A job placement evaluation system

Material Information

A job placement evaluation system under the segmented labor market hypothesis
Martinez, Gaylene
Publication Date:
Physical Description:
ix, 177 leaves : illustrations ; 28 cm


Subjects / Keywords:
Job analysis -- Data processing ( lcsh )
Poor -- Employment -- Colorado -- Jefferson County ( lcsh )
Job analysis -- Data processing ( fast )
Poor -- Employment ( fast )
Colorado -- Jefferson County ( fast )
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )


Includes bibliographical references.
General Note:
Submitted in partial fulfillment of the requirements for the degree, Master of Arts, Department of Economics.
Statement of Responsibility:
by Gaylene Martinez.

Record Information

Source Institution:
University of Colorado Denver
Holding Location:
Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
21121698 ( OCLC )
LD1190.L53 1989m .M37 ( lcc )

Full Text
B.S. Eastern Montana College, 1984
Copyright (C) 1989
A Thesis submitted to the
Faculty of the Graduate School of the
University of Colorado at Denver in Partial Fulfillment
of the Requirements for the degree of
Master of Arts
Department of Economics

This thesis for the Master of Arts degree by
Gaylene Martinez
has been approved for the
Department of Economics
Date /
l. 19

Martinez, Gaylene (M.A. Economics)
A Job Placement Evaluation System Under the Segmented
Labor Market Hypothesis
Thesis directed by Professor Suzanne W. Helburn
The Job Training Partnership Act of 1982 (JTPA)
and the Family Support Act of 1988 (FSA) are two
programs developed by Congress to assist the poor in
the United States to achieve income self-sufficiency.
JTPA grants autonomy to states in the design and
structure of their job training programs and FSA
requires mandatory employment for some welfare clients.
Based on these two provisions, this research involves
the development of a job placement evaluation system
for use by social service agencies to assist the poor
in achieving income self-sufficiency.
This thesis describes the background and
development of a computerized program which can provide
advice and direction on employment to welfare mothers
and other low income women who are job training
participants. Use of this program will assist both the
mother and the social service agency to determine
whether a particular occupation will provide enough
income to meet the family's cost of living (COL)
requirements. The program calculates COL estimates for
a variety of family compositions. These calculations

consider the age, sex, and number of children within a
family as well as the mother's age. The COL categories
include: food, clothing, laundry, child care, transpor-
tation, housing, furniture, furnishings, supplies,
sales taxes, social security taxes, state and federal
income taxes and the child care and earned income tax
credits. The program then computes the hourly wage
that is needed to meet these COL needs. A comparison
of this hourly COL wage and the wage that the mother
will earn from a particular occupation shows whether or
not income self-sufficiency is achievable for the
The computerized program also provides a cost-
effective, internal evaluation procedure for the job
training program. The end result of this evaluation is
a percentage determination of where all job placement
clients fall with respect to four levels: self-
sufficient, potentially self-sufficient, probably not
self-sufficient and not self-sufficient.
The form and content of this abstract are approved,
recommend its publication.
S igned

This thesis is dedicated to all
ingle working mothers who live in poverty
yet live for self-sufficiency.

I wish to thank the following people for their
ideas, criticisms and support throughout the research
and writing of this thesis: Mr. Paul Kludt, Program
Planner/Evaluator of Jeffco Employment and Training;
Dr. Suzanne W. Helburn, Professor of Economics,
University of Colorado at Denver; and Dr. John R.
Morris, Professor of Economics, University of Colorado
at Denver.
I also wish to thank Debra Bornhoft for her
editorial comments and suggestions and Arlene Pacheco
for use of her computer and laser printer. Finally, I
thank all my friends who offered me encouragement and
kept me motivated during this project.

I. INTRODUCTION ............................... 1
Provisions of the Job Training Partnership
Act of 1982 ..............................4
Provisions of the Family Support Act of
1988................................... 10
Job Placement Evaluation System .......... 14
Purpose of the Thesis......................18
Thesis Arrangement.........................19
II. THEORETICAL FOUNDATION...................... 21
Poverty Defined .......................... 21
Self-Sufficiency Defined...................26
Segmented Labor Market Theory ............ 27
III. COST OF LIVING ESTIMATES......................40
Cost of Living Estimates...................40
Food Estimate...........................43
Clothing Estimate.......................46
Laundry Estimate ...................... 46
Household Furniture, Furnishings and
Supplies Estimate.....................48
Sales Tax Estimate......................50
Transportation Estimate.................50
Child Care Estimate. ...................51
Housing Estimate ...................... 52
Total Cost of Living Estimate...........55

