTER (ERtC1

/This document has been reproduced asreceived from the person or organizationoriginating it

C Minor changes have been made to improvereproduction quality

Points of new or opinions slated In this dccuimerit do not necessarily represent officialOERI position or policy

It's Easy To Order Another Copy!

Just dial 1-800-999-6779. Toll free in the United States and Canada.Other areas, please call 1-301-725-7937.

Ask for Metro/Nonmetro Funding Allocation Under Title II-A, Jcb Training PartnershipAct (AGES 9071).

The cost is $8.00 per copy. Please add 25 percent extra for postage to non-U.S. ad-dresses (including Canada). Charge your purchase to your VISA or MasterCard, orwe can bill you.. Or send a check or purchase order (made payable to ERS-NASS)to:

ERS-NASSP.O. Box 1608Rockville, MD 20849-1608.

We'll fill your order by first-class mail.

Metro/Nonmetro Funding Allocation Under Title 11-A, Job TrainingPartnership Act. By John M. Redman. Agriculture and Rural EconomyDivision, Economic Research Service, U.S. Department ofAgriculture. Staff Report No. AGES 9071.

Abstract

The Job Training Partnership Act's (JTPA) Title lI -A Program isthe main Federal effort to enhance the employability of theeconomically disadvantaged. Relying primarily on unemploymentrates, the JTPA allocation formula produced higher expendituresper disadvantaged person in the more rural areas between July1987 and June 1988. This was due to the unusually highunemployment rates experienced by many nonmetro areas at thetime. Metro/nonmetro differences in unemployment rates have beensteadily declining, however, so that this funding advantage israpidly diminishing. Nonmetro areas may be better served duringthe 1990's by a new funding formula emphasizing poverty ratesmore than unemployment rates.

Keywords: Job Training Partnership Act, nonmetro economics,rural economics, unemployment, poverty rate.

Acknowledgments

The author thanks David Sears, Norman Reid, Richard Reeder, PaulSwaim, and Patrick Sullivan of ERS, Greg Knorr of the U.S.Department of Labor, and Jose Figueroa of the National Governors'Association for their helpful comments. The author is alsoindebted to Lisa Mendelson of ERS and Gail Bourchers of the U.S.Department of Labor for their extensive assistance in databasedevelopment and data analysis.

1301 New York Avenue, NW.Washington, DC 20015 -4788 November 1990

iii

Contents

Summary v

Introduction 1

Overview of the JTPA Title II-A Program 1

Distribution of Program Activity by State and SDA Type 6

Relationship of Program Activity Levels to Size ofLabor Force, Unemployment, and Economic Disadvantage 12

"'Alternative Allocation Scenarios 21

Conclusions 24

iv

summary

The Job Training Partnership Act (JTPA) Title II-A program is themain Federal effort to enhance the employability of economicallydisadvantaged youths and adults. Administered by the U.S.Department of Labor (DoL), it is of particular importance to themany communities that have no other ongoing source of trainingfunds and whose local revenue bases are too small to supportsustained independent efforts.

The following report examines metro/nonmetro funding allocationsunder Title II-L during program year (PY) 87 (July 1987 throughJune 1988). Four major questions are addressed:

1. What is the current program structure, particularly thedistribution of program activity and funding between metroand nonmetro areas under the existing method of allocation?

2. How does spending per member of the civilian labor force,per unemployed person, and per economically disadvantagedperson vary between metro and nonmetro areas?

3. What factors are most important in determining spendingpatterns?

4. How do these patterns compare with those produced byallocation formulas relying more heavily on other measuresof economic disadvantage?

During PY87, the Title II-A program had 610 Service DeliveryAreas (SDA's). These are the program's local administrativeunits. Of these 610 SDA's, 261 (43 percent) were entirely inmetropolitan areas (called metro SDA's). Another 129 (about 21percent) were wholly in nonmetro areas (nonmetro SDA's). Theremaining 220 (36 percent) contained both metro and nonmetroareas. Of these, 136 (22 percent of the 610) had less than halfof their population living in nonmetro areas (metro dominantSDA's). The other 84 SDA's (14 percent of the 610) had more thanhalf of their population living in nonmetro areas (nonmetrodominant SDA's).

Most program activity was concentrated in fewer than 20 States.Among the nonmetro dominant and nonmetro SDA's, 17 Statesaccounted for two-thirds of the program terminees (the program'sname for participants who completed training or otherwise leftthe program). Ten of the 14 largest programs are among these 17.

The typical program was of similar size (measured in terms ofcosts and participant numbers) in the metro, metro dominant, andnonmetro dominant SDA categories. The average nonmetro programwas considerably smaller.

There was a smooth decline from metro to metro dominant tononmetro dominant to nonmetro SDA's in average population densityand prevailing wage rates and a smooth increase in unemploymentand poverty rates.

Expenditures per disadvantaged person (that is, a person definedas disadvantaged in Title II-A of the Job Training PartnershipAct) varied widely across SDA's. Disadvantaged individuals livingin different SDA's consequently faced very different odds ofreceiving training.

The unemployment rate appears to be the key factor in explainingnational patterns of expenditure per labor force member and perdisadvantaged person. This appears to have particularly benefitedthe more rural SDA's because they had unusually high unemploymentrates during the period studied. When the unemployment rate iscontrolled for, the number of economically disadvantaged personsliving in an SDA was negatively associated with SDA expendituresper disadvantaged person.

SDA's with lower expenditures per local disadvantaged person wereoften those with a higher percentage of local families livingbelow the poverty line. This suggests that jurisdictions withmore severe problems frequently receive a lower relative level ofJTPA support.

The current regional distribution of JTPA funds was compared withdistributions obtained from application of two alternativeformulas utilizing measures of economic disadvantage only.Application of either alternative produced a substantialreallocation of program activity among regions. Under eitheralternative, however, the New York/New Jersey region wouldreceive substantially larger allocations, and the industrialMidwest would experience a substantial reduction.

If one uses the number of disadvantaged living in the SDA as thesole allocation criterion, metro SDA's enjoyed a measurableincrease in JTPA funds. If one uses a formula that includes boththe SDA poverty rate and the number of disadvantaged, nonmetrodominant SDA's experienced a substantial funding increase at theexpense of a slight reduction in funding for the other threecategories.

For those who work in rural development, the conclusions suggestthat, by virtue of its reliance on unemployment rates, thecurrent JTPA allocation formula produced higher expenditures perdisadvantaged person in the more rural SDA's during the studyperiod. This is a direct result of the unusually highunemployment rates in many nonmetro areas at the time.

Metro/nonmetro differences in unemployment rates have recentlybeen declining, however, so the funding advantage provided by useof unemployment rates may be rapidly diminishing. In contrast,the metro/nonmetro poverty rate differential, defined in terms ofthe percentage of individuals living below the poverty line, hasincreased. In the future, heavier reliance on the poverty rateto allocate JTPA funds may thus offer nonmetro areas a fundingadvantage of comparable size and (probably) of greater stabilitythan that afforded in recent years by the unemployment rate.

vi

Metro/Nonmetro FundingAllocation Under Title II-A,

Job Training Partnership ActJohn M. Redman

Introduction

This report examines the subject of metro/nonmetro fundingallocation under Title II-A of the Job Training Partnership Act(JTPA) during program year (PY) 87 (July 1987 through June 1988).

Four major questions will be considered:

1. What is the current program structure, particularly thedistribution of program activity/funding between metro andnonmetro areas under the existing all cation mechanism?

2. How does spending per member of the civilian labor force,per unemployed person, and per economically disadvantagedperson vary between metro and nonmetro areas?

3. What factors are most important in determining spendingpatterns?

4. How do these patterns compare with those produced byallocation formulas relying more heavily on other measures

of economic disadvantage?

Since the metro/nonmetro dimension of the Title II-A program has

not so far received serious study, this analysis may proveparticularly useful for those who work in the field of rural

economic development.

Overview of The JTPA Title II-A Program

The Title II-A program, technically titled "Training Services forthe Disadvantaged: Adult and Youth Programs," was authorized in1982 as part of the original Job Training Partnership Actlegislative package. This package also included summer youthemployment and training programs (Title II-B), employment andtraining services for dislocated workers (Title III), services

for native Americans and migrant and seasonal farmworkers (Title

IV-A), the Job Corps (Title IV-B), and veterans' employmentprograms (Title IV-C).

Administered by the U.S. Department of Labor (DoL), Title II-A is

the major Federal program designed to enhance the employabilityof the economically disadvantaged through training services.

1

0

Total local expenditure of Federal funds under Title II-A wasabout $1.6 billion in PY87.

The definition of economically disadvantaged used to determine anindividual's eligibility for Title II-A services is given in thelegislation, which specifies six different categories ofeligibles:

1. An individual who receives, or who is a member of a familythat receives, cash welfare payments under a Federal, State,or local welfare program. These programs include Aid toFamilies with Dependent Children, General Assistance, andRefugee Assistance.

2. Individuals whose total family income for the 6 monthsprior to program application was less than the poverty levelor less than 70 percent of the "lower living standard"income level,2 whichever was higher.

3. An individual receiving food stamps.

4. A homeless person as defined under the McKinney HomelessAssistance Act.

5. A foster child on whose behalf State or local governmentpayments are made.

6. Low-income handicapped individuals whose own income is lessthan the income thresholds in number 2 above but whosefamily income exceeds those thresholds.

Title II-A funds are provided as block grants to the States,which administer the JTPA structure. The program's localadministrative districts are termed "Service Delivery Areas" orSDA's. The State is allowed broad discretion in defining SDAboundaries. This has resulted in wide variation across States in

Exclusive of unemployment compensation, child supportpayments, and welfare payments.

2The "lower living standard income" is defined in thelegislation as "that income level (adjusted for regional,metropolitan, urban and rural differences and family size)determined annually by the Secretary [of Labor] based on the mostrecent 'lower living family budget' issued by the Secretary."This budget, in turn, was developed by the Bureau of LaborStatistics (BLS) as an alternative measure of poverty levelincome. Its computation was discontinued several years ago,however. The annual adjustments called for by the legislation arethus adjustments to the last budget issued by BLS. Caseworkers atthe SDA level are provided both poverty line and lower livingstandard income levels for assessing individual eligibility. Theeligibility determination is then made by applying the higher ofthe two measures to the individual's reported family income.

2

the number of SDA's within the State and the average SDApopulation. There were 610 SDA's during PY87.

The amount allotted to each State is determined by a formula thatallocates two-thirds of available funds on the basis of relativeunemployment levels3 and one-third on the basis of the number ofdisadvantaged persons in each State. Seventy-eight percent of theblock grant funding received by each State must be allocated tothe 3DA's by the same formula used to distribute funds across theStates. These funds are termed "basic formula" funds. Theremaining 22 percent is available for State "set-aside"activities:4

1. Eight (of the 22) percent to State education agencies foreducation and training services to eligible participantsthrough cooperative agreements.

2. Six percent for (a) incentive grants to SDA's that exceedprogram performance standards and (b) technical assistanceto SDA's.

3. Five percent for auditing and general programadministration.

4. Three percent for special training for persons aged 55and over who are economically disadvantaged.

Of the basic Title II-A formula funds received by the SDA fromthe State, 70 percent must be spent on training. Of the remaining30 percent, no more than 15 percent may be spent foradministrative costs. The other 15 percent is for supportservices for program participants (such as child care costs or

3Technically, there are two unemployment-based measures. Eachis used to allocate one-third of total program funding among theStates. The first measure is defined as "the relative number ofunemployed individuals residing in areas of substantialunemployment in each State as compared to the total number ofsuch unemployed individuals in all such areas of substantialunemployment in all the States." An area of substantialunemployment is defined, in turn, as "any area of sufficient sizeand scope to sustain a program under part A of Title II of theAct and which has an average rate of unemployment of at least 6.5percent for the most recent twelve months as determined by theSecretary [of Labor]." The second unemployment measure is "therelative excess number of unemployed individuals who reside ineach State as compared to the total excess number of unemployedindividuals in all the States." Excess number means the"...number of unemployed individuals in excess of 4.5 percent ofthe civilian labor force in the State ..." or "...the number ofunemployed individuals in excess of 4.5 percent of the civilianlabor force in areas of substantial unemployment in each State."

4There is no information at the national level on how theseset-aside funds are allocated between metro and nonmetro areas.

3

transportation). Forty percent of the dollars received must bespent on youth programs, and up to 10 percent may be spent onnondisadvantaged individuals with other important barriers toemployment (such as the handicapped or ex-offenders).

Within the SDA, the Private Industry Council (PIC) is the chiefadministrative unit. The PIC is to "provide policy guidance for,and exercise oversight with respect to, activities under the jobtraining plan tor its service delivery area in partnership withthe unit or units of general local government within its servicedeliver- area" (Sec. 103(a) of the act). A majority of the Pr'member 'i.p consists of representatives of the private sector, butit must also have local representation from organized labor,educational and rehabilitation agencies, community-basedorganizations, local economic development agencies, and thepublic employment service. This membership blend is intended topromote close program ties to the private business community,while maintaining direct input from other key institutions.

The PIC must work closely with local elected officials becauseany plan it develops must be approved by the chief electedofficial(s) of each general unit of government within the SDA orby their representative(s). The plan must, in addition, bereviewed and approved by the governor. This rather elaborateprocess is intended to produce a plan that has received broadreview and political support and thus stands a gond chance ofeffective, sustained implementation.

Besides approving the local plans, the governor has primaryresponsibility for ongoing administrative oversight. The governoralso manages the State set-aside programs.

Table 1 provides basic information on the PY86 organizationalstructure and size of SDA's on a State-by-State basis.5 A reviewof these structures suggests that States have taken five basic

5Program data for PY 1986 and 1987 were obtained from the U.S.Department of Labor (DoL). The program year runs from July 1through June 30 with PY87 ending on June 30, 1988. Data arereported to DoL by the individual States. The States, in turn,receive data from each of the local SDA's, the program's basicadministrative unit. The standard reporting form used by allSDA's is called the JTPA Annual Status Report (JASR). The JASRprovides information on activity levels, participantcharacteristics, and program cost. Population data used in thisreport to estimate the percentage of an SDA's total nonmetropopulation in 1986 are county- and minor civil division (MCD)-level estimates from the Bureau of the Census. Counties weredesignated as metro or nonmetro depending on whether they werewithin a metropolitan statistical area (MSA) as defined by theOffice of Management and Budget (OMB) in 1983.

4

Table 1-- Organizational and size characteristics of SDA's by State, PY86

PY86 1986

SDA organization number State Mean pop. Mean pop.of SDA's population per SDA per SDA type

Number

Single-State SDA's: 7 1,010,000

Delaware 632,700 632,700Dist. Columbia 626,100 626,100North Dakota 679,300 679,300South Carolina 3,375,300 3,375,300South Dakota 708,000 708,000Vermont 541,100 541,100Wyoming 507,500 507,500

Modified single-State SDA's: 28 609,574

Alabama 3 4,052,300 1,350,767Alaska 3 533,600 177,867Maine 2 1,173,600 586,800Mississippi 3 2,625,500 875,167Montana 2 818,800 409,400Nebraska 3 1,597,800 532,600Nevn,lit 2 963,200 481,600New Hampshire 2 1,026,900 513,450New Mexico 3 1,479,800 493,267Rhode Island 3 975,000 325,000West Virginia 2 1,918,800 959,400

County-based SDA's: 155 310,678

Colorado 10 3,266,700 326,670Hawaii 4 1,062,300 265,575Idaho 6 1,001,500 167,083Indiana 17 5,503,600 323,741Iowa 16 2,850,800 178,175Kansas 5 2,460,400 492,080Kentucky 9 3,727,900 414,211Maryland 10 4,463,300 446,330North Carolina 26 6,331,600 243,523Tennessee 14 4,802,900 343,064Utah 9 1,665,300 185,033Washington 12 4,462,500 371,875Wisconsin 17 4,784,800 281,459

Modified county-based SDA's: 396 375,674

Arizona 16 3,279,700 204,981Arkansas 10 2,372,200 237,220California 51 26,981,000 529,039Florida 24 11,674,900 486,454Georgia 18 6,104,300 339,128Illinois 26 11,553,200 444,354Louisiana 17 4,501,300 264,782Michigan 26 9,144,600 351,715Minnesota 17 4,213,900 247,876Missouri 15 5,066,000 337,733New Jersey 17 7,619,600 448,212New York 34 17,772,100 522,709Ohio 30 10,752,500 358,417Oklahoma 12 3,305,600 275,467Oregon 7 2,697,900 385,414Pennsylvania 28 11,889,200 424,614Texas 34 16,682,100 490,650Virginia 14 5,781,200 413,371

Town-based SDA States: 24 371,547

Connecticut 9 3,188,700 354,300Massachusetts 15 5,831,900 388,793

U.S. total, 1986 610 241,037,800 395,144

Source: JTPA Annual Status Report data, as compiled by the U.S. Department of Labor,Office of Strategic Planning and Policy Development.

5

CI /3 i

app:oaches6 to defining their SDA jurisdictions. Theseapproaches are:

1. The single-State SDA--all areas in the State are in oneSDA. These SDA's had, on average, about 2-1/2 times the U.S.mean SDA population.

2. The modified single-State SDA--most of the physical areaof the State is in a single SDA, but selected metro areasare designated as separate SDA's. The average SDA populationin these States was about 1-1/2 times the U.S. mean.

3. County-based SDA's--SDA jurisdictional boundaries aredefined exclusively in terms of county boundaries. Thenumber of counties in each SDA varies widely within andbetween States. Average SDA size in these States was about80 percent of the U.S. mean.

4. Modified county-based SDA's--SDA's are defined principallyby county boundary, but selected urbanized areas withinindividual counties, such as a large city, are designated asseparate SDA's. The balance of the affected counties areplaced in different SDA's. Average SDA size was slightlybelow the national mean.

5. Town-based SDA's--Connecticut and Massachusetts define theirSDA's principally in terms of town rather than countyboundaries. SDA average size was also slightly below thenational mean.

No one has examined SDA jurisdictional boundaries across theNation to see how closely they correspond to local labor marketareas.

Distribution of Program Activity by State and SDA Type

Tables 2, 3, and 4 present data regarding the distribution ofprogram activity by what I call SDA type. These categories aredefined in terms of the percentage of total 1986 SDA populationliving in nonmetro areas. Four types7 were defined as follows:

1. Metro--zero percent of the SDA population living innonmetro areas.

2. Metro dominant--0.01 to 49.99 percent of the SDA populationliving in nonmetro areas.

6These categories were developed by the author for descriptivepurposes only. There are no such regulatory definitions.

7These categories were also developed for this study only.There are no such regulatory definitions.