CONTENTS (continued)
Tax Liability Estimate ............... 55
Gordon's Classification .................. 59
Occupation and Wage Survey................68
Job Earnings..............................72
Wage Security.............................74
Future Opportunity........................78
Skill/Education Requirement .............. 79 ,
Overall Job Rank..................... 79
Family Profiles .......................... 81
A. OCCUPATION AND WAGE SURVEY..................101

1987 Poverty Standard of Living Gross
Dual Labor Market Characteristics..........
Monthly Estimated Food Costs ..............
Family Size Food Cost Adjustment ..........
Family Clothing Estimate ..................
Percentage Increase by Family Size For
Household Furnishings, Supplies and
Monthly Cost Estimates for Household
Furnishings Furniture and Supplies . .
Average Weekly Child Care Costs............
Housing Guidelines ........................
Factor Characteristic Evaluation ..........
Family Profiles............................
Job Placement Evaluation Wage Comparisons.
Job Placement Evaluation Ranking System. .

This thesis describes a computerized program
for a job placement evaluation system which was
developed for use by the Jefferson County Employment
and Training Program. For any given client seeking job
placement, the computer program provides estimates of
family income requirements and identifies those
occupations which can currently satisfy or can
potentially satisfy those income requirements. Besides
determining income requirements, the program is also
useful in helping clients find jobs which will promote
job satisfaction for them. In addition to its function
as a counseling tool, the computer program is also
designed to be used for evaluating the effectiveness of
the county's Job Training Partnership Act program.
During the 1980s, numerous changes in public
policy occurred in the United States which regulate the
lives and well being of poor families headed by single
mothers. Two significant changes were the 1982 Job
Training Partnership Act (JTPA) which replaced the
Comprehensive Employment and Training Act (CETA) and
the Family Support Act of 1988 (FSA) which overhauled

the welfare system. Although these programs continue
to provide their same traditional functions, i.e., Aid
to Families with Dependent Children (AFDC) and job
training, Congress did establish specific provisions in
each Act in order (1) to provide for economic relief,
(2) to reduce the overall poverty level in the United
States and (3) to increase the employment of able-
bodied adults within poverty level households, thereby
expanding the tax revenue base and reducing the costs
of these social service programs.
Perhaps the most significant change that -
occurred as a result of FSA legislation is the
provision for mandatory employment. This stipulation
means that all eligible low-income and able-bodied,
single parent or two parent families with children over
three years of age, who received AFDC are now obligated
to seek and eventually secure employment. Congress
included the mandatory employment condition as a
prerequisite for some AFDC recipients to insure that
they would actively participate in achieving and
maintaining their own income self-sufficiency.
At the county level, where welfare reform will
be carried out, FSA requires social service departments
to work with county employment and training offices.
Since the enactment of JTPA in 1982, states and

counties now have the facilities and funds available to
assist AFDC mothers, low income adults and employment-
age youths to secure the necessary vocational
counseling, employment testing, job training and
occupation which would eventually elevate them from
poverty to income self-sufficiency.
The linkage of welfare reform with JTPA will
necessitate the restructuring of many social programs
to accommodate the employment mandate and congressional
intent. The idea for a job placement evaluation system
is relatively new; in fact, this system may be the
first of its kind. The job placement evaluation system
developed in this thesis is intended to provide a
practical method to assist job training agencies to
place their program participants in jobs which will in
time lead to income self-sufficiency. This
computerized program also offers an alternative to a
long-range job training program evaluation.
To fully understand the job placement
evaluation system, it is first necessary to explain the
overall structure of JTPA and the applicable provisions
of FSA which link them both. Since adequate employment
is a necessary prerequisite to reducing poverty, it
follows that job training may also be essential. The
following section explains why JTPA was enacted and

details how states have generally set up their job
training programs.
Provisions of the Job Training Partnership Act
The Job Training Partnership Act (JTPA) was
enacted by the United States Congress in 1982. It was
created from the reorganization of the Comprehensive
Employment and Training Act (CETA). The goals
established by Congress for this new program were to
increase stable employment, to increase employment
earnings, and to reduce welfare dependency of the
economically disadvantaged and dislocated workers.-3=/
JTPA was also designed to correct many of the
problems and abuses that plagued CETA. One major
problem was a lack of strong, long term links between
public agencies and the private sector. The solution
was the creation of a Private Industry Counsel (PIC)
system.^/ PIC provides a forum for representatives
from JTPA agencies and various members of private
industry to discuss and resolve job training problems
and other employment issues. In addition, PIC oversees
and directs the planning of job training programs which
enables the job training agency to organize its
1/ Project Team, JTPA Evaluation Design Project.
Volume 1: Overview, March 1986, pp. 1-5.
2/ JTPA Overview. Volume I, pp. 1-5.