6

Tab

le 2

-Dis

trib

utio

n of

PY

87 a

dult

term

inee

s by

Sta

te, b

ySD

A ty

pe, r

anke

d by

sum

of

term

inee

sin

non

met

ro a

nd n

onm

etro

dom

inan

t SD

A's

Stat

e

Tot

al te

rmin

ees

per

SDA

type

Tot

alte

nnin

ees

Met

ron=

261

Met

rodo

min

ant

n=13

6

Non

met

rodo

min

ant

Non

met

ron=

84n=

129

Ter

min

ee p

erce

nt o

f St

ate

tota

l

Met

ron=

261

Met

rodo

min

ant

n=13

6

Non

met

rodo

min

ant

Non

met

ron=

84n=

129

Sum

of

nonm

etro

& n

onm

etro

Perc

ent o

fdo

min

ant

tota

lte

rmin

ees

Cum

ulat

ive

perc

ent

Mis

siss

ippi

Nor

th C

arol

ina

Ten

ness

eeA

rkan

sas

Ken

tuck

yW

isco

nsin

Mic

higa

nL

ouis

iana

Ohi

oIl

linoi

sIn

dian

aM

isso

uri

Okl

ahom

aT

exas

Min

neso

taW

est V

irgi

nia

Geo

rgia

Vir

gini

aN

ew Y

ork

Penn

sylv

ania

Ore

gon

Cal

ifor

nia

Iow

aSo

uth

Dak

ota

Ari

zona

Was

hing

ton

Mon

tana

Idah

oK

ansa

sC

olor

ado

Mai

neN

ew M

exic

oN

ebra

ska

Ver

mon

tN

orth

Dak

ota

Wyo

min

gU

tah

Ala

ska

New

Ham

pshi

reFl

orid

aM

aryl

and

Haw

aii

Con

nect

icut

Sout

h C

arol

ina

Rho

de I

slan

dN

ew J

erse

yN

evad

aM

assa

chus

etts

Dis

tric

t of

Col

umbi

aD

elaw

are

Ala

bam

a

Tot

al

10,1

5512

,000

11,9

287,

118

8,25

410

,415

23,3

1211

,700

25,1

7625

,610

10,7

539,

659

6,73

425

,166

9,79

94,

778

8,94

67,

083

34,3

5620

,134

7,13

533

,387

5,96

12,

318

5,11

19,

413

1,93

62,

639

2,65

26,

935

1,78

72,

195

2,18

21,

219

982

960

2,71

181

856

821

,560

8,77

91,

205

2,66

35,

945

1.00

07,

732

1,67

54,

606

1,78

81,

156

9,81

6

441,

910

1,29

13,

203

1,95

416

31,

075

2,88

613

,222

4,42

513

,018

14,6

962,

842

3,96

62,

423

10,8

473,

075

374

2,46

91,

122

30,4

4512

,049

3,29

027

,389

0 02,

873

4,65

0 0 023

54,

300

182

717

662 0 0 0

118 0 75

10,9

537,

481

859

1,89

8 065

37,

732 0

4,07

91,

788 0

3,53

1

209,

010

Num

ber 0

1,55

53,

926

1,02

21,

549

2,07

64,

717

2,26

47,

452

6,22

33,

548

1,37

9 010

,156

2,64

132

52,

539

2,31

937

84,

821

874

3,13

83,

129 0 0

2,54

5 075

858

51,

019 0 0

150 0 0 0

1,74

227

8 010

,126 824 0

610

5,94

534

7 01,

675

527 0

1,15

66,

285

100,

603

8,86

43,

249

3,22

32,

856

4,74

63,

790

1,18

83,

550

2,56

045

93,

654

1,15

32,

844

2,92

82,

978

4,07

92,

796

2,86

11,

412

1,35

1 0 036

52,

318 0

1,34

81,

530 0

1,29

8 01,

605

1,38

51,

370

1,21

998

296

0 0 049

3 012

0 015

5 0 0 0 0 0 0 0 0

75,6

89

03,

993

2,82

53,

077

884

1,66

34,

185

1,46

12,

146

4,23

270

93,

161

1,46

71,

235

1,10

5 01,

142

781

2,12

11,

913

2,97

12,

860

2,46

7 02,

238

870

406

1,88

153

41,

616 0 93 0 0 0 0

851

540 0

481

354

346 0 0 0 0 0 0 0 0 0

56,6

08

13 27 16 2 13 28 57 38 52 57 26 41 36 43 318 28 16 89 60 46 82 0 0 56 49 0 0 9 V 10 33 30 0 0 0 4 0 13 51 85 71 710 65 100 0 89 100 0 36 47

Perc

ent

0 13 33 14 19 20 20 19 30 24 33 14 0 40 27 72E 33

1

24 129

520 0 27 0 29 22 150 0 7 0 0 0 64 34 0

479 0 23 100 35 0

100 11 0

100

64 23

87 27 27 40 57 365 30 10 2 34 12 42 12 30 85 31 40 4 7 0 0 6

100 0 14 79 0 49 0 90 63 63 100

100

100 0 0 87 0 1 0 6 0 0 0 0 0 0 0 0 17

0 33 24 43 11 16 18 12 9 17 7 33 22 5 11 0 13 11 6 10 42 9 410 44 9

21 71 20 23 0 4 0 0 0 0 31 66 0 2 4 29 0 0 0 0 0 0 0 0 0 13

- --

Num

ber

- --

8,86

47,

242

6,04

85,

933

5,63

05,

453

5,37

35,

011

4,70

64,

691

4,36

34,

314

4,31

14,

163

4,08

34,

079"

3,93

83,

642

3,53

33,

264

2,97

12,

860

2,83

22,

318

2,23

82,

218

1,93

61,

881

1,83

21,

616

1,60

51,

478

1,37

01,

219

982

960

851

540

493

481

474

346

155 0 0 0 0 0 0 0 0

132,

297

Perc

ent

6.7

5.5

4.6

4.5

4.3

4.1

4.1

3.8

3.6

3.5

'3.3

3 3

3.3

3.1

3.1

3.1

3.0

2.8

2.7

2.5

2.2

2.2

2.1

1.8

1.7

1.7

1.5

1.4

1.4

1.2

1.2

1.1

1.0 .9 .7 .7 .6 .4 .4 .4 .4 .3 .1 0 0 0 0 0 0 0 0

6.7

12.2

16.7

21.2

25.5

29.6

33.7

37.5

41.0

44.6

47.9

51.1

54.4

57.5

60.6

63.7

66.7

69.4

72.1

74.6

76.8

79.0

81.1

82.9

84.6

86.2

87.7

89.1

90.5

91.7

92.9

94.1

95.1

96.0

96.7

97.5

98.1

98.5

98.9

99.3

99.6

99.9

100.

010

0.0

100.

010

0.0

100.

010

0.0

100.

010

0.0

100.

0

Sour

ce: I

TPA

Ann

ual S

tatu

s R

epor

t dat

a, a

s co

mpi

led

by th

e U

.S. D

epar

tmen

t of

Lab

or,O

ffic

e of

Str

ateg

ic P

lann

ing

and

Polic

y D

evel

opn,

ent.

14B

EST

CO

PY I

MR

E

Tab

le 3

--D

istr

ibut

ion

of P

Y87

wel

fare

adu

lt te

rmin

ees

by S

tate

, by

SDA

type

, ran

ked

by s

um o

f te

rmin

ees

in n

onm

etro

and

non

met

ro d

omin

ant S

DA

's

Stat

e

Tot

al te

rmin

ees

per

SDA

type

Ter

min

ee p

erce

nt o

f St

ate

tota

lSu

m o

fno

nmet

roT

otal

wel

fare

Met

roN

onm

etro

Met

roN

onm

etro

& n

onm

etro

Perc

ent o

fC

umul

ativ

ete

rrni

nces

Met

rodo

min

ant

dom

inan

tN

onm

etro

Met

rodo

min

ant

dom

inan

tN

onm

etro

dom

inan

tto

tal

perc

ent

n=26

1n=

136

n=84

n=12

9n=

261

n=13

6n=

84n=

129

term

inee

s

Num

ber

Perc

ent

- --

Num

ber

- --

Perc

ent

Mic

higa

n10

,267

5,42

22,

406

486

1,95

353

235

192,

439

7.2

7.2

Wis

cons

in4,

694

1,57

497

51,

488

657

3421

3214

2,14

56.

313

.5O

hio

11,3

376,

250

2,97

31,

181

933

5526

108

2,11

46.

219

.7M

issi

ssip

pi1,

875

135

01,

740

07

093

01,

740

5.1

24.8

Min

neso

ta4,

720

1,66

81,

352

1,26

343

735

2927

91,

700

5.0

29.8

New

Yor

k10

,899

9,01

619

160

81,

084

832

610

1,69

25.

034

.8N

orth

Car

olin

a2,

743

696

390

687

970

2514

2535

1,65

74.

939

.7Il

linoi

s9,

787

5,98

12,

216

143

1,44

761

231

151,

590

4.7

44.3

Wes

t Vir

gini

a1,

552

126

122

1,30

40

88

840

1,30

43.

848

.2Pe

nnsy

lvan

ia7,

703

4,84

51,

755

457

646

6323

68

1,10

33.

251

.4C

alif

orni

a10

,687

8,89

578

60

1,00

683

70

91,

006

3.0

54.4

Ten

ness

ee2,

074

415

667

491

501

2032

2424

992

2.9

57.3

Ark

ansa

s1,

119

2713

453

142

72

1247

3895

82.

860

.1M

isso

uri

1,87

676

319

324

167

941

1013

3692

02.

762

.8G

eorg

ia2,

148

621

631

566

330

2929

2615

896

2.6

65.5

Tex

as4,

915

2,19

01,

851

662

212

4538

134

874

2.6

68.0

Iow

a1,

854

01,

013

140

701

055

838

841

2.5

70.5

Vir

gini

a1,

866

376

703

642

145

2038

348

787

2.3

72.8

Lou

isia

na1,

769

704

296

506

263

4017

2915

769

2.3

75.1

Ken

tuck

y1,

143

190

232

584

137

1720

5112

721

2.1

77.2

Okl

ahom

a1,

064

344

046

925

132

044

2472

02.