programs around the needs of business as well as its
clients. This increase in communication between the
public and private sectors results in favorable
circumstances for all participants to achieve their
individual goals, i.e., job training for the client,
successful placement of the client for the job training
agency, and the resolution of an employer's labor
JTPA is funded by Congress and block grants are
allocated to every state. The determination of grant
amounts is based on individual state population and the
percentage of its population which falls under
"official" poverty standards. The state then allocates
funds to Service Delivery Areas (SDAs) within its
boundaries each of which serve specific geographic
areas. The following guidelines show how a state
generally designs its job training programs and
determines its SDAs.^/
1. State officials set goals, objectives and
performance standards of an SPA. The factors
considered in this process are:
The overall state environment
economic, political and
social conditions, as well as
state and federal
constraints; and
3/ JTPA Overview. Volume I, pp. 5-10.

how SDAs will be governed.
2. State officials determine services and
service providers. In determining SDAs, state
officials look at the following conditions:
Labor Unemployment rate; the number of
Conditions: unemployment insurance claims; percent of
the population on welfare; jobs with
increasing demand; jobs with declining or
stable demand;
Growth vs. decline; number of building
permits issued; number of new businesses
started during the preceding 12 months;
number of business closings and
bankruptcies over the preceding year;
Population: total population, specific
age groups; race/ethnic groups; sex;
income per capita;
Household: overall or by key target
groups; education levels; religious
groups; employment status by key target
Social Crime rates; degree of alcoholism/drug
Trends: abuse in the area; school dropout rates;
percent of population below poverty level;
percent of population on welfare; percent
of households with a female head; the
degree of serious health problems;
Political Relationship between local government and
Conditions: business; the legacy of CETA in the area;
the role of local government and business
in the support of past CETA program; the
number of jurisdictions the SDA serves;
the influence of local government(s) and
local business on SDA operations;

political influence of SDA sub-
contractors .
3. Resources are allocated to each SDA based
upon local conditions and participant/emplover needs.
At this point the individual SDA receives funding and
develops its own program based on the following:
its individual mission, goals and
performance standards?
the planning process;
Revenue, allocative, treatment and
access decisions;
the method of subcontracting and
selection of service providers; and
- the performance control system and
the nature of compliance and
technical assistance efforts.
4. Implementation is monitored through the
SDA's performance control system. The SDA provides
information to state officials on how well it has
achieved its objectives as well as the objectives of
the state and Federal job training programs.
The state of Colorado has generally followed
these guidelines in implementing its job training
4/ Subcontractors are employment service providers
contracted by the SDA to provide job training oppor-
tunities for JTPA participants and to establish contact
with local business. Some SDAs are heavily populated or
cover a large geographic area. Subcontracting allows
the SDA to cover the largest area, provide service to the
most people, and establish contact with the greatest
number of businesses.

program. Currently, there are thirteen SDAs in
Colorado and each SDA has enormous autonomy in the
design, structure and evaluation of its job training
program as well as the selection of its subcontractors.
The computerized job placement evaluation
system would accommodate the state and SDA job training
environment because it can assist the SDA in meeting
its goals and also provide a valuable tool for program
evaluation. For this thesis the Jeffco Service
Delivery Area was the primary focus.
Jeffco represents approximately 500,000 people
living within Jefferson, Gilpin and Clear. Creek
counties. It is estimated 52,000 people living within
the SDA are eligible for job training programs. The
Jeffco SDA will actually assist 2,800 participants
between July 1, 1988 and June 30, 1990.-^/
The major goals of the Jeffco SDA are to
increase economic development within the three-county
area and to place its clients into unsubsidized
employment. The program targets specific segments of
the population which include welfare recipients,
5/ Job Training Plan Jeffco Service Delivery Area
Consortium for Job Training Partnership Act Program Years
1988 and 1989 fJulv 1. 1988 through June 30. 1990^, April
12, 1988, pp. 1-3.