179

.3W

ashi

ngto

n3,

045

1,53

587

838

724

550

2913

863

21.

981

.2O

rego

n1,

358

604

145

060

944

110

4560

91.

883

.0In

dian

a1,

505

452

487

483

8330

3232

656

61.

784

.6M

onta

na56

00

040

715

30

073

2756

01.

686

.3V

erm

ont

487

00

487

00

010

00

487

1.4

87.7

Mai

ne53

769

046

80

130

870

468

1.4

89.1

Kan

sas

770

7124

036

198

931

4713

459

1.3

90.4

Ari

zona

1,35

890

30

045

566

00

3445

51.

391

.8So

uth

Dak

ota

351

00

351

00

010

00

351

1.0

92.8

Neb

rask

a63

025

236

342

040

654

034

21.

093

.8Id

aho

461

016

90

292

037

063

292

.994

.'7C

olor

ado

1,61

21,

156

203

025

372

130

1625

3.7

95.4

Nor

th D

akot

a22

00

022

00

00

100

022

0.6

96.1

New

Mex

ico

317

990

202

1631

064

521

8.6

96.7

Mar

ylan

d3,

756

3,27

726

434

181

877

15

215

.697

.3U

tah

'Vyo

min

g69

226

465

020

14

670

2920

1.6

97.9

194

00

194

00

010

00

194

.698

.5N

ew H

amps

hire

167

180

149

011

089

014

9.4

98.9

Ala

ska

176

061

011

50

350

6511

5.3

99.3

Haw

aii

344

255

00

8974

00

2689

.399

.5Fl

orid

a3,

790

1,83

51,

870

085

4849

02

85.2

99.8

Con

nect

icut

1,26

496

322

774

076

186

074

.210

0.0

Dis

tric

t of

Col

umbi

a15

515

50

00

100

00

00

010

0.0

Nev

ada

148

014

80

00

100

00

00

100.

0So

uth

Car

olin

a1,

550

01,

550

00

010

00

00

010

0.0

Del

awar

e38

6(

386

00

010

00

00

010

0.0

Mas

sach

uset

ts2,

042

1,79

624

60

088

120

00

010

0.0

Rho

de I

slan

d41

528

513

00

069

310

00

010

0.0

New

Jer

sey

2,79

72,

797

00

010

00

00

00

100.

0A

laba

ma

1,01

953

848

10

053

470

00

010

0.0

Tot

al12

8,19

867

,324

26,8

7218

,348

15,6

5453

2114

1234

,002

Sour

ce: J

TPA

Ann

ual S

tatu

s R

epor

t dat

a, a

s co

mpi

led

by th

e U

. S.

Dep

artm

ent o

f L

abor

, Off

ice

of S

trat

egic

Pla

nnin

g an

d Po

licy

Dev

elop

men

t.

Tab

le 4

-Dis

trib

utio

n of

PY

87 y

outh

term

inee

s by

Sta

te, b

y SD

A ty

pe, r

anke

d by

sum

of te

rmin

ees

in n

onm

etro

and

non

met

ro d

omin

ant S

DA

's

Stat

e

Total ter minces per SDA type

Terminee percent of State total

Sum

of

Tot

alno

nmet

ro

yout

hM

etro

Non

met

roM

etro

Non

met

ro&

non

nict

roPe

rcen

t of

Cum

ulat

ive

term

inee

sM

etro

dom

inan

tdo

min

ant

Non

met

roM

etro

dom

inan

tdo

min

ant

Non

met

rodo

min

ant

tota

lpe

rcen

t

n=26

1n=

136

n=84

n=12

9r=261

n=13

6n=

84n=

129

term

ince

c

Number

Percent

'Number-

Percent

Mississippi

10,502

1,666

08,836

016

084

08,836

8.1

8.1

North Carolina

11,992

2,084

1,551

4,266

4,091

17

13

36

34

8,357

7.6

15.7

Kentucky

9,588

1,488

1,202

5,813

1,085

16

13

61

11

6,898

6.3

22.0

Tennessee

11,272

1,434

3,134

?,500

3,204

13

28

31

28

6,704

6.1

28.1

Louisiana

14,048

4,187

3,184

4,927

1,750

30

23

35

12

6,677

6.1

34.2

Arkansas

5,721

196

623

2,995

1,907

311

52

33

4,902

4.5

38.7

Wisconsin

8,267

2,397

1,886

2,476

1,508

29

23

30

183,984

3.6

42.3

Georgia

7,884

1,556

2,517

2,512

1,299

20

32

32

16

3,811

3,5

45.8

Illinois

23,415

14,539

5,100

238

3,538

62

22

115

3,776

3.4

49.2

Michigan

15,226

8,367

3,373

569

2,917

55

22

419

3,486

3.2

52.4

Ohio

21,803

13,098

5,529

1,322

1,854

60

25

69

3,176

2.9

55.3

Oklahoma

5,022

1,894

02,367

761

38

047

15

3,128

2.9

58.2

Texas

21,824

9,880

8,828

2,048

1,068

45

40

95

3,116

2.8

61.0

Missouri

7,249

3,387

854

954

2,054

47

12

13

28

3,008

2.7

63.8

Virginia

6,78F

1,485

2,333

2,350

620

22

34

35

92,970

2.7

66.5

Indiana

7,511

2,273

2,660

2,087

491

30

35

28

72,578

2.4

68.8

California

29,704

25,411

1,719

02,574

86

60

92,574

2.3

71.2

Minnesota

5,443

1,904

988

1,803

748

35

18

33

14

2.551

2.3

73.5

West Virginia

2,799

233

109

2,457

08

488

02,457

2.2

75.8

New York

21,252

18,645

242

796

1,569

88

14

72,365

2.2

77.9

Pennsylvania

15,839

10,821

2,921

1,017

1,080

68

18

67

2,097

1.9

79.8

New Mexico

2,466

637

01,641

188

26

067

81,829

1.7

81.5

Oregon

4,853

2,689

382

01,782

55

80

37

1,782

1.6

83.1

Arizona

4,611

2,891

00

1,720

63

00

37

1,720

1.6

84.7

Iowa

3,993

02,294

205

1,494

057

537

1,699

1.6

86.2

Washington

7,099

3,564

1,889

1,047

599

50

27

15

81,646

1.5

87.7

Kansas

2,472

577

365

1,012

518

23

1541

21

1,530

1.4

89.1

South Dakota

1,403

00

1,403

00

0100

01,403

1.3

90.4

Colorado

4,446

2,583

530

01,333

58

12

030

1,333

1.2

91.6

Idaho

1,766

0504

01,262

029

071

1,262

1.2

92.8

Maine

1,248

59

01,189

05

095

01,189

1.1

93.9

Montana

1,126

00

903

223

00

80

20

1,126

1.0

94.9

Vermont

851

00

851

00

0100

0851

.895.7

Nebraska

1,075

263

46

766

024

471

0766

.7

96.4

Wyoming

684

00

684

00

0100

0684

.6

97.0

Maryland

5.799

4,506

728

190

375

78

13

36

565

.5

97.5

Alaska

894

0354

0540

040

060

540

.5

98.0

North Dakota

494

00

494

00

0100

0494

.5

98.5

Utah

1,537

106

975

0456

763

030

456

.4

98.9

New Hampshire

557

105

0452

019

081

0452

.4

99.3

Florida

14,936

7,663

6,939

0334

51

46

02

334

.3

99.6

Hawaii

1,401

1,132

00

269

81

00

19

269

.2

99.8

Connecticut

2,138

1,418

546

174

066

26

80

174

.2

100.0

Rhode Island

831

588

243

00

71

29

00

00

100.0

Massachusetts

3,015

2,607

408

00

86

14

00

00

100.0

Delaware

987

0987

00

0100

00

00

100.0

Nevada

2,091

02,091

00

0100

00

00

100.0

New Jersey

2,797

2,797

00

0100

00

00

0100.0

Sout

h C

arol

ina

5,92

60

5,926

00

0100

00

00

100.0

District of Columbia

362

362

00

0100

00

00

0100.0

Alabama

9,164

3,221

5.943

00

35

65

00

00

100.0

Total

354,171

164,713

79,903

64,344

45,211

47

23

18

13

109,555

Source: JTPA Annual Status Report data, as compiled by the U.S. Department of Labor, Office of Skrategic Planning and Policy Development.

BE

ST C

OPY

AV

AIL

AB

LE

1 9

3. Nonmetro dominant--50 to 99.99 percent of the SDA populationliving in nonmetro areas.

4. Nonmetro--100 percent of the population living in nonmetroareas.

Program activity is presented in terms of the total number of alladult terminees, the number of welfare adult terminees, and thenumber of youth terminees.8 "Terminee" is a program term used todenote a person who completed training or otherwise left the SDAprogram without completing training. Welfare adult terminees area subset of all adult terminees (that is, adult terminees who hadbeen receiving welfare upon entry into the program). Youthterminees are terminees who were less than 21 years of age whenthey entered the program. Though they may be receiving welfaresupport, they are included in the youth rather than welfare adultcategory.

In PY87, there were 261 metro SDA's and 129 nonmetro SDA's. Theremaining 220 SDA's were distributed very evenly between thesetwo extremes. Metro SDA's accounted for slightly less than halfof the all-adult and youth terminees and a little more than halfof the welfare-adult terminees. Nonmetro SDA's accounted forabout one-eighth of the terminees in each group.