offenders, youth, handicapped, teen parents and school
Eligibility requirements for each target group
are based on the individual1s need for services and the
potential benefit he/she will derive from such
services. In other words, both the need and the
potential benefit must be achievable within a specific
time frame and within program limits. All eligible
participants undergo extensive testing to assess their
vocational, aptitude, interest and education levels.
Based on the test results, clients are directed into
one of four primary training programs; each designed to
serve specific levels of need. These training programs
include classroom training to acquire occupational
skills; on-the-job training for employers willing to
employ new hires; a job-finding club to teach job
search techniques and assist clients in self placement;
and temporary subsidized employment with either private
or public for-profit or non-profit organizations.In
addition, Jeffco offers support to its clients through
vocational counseling, job placement, transportation,
day care, medical services, etc.
Since the Jeffco SDA targets welfare recipients
for job training assistance, the FSA legislation will
6/ Ibid.

also affect these participants. Therefore, the
applicable provisions of FSA need to be examined. The
following section discusses Title II of FSA which
pertains to job opportunities and basic skills
Provisions of the Family Support Act of 1988
The Family Support Act of 1988 covers a wide
range of public service areas including child support
and the establishment of paternity, job opportunities
and basic skills, and family support services. The
most significant change for low income individuals is
the provision in the Act which mandates employment for
some AFDC recipients. Title II, Job Opportunities and
Basic Skills Training Program (JOBS) is the section
under FSA which outlines state responsibility for job
training. Thus, the main focus in this section is on
Title II and its implications for job training.
In general, the state has enormous autonomy in
deciding who must participate in JOBS. However, in
order to get the maximum allowable federal grant, the
state must target at least 55% of its JOBS funds toward
those clients who have received three continuous years
of AFDC payments, parents under 24 years who have
either not completed high school or have little work
experience, and family members where the "youngest

child is within two years of being ineligible for AFDC
because of age."2/
Congress also provided guidelines to determine
which clients would be exempt from participation.
Exempt clients are those who are:
(a) ill, incapacitated, or
(b) needed in the home because of
the illness or incapacity of
another household member;
(c) a parent or relative who
personally provides care to a
child under 1;
(d) a parent or relative who
personally provides care to a
child under 6, unless the
state guarantees child care
and limits participation to
20 hours or less per week;
(e) working 30 or more hours per week;
(f) a child under 16 or attending
full-time an elementary,
secondary, vocational, or
technical school;
(g) pregnant, after the first
three months of pregnancy;
(h) residing in an area of the
state where the program is
not available
7/ Mark Greenberg, JOBS Provisions of the Family
Support Act of 1988: Ten Early Choices. Center for Law
and Social Policy, November 1988, p. 3.
8/ Greenberg, Family Support Act of 1988. p. 8.

Aside from these specifically exempt clients, states
may also exempt eligible AFDC recipients if adequate
funds are not available to include everyone.
Each state was granted enormous autonomy in the
design of their JOBS programs. However, the state must
include the following services as part of its JOBS
programs: basic, remedial and high school education;
job skills training; job readiness activities, job
development and job placement, job search assistance,
on the job training, work supplementation, and
community work experience.Aside from these
requirements, the state decides how extensive and
expensive each service will be and the number of people
who will benefit from each program.
Finally, Title II provides social service
agencies with the power and authority to impose
penalties against any eligible, non-exempt client who
refuses to participate in the JOBS program or to accept
a satisfactory job offer. The penalties for participa-
tion refusal range from reductions in AFDC aid until
compliance is met to a mandatory reduction of AFDC
payments "for six months or until the failure to comply
ceases, whichever is longer."-^/ However, the state
9/ Greenberg, Jobs Provisions, p. 4.
10/ Greenberg, Jobs Provisions, p. 5.

cannot penalize a client for noncompliance "if the job
would result in a net loss of cash income to the
family" unless the state provides supplemental income
to make up the loss in wages.3i/
Under FSA, all states must have a JOBS program
which provides AFDC recipients with job training and
basic skill and education; and, the program must be
operational by October 1990.-3^/ However, because
federal funding for FSA was not expanded from previous
levels, money for an extensive JOBS program is
The lack of funds creates pressure on social
service agencies because.they are also required by law
to have definite JOBS participation rates among its
eligible, nonexempt clients.3^/ if the state fails to
reach the federal quotas after 1991, federal funding
for AFDC will be reduced. This puts welfare agencies
11/ Greenberg, JOBS Provisions, p. 6.
12/ Mark Greenberg, Family Support Act of 1988:
JOBS Program and Related Amendments. Requirements, Issues
and Options. Center for Law and Social Policy, November
1988, p.l.
13/ Ibid., p. 9. The eligibility participation
rates are gradual and in fiscal year 1990 a 7% JOBS
participation rate for nonexempt AFDC clients is
required. In 1991, the rate increases to 7% each half of
the fiscal year and in 1992, the rate increases to 11%
for each quarter of the fiscal year. By 1995, social
service agencies must have a JOBS participation rate of
20% for each month.

in a catch-22 situation: the agency must assist a
certain percentage of eligible clients in order to
receive the maximum federal funds; yet, the agency may
not have enough funds to provide this service to all
participants which fall within the federal quota.
The job placement evaluation system developed
in this thesis would alleviate some of the financial
burden placed on welfare agencies and other job
training programs. This system can also assist social
service agencies in meeting internal program goals as
well as provide a valuable tool for program evaluation.
Job Placement Evaluation System
The idea for the job placement evaluation
system developed in this thesis was originally
suggested by Paul Kludt, Program Planner/Evaluator,
Jeffco Employment and Training, Jefferson County,
Colorado. In "Job Placement Rating: A Proxy for Impact
Evaluation,"!i/ Kludt discussed an alternative to the
usual evaluation procedures based on the hypothesis
that the criteria for the effectiveness of job training
and placement programs is the client's placement in a
job with adequate pay and job satisfaction. He argued
14/ Paul R. Kludt, Job Placement Rating; A proxy for
Impact Evaluation. Jeffco Employment & Training,
Jefferson County Colorado.

that if jobs could be classified with regard to
quality, the county should try to place clients in jobs
which meet these standards. From his research, Kludt
determined that any method which would objectively
define job quality at placement would have two sig-
nificant advantages: (1) it would significantly aid
administrators and clients in the counseling and job
placement process and (2) it would potentially elimina-
te the need for a costly long-range evaluation process.
The first advantage of the job placement
evaluation system is its ability to provide job
training administrators with an objective method of
comparing the cost of living needs of program par-
ticipants, with respect to job placement wages and the
skill and education required for a particular job.
With this information, a job training participant can
be directed by his/her counselor toward those
occupations that meet the participant's cost of living
needs and match his/her skill and educational levels.
This advantage of this type of counseling is
important because it eliminates some of the uncertain-
ties of determining the proper job placement for the
client and it provides a more concrete foundation for
administrative decision-making. The information also

allows program participants to take an active part in
deciding their future occupations, not only in terms of
their financial requirements and educational
compatibility; but, also in terms of where his/her
personal occupational interests are centered. This is
an important consideration because it increases the
probability that the program participant will stay
employed in an occupation that satisfies his/her
interest needs as well.
The second advantage of the job placement
evaluation system is its potential to provide program
administrators with an economical procedure to
internally evaluate the job training program while
avoiding long-term evaluation projects. The benefit to
the job training agency is a substantial savings in
money, time and manpower. The evaluation can determine
the actual number of job training participants that
achieve income self-sufficiency, the number of
participants that could potentially reach self-
sufficiency in the near future, the participants which
probably will not satisfy their income self-sufficiency
needs unless a different job that pays higher wages is
found, and those that definitely will not meet their
income self-sufficiency requirements from the
occupation in which they are placed.

Kludt developed a theoretical approach to
classifying job quality. The foundation for this
system was a "quality of employment scale" which
categorized occupations in accordance with characteris-
tics identified in the literature on segmented labor
The quality scale developed by Kludt included
the following categories: marginal, stable/low, transi-
tional, self-sustaining, and cycle breaking. Each
quality level attempted to rank employment according to
income, stability, duration, entry requirements, job
satisfaction, future opportunities, the existence of
employee rights in the work place, and the attitude the
employer has toward the employee.
The job placement evaluation system proposed by
Kludt offered a cost-efficient means to provide a
wealth of information to job training program
administrators. However, the problem of how to measure
or quantify "quality of employment" posed a serious
obstacle to implementing the job placement evaluation
system. This thesis presents a computer program which
operationalizes Kludt's theoretical approach by
estimating, as objectively as possible, the "quality of
job placement" for job training agencies and creating a
procedure for evaluating the job training program.