Tables 2, 3, and 4 are arranged by State according to the sum oftheir terminees in nonmetro and nonmetro dominant SDA's. Thisordering was chosen to highJight States with the largest programsin predominantly nonmetro areas. Seventeen States, located almostexclusively in the Southeast and industrial Midwest, accountedfor two-thirds of the total terminees in these two SDA types. Tenof the 14 largest programs were in these 17 States.

Alabama and South Carolina are anomalies within thisclassification scheme. Each State has sizeable nonmetropopulations served by single, geographically extensive SDA's withpopulations more than 50 percent metropolitan. Consequently,these two States show no terminees in either the nonmetro ornonmetro dominant categories despite large aggregate nonmetropopulation percentages.9

Table 5 displays simple means for basic program characteristicsby SDA type across all 610 SDA's. Program cost is a measure ofFederal funds expended on training for each terminee group by theSDA during the program year. Separate cost data are collected

Data on the number of terminees and on program cost measuresare from the JTPA Annual Status Report (JASR) data provided bythe Department of Labor.

9In PY88, South Carolina disaggregated its single SDA intoseveral smaller SDA's. For the program year under study, however,there was a single SDA for the entire State.

r10 (,

Tab

le 5

-Pro

gram

cos

t and

par

ticip

atio

n

Item

Uni

t

PY86

unw

eigh

ted

U.S

. mea

nn=

610

PY86

unw

e;,,h

ted

mea

nby

SD

A'p

ePY

87PY

87 u

nwei

ghte

d m

ean

by S

DA

type

Met

ron=

262

Met

rodo

min

ant

n=13

5

Non

met

rodo

min

ant

n=84

Non

met

ron=

129

unw

eigh

ted

U.S

. mea

nn=

610

Met

ron=

261

Met

rodo

min

ant

n=13

6

Non

met

rodo

min

ant

n=84

Non

met

ron=

129

Adu

lt:M

ean

tota

l pro

gram

cos

tD

olla

rs1,

441,

253

(1)

1,68

2,30

5 (1

)1,

468,

316

1,58

8,78

682

9,15

61,

411,

575

1,60

5,68

11,

425,

618

1,68

2,73

482

7,47

9

SDA

type

mea

n/U

.S. m

ean

Rat

io1.

001.

171.

021.

10.5

71.

001.

141.

011.

19.5

9

Part

icip

ants

Num

ber

965

1,03

01,

043

1,17

861

195

21,

005

980

1,26

461

3

Ter

min

ees

Num

ber

722

809

763

808

443

724

801

740

901

439

SDA

type

mea

n/U

.S. m

ean

Rat

io1.

001.

121.

061.

12.6

11.

001.

1!1.

021.

24.6

1

Wel

fare

:M

ean

tota

l pro

gram

cos

t--

----

----

----

----

Part

icip

ants

Num

ber

291

347

294

294

171

292

342

279

325

179

Ter

min

ees

Num

ber

205

255

203

188

117

210

258

198

218

121

SDA

type

mea

n/U

.S. m

ean

Rat

io1.

001.

24.9

9.9

2.5

71.

001.

230.

941.

04.5

8

You

th:

Mea

n to

tal p

rogr

am c

ost

Dol

lars

1,02

8,31

41,

182,

717

1,09

3,97

61,

191,

589

539,

685

1,05

2,45

31,

187,

563

1,10

0,69

81,

300,

833

566,

493

SDA

type

mea

n/U

.S. m

ean

Rat

io1.

001.

151.

161.

16.5

21.

001.

131.

051.

24.5

4

Part

icip

ants

Num

ber

763

837

802

952

448

762

805

798

1,02

946

2

Ter

min

ees

Num

ber

575

659

584

669

332

587

645

588

766

350

SDA

type

mea

n/U

.S. m

ean

Rat

io1.

001.

151.

021.

16.5

81.

001.

101.

001.

30.6

0

= N

ot A

vaila

ble.

(1)

One

SD

A (

whi

ch w

as a

met

ro S

DA

) w

as d

ropp

ed f

rom

the

PY86

adu

lt co

st c

alcu

latio

ns b

ecau

se it

s re

port

ed v

alue

was

far

bel

ow a

ll ot

hers

.

Sour

ce: J

TPA

Ann

ual S

tatu

s R

epor

t dat

a, a

s co

mpi

led

by th

e U

.S. D

epar

tmen

t of

Lab

or, O

ffic

e of

Str

ateg

ic P

lann

ing

and

Polic

y D

evel

opm

ent.

(

BE

ST C

ITY

AV

P,E

,411

.E

only for all adults and for youth. Costs for welfare adults arenot segregated from the all-adult total. The number ofparticipants is the total number of people who were in theprogram during the program year, whether or not they finishedtraining. The difference between participants and terminees in agiven program year is the number of people who were in theprogram at the end of the program year but who had not yetcompleted training. For example, if a person began training onJune 1, 1986, but did not complete the program until August 1,1986, he or she would be counted as a participant in both PY85and PY86 and as a terminee in PY86.

The data in table 5 indicate that metro dominant and nonmetrodominant mean program cost and size were comparable with metrolevels in both the all-adult and youth programs. Metro programsdid, however, have a considerably higher average number ofwelfare adult terminees. The average nonmetro program wasconsiderably smaller on all measures than those in each of theother three SDA-type categories.

Five socioeconomic measures, displayed in table 6, are containedin the SDA-level data provided by the Department of Labor. Theseestimates are made by DoL for modeling purposes and are notreported directly by the SDA's.

The socioeconomic data indicate a steady decline in populationdensity from metro SDA's across the intermediate SDA types to thenonmetro SDA's. This is the expected finding. There is alsoconsistent variation from metro through nonmetro on the fourother variables. The more nonmetro the SDA type, the lower theprevailing wage rates and the higher the rates of unemploymentand families in poverty.

Relationship of Program Act:Lvity Levels to Size ofLabor Force, Unemployment, and Economic Disadvantage

Table 7 provides some summary measures of PY87 program activityas it related to the size of the civilian labor force, the numberof unemployed individuals, and the number of economicallydisadvantaged within each SDA. Data for program activity aretaken from the JTPA Annual Status Report (JASR) data. Civilianlabor force, unemployment, and economically disadvantaged datawere also provided by DoL.

Data on the number of economically disadvantaged are from aspecial tabulation of 1980 Census of Population data provided toDoL by the Bureau of the Census. These counts are consistent withthe definition of economically disadvantaged used in the Title1I-A allocation formula. DoL has used these specific totals forseveral program years to allocate Title II funds to the States.The States, in turn, have used them to allocate funds amongSDA's. The reason for their repeated use is that ro more recentsource exists of county-level data on the number of economicallydisadvantaged.

12

Tab

le 6

--So

cioe

cono

mic

cha

ract

eris

tics

of s

ervi

ce d

eliv

ery

area

s by

SD

A ty

pe, P

Y86

and

PY

87

PY86

PY86

unw

eigh

ted

mea

nsPY

87PY

87 u

nwei

ghte

d m

eans

unw

eigh

ted

by S

DA

type

unw

eigh

ted

by S

DA

type

Mea

sure

U.S

.m

ean

n=61

0

Met

ro

n=26

2

Met

rodo

min

ant

n=13

5

Non

met

rodo

min

ant

n=84

Non

met

ron=

129

U.S

.m

ean

n=61

0M

etro

n=26

1

Met

rodo

min

ant

n=13

6

Non

met

rodo

min

ant

n=84

Non

met

ro

n=12

9

Ave

rage

yea

rly

earn

ings

per

job,

Dol

lars

all s

ecto

rs17

,540

19,3

1017

,150

15,8

1015

,470

18,1

6720

,120

17,6

7016

,300

15,9

50

Ave

rage

yea

rly

earn

ings

per

job,

reta

il/w

hole

sale

12,1

1613

,700

11,6

0010

,600

10,4

0012

,489

14,3

0011

,900

10,8

0010

,600

Perc

ent

Une

mpl

oym

ent r

ate

7.81

6.48

7.97

8.36

9.96

7.40

6.10

7.61

8.15

9.30

Perc

enta

ge o

f fa

mili

es b

elow

the

pove

rty

leve

l, 19

809.

598.

009.

5011

.40

11.8

09.

598.

009.

5011

.40

11.8

0

1,00

0 pe

r sq

uare

mile

Popu

latio

n de

nsity

.74

1.61

.15

.06

.04

.74

1.61

.15

.06

.04

Not

e: V

alue

s fo

r th

e pe

rcen

tage

of

fam

ilies

bel

ow th

e po

vert

y le

vel a

nd f

or p

opul

atio

n de

nsity

wer

e on

e-tim

e es

timat

es a

nd a

re th

e sa

me

for

both

pro

gram

year

s.