Purpose of the Thesis
The computerized program developed in this
thesis has two purposes: (1) to establish a reliable
job placement evaluation system based on segmented
labor market theory; and (2) to provide a cost-saving
substitute for evaluating the overall effectiveness of
a job training program. Specifically, the program data
focuses on the Jefferson County Employment and Training
Program, the cost of living requirements for Jefferson
Colorado and the occupations and wages available in
this county as well as the immediate urban area.
This computerized program for a job placement
evaluation system determines:
(1) whether a specific job will
provide a given family with the
necessary income to meet or exceed
a poverty standard of living;
(2) whether the occupation will
potentially meet a particular
family's income needs;
(3) whether the education and skill
level of a client and a particular
occupation are compatible; and
(4) whether the job training program
meets its overall self-sufficiency
placement objective.
The job placement evaluation system required
collecting a large amount of information including:
(1) the prerequisite education and skill levels for a
large number of occupations; (2) the current maximum

and minimum wages for each occupation; (3) poverty cost
of living estimates for a variety of family composi-
tions; and (4) the federal, state and local tax inform-
The job placement evaluation system is set up
on a Lotus 123 spreadsheet and is expandable to include
more calculations and information as needed. The
program can also be easily modified to include cost of
living estimates for geographic areas other than
Jefferson County, Colorado.
Thesis Arrangement
The following chapters discuss more fully each
area in greater detail. Chapter II provides a theo-
retical basis for this study by defining poverty and
self-sufficiency, and examines the literature on the
segmented labor market theory. Chapter III outlines
the procedures for estimating cost of living at a
poverty standard of living which were necessary to
develop the job training evaluation system. Chapter IV
discusses David Gordon's classification method and
details the methodology used to develop the
occupational classification and wage survey data.
Chapter V explains the job placement evaluation system,
provides profiles for three different family
compositions based on the computerized program, and

shows how the system evaluates the job training
program. Finally, Chapter VI summarizes the results of
this project and concludes with possible policy
implications and directions for further research.

Theoretical Foundation
This chapter discusses the theoretical
background material essential to this project. The
development of the job placement evaluation system
required significant information which included:
* a definition of poverty and
the criticisms of this
* A definition of self-sufficiency; and
* A background of the economic
literature on the segmented
labor market hypothesis.
Poverty Defined
The following is the United States government's
"official" definition of poverty:
Poverty occurs at a level which is
three times the amount of income,
before taxes, needed to buy enough
thrifty foodstuffs to feed a family
(adjusted for family size) . The
poverty thresholds are updated every
year to reflect changes in the Consumer
Price Index (CPI) .-3^/
15/ The definition of poverty originated from the
United States Department of Agriculture's (USDA) 1955
Survey of Food Consumption and its 1961 Economy Food
Plan. From these surveys a poverty index was developed
by the Social Security Administration in 1964. (The
index was subsequently revised in 1969 and 1980). The
poverty index serves as the basis for determining poverty
thresholds for families by size of household and is

Table 1 shows poverty levels for 1987 and the projected
thresholds for 1988.
1987 Poverty Standard of Living Gross Incomes
Size 1987 Threshold^
1 $ 5,778.00
2 $ 7,397.00
3 $ 9,056.00
4 $11,611.00
5 $13,737.00
6 $15,509.00
7 $16,649.00
8 $19,515.00
9 $23,105.00
1988 Projected Threshold2
$ 6,017.00
$ 7,703.00
$ 9,431.00
1 Money Income and Poverty Status in the
United States: 1987. U.S. Department of Commerce,
Bureau of the Census, pp. 41-42.
2 Preliminary Estimate of Poverty Thresholds
in 1988. U.S. Department of Commerce, Bureau of the
Census, January 19, 1989.
Although this is the official definition of
poverty used in the United States, it is restrictive
because (1) the definition fails to distinguish between
adjusted yearly for inflation).
These thresholds differ from the Poverty Income
Guidelines published in the Federal Register. The
guidelines are based upon these thresholds but are
adjusted to determine income eligibility requirements for
publicly funded social programs such as food stamps and