Sour

ce: J

TPA

Ann

ual S

tatu

s R

epor

t dat

a, a

s co

mpi

led

by th

e U

.S. D

epar

tmen

t of

Lab

or, O

ffic

e of

Str

ateg

ic P

lann

ing

and

Polic

y D

evel

opm

ent.

r4

BE

ST C

OPY

Ani

tIL

LE

Table 7Measures

of a

ssoc

iatio

n am

ong

prog

ram

act

ivity

, lev

els

of e

cono

mic

dis

adva

ntag

e,em

ploy

men

tan

d un

empl

oym

ent b

y SJ

A ty

pe,PY87

Item

Num

ber

of c

ases

Wei

ghtin

gU

.S.

mea

n

PY87

mea

n by

SD

A ty

pe

Met

ron=

256

Met

rodo

min

ant

n=13

6

Non

met

rodo

min

ant

n=83

Non

met

ron=

124

Dol

lars

Adu

lt pr

ogra

m e

xpen

ditu

re p

er m

embe

rn=

598

Unw

eigh

ted

96

810

13of

the

civi

lian

labo

r fo

rce

Wei

ghte

d10

89

1213

Tot

al p

rogr

am e

xpen

ditu

re p

er m

embe

rn=

598

Unw

eigh

ted

1511

1518

21of

the

civi

lian

labo

r fo

rce

Wei

ghte

d17

1416

2222

.Z

1, lt

pro

gram

exp

endi

ture

per

n=59

7U

nwei

ghte

d12

010

611

413

214

4un

empl

oyed

per

son

Wei

ghte

d12

812

312

014

214

3

Adu

lt pr

ogra

m e

xpen

ditu

re p

er e

cono

mic

ally

n=59

9U

nwei

ghte

d24

2224

2427

disa

dvan

tage

per

son

Wei

ghte

d24

2225

2628

Tot

al p

rogr

am e

xpen

ditu

re p

er e

cono

mic

ally

n=59

9U

nwei

ghte

d41

3743

4245

disa

dvan

tage

d pe

rson

Wei

ghte

d42

3943

4546

Adu

lt pr

ogra

m e

xpen

ditu

re p

er e

cono

mic

ally

n=59

9U

nwei

ghte

d42

3944

4348

disa

dvan

tage

d ad

ult

Wei

ghte

d43

4044

4548

You

th p

rogr

am e

xpen

ditu

re p

er e

cono

mic

ally

n=59

9U

nwei

ghte

d14

012

514

915

015

2di

sadv

anta

ged

yout

hW

eigh

ted

150

134

161

167

166

Not

es: S

ee f

ootn

otes

at e

nd o

f ta

ble

Con

tinue

d--

Tab

le 7

-Mea

sure

s of

ass

ocia

tion

amon

g pr

ogra

m a

ctiv

ity, l

evel

s of

eco

nom

ic d

isad

vant

age,

em

ploy

men

t,an

d un

empl

oym

ent b

y SD

A ty

pe, P

Y87

-Con

tinue

d

Mea

n by

SD

A ty

pe

Item

Num

ber

of c

ases

Wei

ghtin

gU

.S.

mea

nM

etro

n=25

6

Met

rodo

min

ant

n =

136

Non

met

rodo

min

ant

n=83

Non

met

ron

= 1

24

Perc

ent

Adu

lt te

rmin

ees

as a

per

cent

age

of th

en=

598

Unw

eigh

ted

.44

.31

.44

.54

.63

civi

lian

labo

r fo

rce

Wei

ghte

d.5

3.4

5.5

0.6

7.7

0

All

term

inee

s as

a p

erce

ntag

e of

the

n=59

8U

nwei

ghte

d.8

0.5

7.7

9.9

91.

14

civi

lian

labo

r fo

rce

Wei

ghte

d.9

7.8

0.8

81.

281.

28

Adu

lt te

rmin

ees

as a

per

cent

age

of a

lln=

597

Unw

eigh

ted

6.16

5.24

6.08

7.23

1.42

unem

ploy

edW

eigh

ted

7.24

6.91

6.70

8.23

8.08

Adu

lt te

rmin

ees

as a

per

cent

age

of a

lln=

599

Unw

eigh

ted

1.22

1.06

1.29

1.35

1.37

econ

omic

ally

dis

adva

ntag

edW

eigh

ted

1.34

1.22

1.37

1.48

1.53

All

term

inee

s as

a p

erce

ntag

e of

all

n=59

9U

nwei

ghte

d2.

171.

922.

292.

382.

44

econ

omic

ally

dis

adva

ntag

edW

eigh

ted

2.37

2.16

2.39

2.65

2.72

Adu

lt te

rmin

ees

as a

per

cent

age

of e

cono

mic

ally

n=59

9U

nwei

ghte

d2.

181.

912.

322.

372.

43

disa

dvan

tage

d, a

ges

22 a

nd o

ver

Wei

ghte

d2.

402.

222.

462.

632.

68

All

yout

h te

rmin

ees

as a

per

cent

age

of e

cono

mic

ally

n=59

9U

nwei

ghte

d7.

806.

748.

078.

669.

11

disa

dvan

tage

d, a

ges

16 to

21

Wei

ghte

d9.

007.

968.

9610

.40

10.9

0

Not

e: O

ne S

DA

was

dro

pped

fro

m th

e ca

lcul

atio

n of

the

two

civi

lian

labo

r fo

rce

mea

sure

s du

e to

an

extr

eme

valu

e. T

wo

SDA

's w

ere

excl

uded

for

the

sam

e re

ason

s fr

om th

e ad

ult/u

nem

ploy

ed c

alcu

latio

n.

Sour

ce: U

.S. D

epar

tmen

t of

Lab

or. O

ffic

e of

Str

ateg

ic P

lann

ing

and

Polic

y D

evel

opm

ent.

C 9

('

4", t

";

The civilian labor force and unemployment rate data were providedto the D0L program office by the Bureau of Labor Statistics(BLS). These data are for PY87 (July 1987-June 1988) .10

Seven expenditure-based ratios are presented in table 7. Theseratios were calculated by dividing the relevant program costvariable by the relevant labor force or disadvantaged count. Theterminee-based measures in the continuation of table 7 use thesame denominator as these seven cost measures but employ termineecounts in the numerator instead of expenditure levels.

In addition, I weighted the results to measure the effect ofprogram size on the expenditure ratios. For the expenditure-based ratios, the weighted values use total SDA program costs forthe relevant terminee population (all adult or youth) as theweighting variables. For the terminee-based ratios, termineecounts are used.

The patterns displayed by these ratios are strong and consistent.With few exceptions, both weighted and unweighted values show anincrease in level across SDA types from metro to nonmetro. In nocase was the metro SDA value higher than the value for either thenonmetro dominant or nonmetro SDA categories. The nonmetro SDAmean value was always at least 16 percent higher than the metroSDA mean. It was, indeed, approximately double on thoseunweighted measures that used labor force or unemployment countsin the denominator.

A comparison of the weighted and unweighted values shows that thelarger programs within each category had higher than averagevalues on almost all measures. This influence was strongest inthe metro and nonmetro dominant categories. The metro SDAincrease was sharpest relative to the nonmetro SDA increase onthose ratios using labor force or unemployment counts in thedenominator. Size had much less influence on relativemetro/nonmetro values for those ratios related to thedisadvantaged.

"The number of SDA's in this labor force/disadvantageddatabase is 599, 11 SDA's fewer than the number submitting JASRreports in FY87. This discrepancy arises because the number ofSDA's dropped from 610 in PY87 to 599 in PY88, and DoL used thePY88 structure in compiling the labor force/disadvantaged data.These two data structures were reconciled without greatdifficulty, however. On variables utilizing the labor force data,there were one or two SDA's with very extreme values. Thesecases were dropped in calculating national and SDA type meansbecause they substantially affected the mean values. The numberof SDA's on a particular measure in table 7 may thus be slightlyless than 599..-

16J

The above results prompt the question: why are these variationsso systematic? The data permit examination of some hypothesesregarding this issue.

As table 7 indicates, the largest metro to nonmetro differentialsare found in the total expenditures per member of the civilianlabor force. The correlations among the data elements show thatvariatiel in the unemployment rate was a strong predictor ofvariation in expenditures per member of the civilian labor force.This relationship is shown in table 8. The quartiles are definedin terms of expenditures per member of the civilian labor force.Among the four quartiles, the top quartile contains the 25percent of SDA's (about 150) with the highest expenditures percivilian labor force member. The table shows how the unemploymentrate drops sharply across the quartiles. This suggests that thehigher the SDA's unemployment rate, the more JTPA funding itreceived per member of its labor force. This finding isconsistent with the heavy reliance on interstate differences inunemployment rates to allocate Title II-A funds among the States.

I also performed a simple linear regression to measure thisassociation more systematically. The results suggest that thevariation in the unemployment rate explained about 71 percent ofthe variation in total program expenditures per labor forcemember."

A second finding, also depicted in table 8, was that expendituresper disadvantaged person living in the SDA rose sharply withexpenditure per labor force member. SDA's in the highest quartileof expenditures per labor force member had mean expenditures perdisadvantaged person over twice the mean of the lowest quartile.If unit program costs were comparable from one SDA to another, aneconomically disadvantaged person living in a top-quartile SDA(with high expenditures per labor force member) had about twice

VAs noted in footnote 3, unemployment rate thresholds of 6.5percent and 4.5 percent are used to determine State eligibilityfor two-thirds of available funds. A finer level of analysiscentered around these thresholds was not pursued because theexplanatory power of the simple unemployment rate was very high,and use of this single rate facilitated presentation of thefindings. However, the threshold data clearly reinforce theinterpretation presented in the main text. For each threshold,the more nonmetro the SDA category, the higher the percentage ofSDA's with unemployment rates above that threshold. Thepercentage of SDA's with unemployment rates above 6.5 percentincreases steadily from 31 percent for metro SDA's to 53 percentfor metro dominant, 60 percent for nonmetro dominant, and 72percent for nonmetro SDA's. For the 4.5-percent threshold, thecorresponding percentages are 64 percent, 89 percent, 90 percent,and 93 percent. Since expenditure ratios do not appear related tothe third formula criterion (that is, the number of economicallydisadvantaged), incorporation of this threshold information wouldprobably explain a substantial part of the remaining variation inthe spending ratios, as shown in table 8.

17

the chance of receiving training as one living in a bottom-quartile SDA (with low expenditures per labor force member).

Table 9 shows that, overall, total expenditures per disadvantagedperson varied across SDA's by a factor of 14. Mean expenditures

per disadvantaged person in the highest expenditure quartile wasalmost three times the mean expenditures in the lowest quartile.

When differences in unemployment rates are controlled for, there

is a significant negative association between expenditures perdisadvantaged person and the number of disadvantaged personsliving in the SDA (table 9) .12 Since the number of disadvantagedis the third specific criterion used in the Title II-A allocationformula, it seems anomalous that its allocative effect appears to

be more than offset by the influence of unemployment rates, which

had a slight negative correlation with the absolute number of

disadvantaged.

There was also a weak negative association between theexpenditures per disadvantaged person and the percent of SDAfamilies with incomes below the poverty line. Unlike the number

of disadvantaged, however, the poverty rate had a relativelystrong positive association with both the expenditures per laborforce member and the unemployment rate.13 Since the poverty rate

is not a part of Title II-A's allocation formula, it is likely

that higher poverty rate SDA's tended to have higher expendituresper labor force member due solely to their higher unemploymentrates.

To test this interpretation more formally, I regressedexpenditures per labor force member against the unemployment rateand the poverty rate. The results strongly suggest that thepoverty rate provided no additional explanatory power beyond that

provided by the unemployment rate.14 In other words, the povertyrate, by itself, did not appear to directly influence patterns of

program expenditure. Higher expenditures per labor force member

12When expenditures per disadvantaged person were regressedagainst both the unemployment rate and the number of economicallydisadvantaged persons living within the SDA, the variable for the

number of economically disadvantaged had a negative t-ratio,which was significant at the 0.001 level. In contrast, the t-ratio for the unemployment rate variable was positive and alsosignificant at the 0.001 level.

13The Pearson correlation coefficients between the unemploymentrate and the poverty rate and between the unemployment rate andthe expenditures per member of the civilian labor force bothexceeded 0.5.

%When the expenditures per labor force member were regressedagainst the unemployment rate alone, the R-square value was 0.71.

It remained virtually the same when the poverty rate was included

as a second independent variable.

18

Tab

le 8

Unw

eigh

ted

mea

n un

empl

oym

ent r

ates

and

exp

endi

ture

per

dis

adva

ntag

ed p

erso

n by

quar

tiles

of

expe

nditu

re p

er m

embe

r of

the

civi

lian

labo

r fo

rce,

PY

87

Item

Uni

tT

op q

uart

ileSe

cond

qua

rtile

Thi

rd q

uart

ileB

otto

m q

uart

ile

Exp

endi

ture

per

mem

ber

of th

eci

vilia

n la

bor

forc

eD

olla

rs28

1611

6

Une

mpl

oym

ent r

ate

Perc

ent

10.5

7.3

5.7

4.3

Exp

endi

ture

per

dis

adva

ntag

ed p

erso

nD

olla

rs55

4536

27

Sour

ce: U

.S. D

epar

tmen

t of

Lab

or, O

ffic

e of

Str

ateg

ic P

lann

ing

and

Polic

y D

evel

opm

ent,

Div

isio

n of

Per

form

ance

Man

agem

ent a

nd E

valu

atio

n.

Tab

le 9

Unw

eigh

ted

mea

n ex

pend

iture

per

dis

adva

ntag

ed p

erso

n an

d un

wei

ghte

d m

ean

num

ber

ofdi

sadv

anta

ged

pers

ons

by q

uart

iles

of e

xpen

ditu

re p

er d

isad

vant

aged

per

son,

PY

87

Item

Uni

tT

op q

uart

ileSe

cond

qua

rtile

Thi

rd q

uart

ileB

otto

m q

uart

ile

Exp

endi

ture

per

dis

adva

ntag

ed p

erso

n

Num

ber

of d

isad

vant

aged

per

sons

Dol

lars

Num

ber

63

45,3

84

44

57,2

67

33

78,9

07

23

89,4

51

Sour

ce: U

.S. D

epar

tmen

t of

Lab

or, O

ffic

e of

Str

ateg

ic P

lann

ing

and

Polic

y D

evel

opm

ent.

r4

in areas with higher poverty rates probably occurred becausepoverty rates were positively associated with unemployment rates.

The positive interaction between poverty and unemployment rateswas not strong enough, however, to produce above-averageexpenditures per disadvantaged person in areas of higher povertyrates. In fact, as noted, there was a weak negative correlationbetween the poverty rate and the expenditures per disadvantagedperson. This outcome is likely attributable to the higher ratioof economically disadvantaged persons in areas with higherpoverty rates relative to the number in the labor force. Thishigher ratio more than offset the advantage created for higherpoverty areas by use of the unemployment rate in the allocationformula.

As for poverty in nonmetro areas (table 10), expenditure levelsin SDA's containing "persistent poverty counties" can be comparedwith those which have some nonmetro population but are notpersistent poverty areas. Persistent poverty counties constituteone category of the ERS Ross-Green typology of nonmetro countiesand are defined as those with average per capita incomes in thebottom quintile of all U.S. counties in 1950, 1959, 1969, and1979. These counties are located in 73 of the 343 metro dominant,nonmetro dominant, and nonmetro SDA's. These 73 SDA's had, in

turn, mean poverty rates 6 percentage points higher (15.6compared with 9.5) than the 270 metro dominant, nonmetrodominant, and nonmetro SDA's containing no persistent povertycounties. As shown in table 10, the persistent poverty SDA'sconform to the general patterns found above for the poorercounties in all SDA's: mean unemployment rates were 1.1percentage points higher, and expenditures per labor force memberwere 24 percent higher, but expenditures per economicallydisadvantaged person were 16 r "rcent lower.

Table 10Unemployment rates, average expenditure per disadvantaged person and per memberof the civilian labor force, in and out of "persistent poverty county" SDA's, PY87

SDA's containing one or more Metro dominant, nonmetroItem "persistent poverty counties" dominant, and nonmetro SDA's

containing no "persistent poverty county"

Unemployment rate

Percentage of families below thepoverty line

Expenditures per disadvantagedperson

Expenditure per member of thecivilian labor force

Percent

8.7

15.6

Dollars

38

7.6

9.5

45

21 17

Source: JTPA Annual Status Report Data, as compiled by the U.S. Department of Labor, Office of Strategic Planning and Policy Development.

20

t.)

1

A companion piece to this reports shows that metro SDA's tendto have higher unit costs per terminee. Unit cost per terminee ina given program year is defined as total Federal expendituresunder the Title II-A program divided by the number of programterminees. Higher unit costs might come from differences inparticipant characteristics (for example, a more or a less job-ready clientele), differences in cost levels (for instance, theneed to pay higher trainer salaries and rents), differences inoperating efficiency, or differences in the types of servicesprovided. Whatever the reasons, not only is a smaller absoluteamount spent per disadvantaged person in the typical metro SDA,but a dollar of expenditure does not go as far toward trainingthe average terminee.

In summary, then, the unemployment rate appears at present to bethe key factor in defining patterns of the expenditures permember of the civilian labor force and the expenditures perdisadvantaged person in the Title li-A program. This patternappears to particularly benefit the more rural SDA's because theytend to have higher than average unemployment rates and lowerthan average unit costs. Secondly, expenditures per disadvantagedperson vary widely across SDA's and are negatively associatedwith the number of disadvantaged. The wide variance inexpenditures per disadvantaged person means that disadvantagedpersons living in different SDA's face very different odds ofreceiving training. Finally, expenditures per disadvantagedperson are somewhat lower in higher poverty areas. This suggeststhat jurisdictions least able to afford training on their ownreceive a lower level of support per disadvantaged person.

Alternative Allocation Scenarios

The apparent negative correspondence between the expenditures pereconomically disadvantaged person and both the local poverty rateand the number of economically disadvantaged raises the questionof how funding allocation would be changed if funds weredistributed in a manner that more directly reflected thesemeasures of economic distress. And further, in what ways wouldsuch a reallocation affect the metro/nonmetro distribution ofprogram activity? To address these issues, two scenarios can bepropounded:

Scenario 1: Program funds are allocated to SDA's based on thenumber of economically disadvantaged in each SDA.

Scenario 2: Program funds are allocated to SDA's based on thenumber of economically disadvantaged weighted by the poverty ratein each SDA.

15See John M. Redman, Metro/Nonmetro Program Performance UnderTitle II-A, Job Training Partnership Act, AGES 9072, Econ. Res.Serv., U.S. Dept. Agr., Dec. 1990.

21

The criterion used in scenario 1 is the same as that currently

used by the Title II-A program in the third part of the existing

allocation formula, that is, the number of economicallydisadvantaged in the locality. The criterion used in scenario 2

will distinguish between SDA's that had, in 1980, similar numbers

of economically disadvantaged but different poverty rates. If two

SDA's have the same absolute number of disadvantaged, the SDA

with the higher poverty rate will be favored.

Because aggregate State poverty rates had not been calculated by

DoL, the computations of the results of these two scenarios were

performed in a manner very different from the existing Title II-A

allocation procedure. As discussed in the earlier overviewsection, the current funding plan employs a two-tier approach.

First, each State government receives an allocation from theFederal Government; the State then allocates funds to theindividual SDA's. For our two scenarios, we simply bypass the

State and allocate funds directly to the SDA, based on the SDA's

values for the applicable criteria. If a two-tier approach wasapplied to this exercise, the results might be very different.

There is no a priori way to estimate the magnitude of this

potential difference.

The procedure used here is conceptually similar, however, to the

"bottom up" approach to funding contained in both theadministration's and in Senator Simon's proposed amendments to

the JTPA legislation. Under that approach, funding allocations

are made to each SDA instead of to each State. The individual

SDA's share of the total funding is determined by comparing its

values on the allocation criteria (for instance, number ofeconomically disadvantaged) against those for all other SDA's. A

State's total funding becomes simply the sum of Title II-A funds

allocated to its individual SDA's.

Table 11 shows funding by region under each scenario expressed as

a percentage of PY87 Title II-A expenditure, as reported on the

JASR submissions. Under the first scenario (using the absolute

number of disadvantaged), New England and Ncei York/New Jersey

experience substantial increases in their allocations, while theindustrial Midwest (the East North Central region) experiences asharp decline. This result suggests that the Northeast has lowunemployment levels relative to the number of economicallydisadvantaged, while the Midwest tends to have high unemploymentrelative to the number of disadvantaged.

The results using the second scenario (weighting the number ofdisadvantaged by the poverty rate) are substantially different.

Here, the Southeast and West South Central regions join the NewYork/New Jersey area as major beneficiaries of the funding

change. This outcome suggests three things. First, the New

York/New Jersey area appears (based on 1980 data) to have bothrelatively large numbers of disadvantaged and relatively highpoverty rates, since its funding levels increase substantiallyunder either scenario. Second, the Southeast and West SouthCentral regions have very high poverty rates relative to their

share of economically disadvantaged. Including the poverty rate

22

Tab

le 1

1Fun

ding

allo

catio

n le

vels

und

er a

ltern

ativ

e al

loca

tion

scen

ario

s, P

Y87

Reg

ion/

SDA

type

Exi

stin

gal

loca

tion

Scen

ario

1Sc

enar

io 2

(cri

teri

on: n

umbe

r of

(cri

teri

on: n

umbe

r of

dis

adva

ntag

eddi

sadv

anta

ged

indi

vidu

als

wei

ghte

din

divi

dual

s)by

the

pove

rty

rate

)

Reg

ion:

----

Perc

ent-

---

Perc

ent o

f ex

istin

g al

loca

tion

New

Eng

land

100

142

107

New

Yor

k/N

ew J

erse

y10

012

814

2M

id-A

tlant

ic10

096

88So

uthe

ast

100

107

13::.

'

Eas

t Nor

th C

entr

al10

077

60W

est S

outh

Cen

tral

100

9412

1M

idw

est C

entr

al10

011

397

Mou

ntai

n/W

est N

orth

Cen

tral

100

105

87So

uthw

est

100

107

91N

orth

wes

t10

091

66

SDA

type

:M

etro

100

109

99M

etro

dom

inan

t10

095

97N

onm

etro

dom

inan

t10

089

108

Non

met

ro10

087

99

Sour

ce: J

TPA

Ann

ual S

tatu

s R

epor

t Dat

a, a

s co

mpi

led

by th

e U

.S. D

epar

tmen

t of

Lab

or, O

ffic

e of

Str

ateg

ic P

lann

ing

and

Polic

y D

evel

opm

ent.

S

BE

ST C

OPY

AV

AIL

AB

LE

.r.

in the formula makes a very big difference for each of theseregions. Third, New England benefits considerably from relianceon absolute numbers of disadvantaged but loses most of thisadvantage when the poverty rate is included.

At the other extreme, the Northwest joins the industrial Midwestas a big loser under the second scenario. This indicates thatthe industrial Midwest has not only relatively low numbers ofeconomically disadvantaged but relatively low poverty rates as

well. The Northwest shows this same pattern, though to asomewhat lesser extent than the Midwest.

Also measured was the differential metro/nonmetro effects ofthese two scenarios. The results are also presented in table 11.Under the first scenario (using the number of economicallydisadvantaged alone), metro SDA's experience a moderate increasein the total amount of program funds made available to them.This increase comes at the expense of the remaining three SDAcategories. The size of the decrease in these three othercategories increases as the category becomes more nonmetro.This outcome gives a broad indication of the benefit nonmetroareas receive from emphasis on the unemployment rate in theactual program formula. As these findings indicate, the effect ofthe unemployment rate more than offsets the effect of includingthe absolute number of disadvantaged as the third leg of theTitle II-A funding formula.

Under the second scenario (the number of disadvantaged weightedby the poverty rate), little change from the existing allocationoccurs, except that nonmetro dominant SDA's receive an increasedshare of funding. Here, the poverty rate appears largely tosubstitute for the unemployment rate in directing funding tononmetro areas well in excess of what is received using only theabsolute number of disadvantaged.

Nonmetro areas might be better served over the long term by aformula emphasizing the poverty rate rather than the unemploymentrate. In recent years, the metro/nonmetro difference inunemployment rates has been steadily declining and for the year1989 stood at just 0.5 percentage point. This reduction has notproduced a similar narrowing of poverty rates, however. In fact,the metro/nonmetro poverty rate differential, defined in terms of

the percentage of individuals living below the poverty line,increased from 3.5 percentage points to 4.4 percentage pointsbetween 1980 and 1987. Use of the poverty rate, then, may offernonmetro areas comparable but more stable funding advantagesrelative to those offered in recent years by the unemploymentrate.

Conclusions

The JTPA program is the main Federal effort to enhance theemployability of the economically disadvantaged. It is of centralimportance to the many areas with no other ongoing source of

24

employment and training funding and whose local revenue bases aretoo small to support sustained independent efforts.

For those who work in the field of rural development, theforegoing review suggests that, because of its reliance onunemployment rates, the current Title II-A allocation formulahas, up to this point, produced higher expenditures perdisadvantaged person in the more rural SDA's. This has been adirect result of the unusually high unemployment ratesexperienced by nonmetro areas in recent years.

This finding is not necessarily inconsistent with past studies16which argued that use of standard unemployment rates firallocating Federal funding discriminates against nonmetro areas.Such discrimination arises, it is contended, because standardunemployment rates fail to capture various dimensions of truelabor distress more prevalent in nonmetro areas (such asinvoluntary part-time employment, higher levels of discouragedworkers, or low and irregular income among farm proprietors).17Echoing this theme, a recent Joint Economic Committee report onemployment and training programs stated:

Allocation of Federal program funds reveals an urban bias.Despite the fact that nonmetro unemployment andunderemployment rates exceed metro measures by a third to ahalf, nonmetro areas receive only about 13 percent ofemployment and training funds. Federal procurement programsalso show a pronounced urban leaning. Another illustrationof implicit Federal discrimination is the funding formulafor the Job Training Partnership Act of 1982. Two-thirds ofthe funding was allocated according to unemployment figuresof little relevance to the rural employment picture.Because of under-reporting of unemployment and the way theformula was designed, rural areas were denied over $100

16For a discussion of this issue, see Sigurd R. Nilsen,Assessment of Employment and Unemployment Statistics forNonmetropolitan Areas, RDRR-18, Econ. Res. Serv., U.S. Dept.Agr., Dec. 1979. For a recent discussion of the use of andalternatives to standard unemployment rates see Richard J.Reeder, "Targeted State Aid to Distressed Rural Communities,"Publius, 19 (2), spring 1989, pp. 143-160.

VThe standard unemployment rate calculated and published by

the Bureau of Labor Statistics (BLS) is total unemployed as apercentage of the labor force, including the resident ArmedForces (U-5a). Alternative rates are also made available by BLSbut are not available at the county level. These alternativesinclude components which reflect additional dimensions of labordistress (such as the number of discouraged workers or thoseemployed part-time for economic reasons). The standardunemployment rate is thus only one of several measures which BLSitself has developed and one of the many others (for example, percapita income or the percentage change in unemployment) whichmight also be used in formula allocations.

254

'-ft 4

million in funding from 1983 through 1985, according to apreliminary analysis by the GenerDl Accounting Office."

From this perspective, the higher Title II-A expenditure levels

in nonmetro areas are still well below levels justified ifcongressional intent was to use the unemployment rate as a proxy

for overall labor distress. Within the context of the Title II-Aprogram, however, use of the simple unemployment rate held some

advantage for nonmetro areas compared with alternative criteriawhich might have been applied, such as the number of economicallydisadvantaged or the number of unemployed.

Practitioners should be aware, however, that the Title II-Aallocation formula is one focus of an ongoing congressionalreview of the JTPA legislation. This review has importantramifications for the future of rural employment and training

activity. Although it is still unclear if or how the formulamight be specifically modified, two major proposals have alreadybeen presented, one from the Bush administration and one from

Senator Simon of Illinois.

These proposals reflect considerable "second guessing" regardingthe use of unemployment rates as the dominant allocation

criterion.19 A major concern has been the lack ofcorrespondence between the incidence of unemployment and ofeconomic disadvantage--a lack of correspondence evident in thedata presented above. As a consequence, both proposals woule.greatly reduce the role of unemployment in setting allocations.Instead, both would emphasize the number and concentration ofeconomically disadvantaged within the SDA.

As discussed above, increased emphasis in the allocation formula

on the number of economically disadvantaged would tend to work in

favor of metro areas whereas emphasis on a poverty rate-basedmeasure could work to both the short- and longer-term advantageof the more nonmetro areas. The net effect of the formula change

on metro/nonmetro allocation will thus likely depend on thebalance struck between the number and the concentration of

economically disadvantaged.

"Dale Jahr, "The Rural Political Economy: Change andChallenge," U.S. Congress, Joint Economic Committee, Sept. 1988.

19See, for example, Abt Associates Inc., "An Assessment ofFunding Allocation Under the Job Training Partnership Act,"Contract No. J-9-M-5-0051, U.S. Dept. of Labor, Aug. 1986.

26A

U.S. Government Printing Office 1990 2g2-9,,g/2154:74/