short-run and long-run poverty; (2) it defines poverty
only in monetary terms; and (3) the definition does not
consider the variations in cost of living needs which
arise within different geographic areas and different
family compositions.
Poverty is a chronic and persistent problem in
the United States and the traditional cure is to
provide the poor with an adequate amount of money to
meet the socially acceptable minimum survival
necessities. However, this curative measure fails to
realistically focus on the problem of poverty, let
alone resolve it. Because the government has so
rigidly defined poverty, it does not distinguish
between types of poverty, i.e., short-run and long-run
poverty. There is a definite difference between being
"broke" (a short-run lack of money) and being poor
(long-run spiritual bankruptcy). This distinction needs
to be clarified in order to better understand the
nature and scope of poverty. A person who is
temporarily short of cash (perhaps because of a job
loss) can have his/her poverty status alleviated by
public assistance. However, mere cash does not ease
the poverty burden for someone who is facing a lifetime
of being "broke," not only in terms of money; but, also
in terms of attitude and circumstance. Given this

situation, the definition of poverty must also be
redefined over time.
The second restriction of the official
definition is its failure to characterize poverty in
non-monetary terms. This definition does not show that
poverty is also a lack of education, skill, and job
training. Equally important to these tangible charac-
teristics is the definition's failure to include those
intangible characteristics that also characterize
poverty such as the lack of self-esteem, the feeling of
worthlessness, the inability to recognize and take
advantage of opportunity, and the general belief that
alternatives to poverty are nonexistent.
Finally, the third limitation of the official
definition is the failure to take into consideration
the variations in regional costs or the different cost
of living needs arising from a variety of family
compositions. For example, consider the following
three person families: Family A consists of one parent
and two children each under the age of five years;
Family B consists of one parent and male teenagers aged
14 and 16. Given the differences of these two family
compositions, it is unrealistic to consider that both
families would have the same needs.

To further illustrate this point, it is
generally accepted that teenage children require more
food intake than children under five years.As a
result, Family B spends more money on food items than
Family A. At the same time, teenage children usually
do not require child care; but preschool children do
require child care if the mother is employed. There-
fore, given Family A's situation, child care costs play
an important role in its particular cost of living
Although the government's definition links
poverty exclusively to money income and it is
definitely arguable whether the root of all poverty is
merely a lack of money, the fact remains that an
insufficient amount of money to purchase minimum life
necessities is a significant determinant of poverty.
Therefore, only income-related poverty as defined in
16/ See Carolyn S. Edwards, USDA Estimates of the
Cost of Raising a Child: A guide to Their Use and Inter-
pretation. U.S. Department of Agriculture, Agricultural
Research Service, Miscellaneous Publication Number 1411,
In this example, a 14 year old teenage male
consumes an average of $3,300 per year in food and the 16
year old consumes $3,518 per year. Children under 5
years consume between $2,723 and $2,822 per year
depending upon the actual age of the child. Given this
situation, Family A spends an average of $5,545 per year
on food for its children while Family B spends $6,818
per year to feed its children.

the government's official definition is considered
Another reason to focus on income-related
poverty is the advent of the welfare reform movement.
Under the new welfare legislation, low-income women who
receive public assistance will be required to find
self-supporting occupations. As a result, it becomes
imperative that any employment satisfies the following
conditions: (1) the income from the job meets the cost
of living needs plus tax liabilities for the family;
(2) the prerequisite education and skill level for the
job and the employee are compatible; (3) the job
provides immediate, or at the very least some future
opportunity for pay increases so that the family can
become and remain income self-sufficient; and (4) the
job offers some possibilities for advancement and other
characteristics of a good job which will induce a woman
to remain on the job.
Self Sufficiency Defined
A definition of self-sufficiency includes many
concepts that are not necessarily relevant for this
project. So, we will limit the parameters of self-
sufficiency and take into consideration only those
concepts which are minimally necessary to meet or
exceed the official poverty definition. In other

words, a person is self-sufficient if he/she is
employed, has an income level above the poverty level
thresholds established by the United States Department
of Commerce, and is able to generate enough income to
meet and exceed his/her cost of living requirements by
at least 3 percent.iZ/
For these reasons, a no-nonsense approach in
the determination and alleviation of poverty is needed.
Consequently, any method used must focus on the cost of
living requirements in the specific geographic areas
for particular family compositions.
Segmented Labor Market Theory
The theoretical foundation for the development
of the job placement evaluation system is found within
the theory of segmented labor markets. Objections to
orthodox labor theory arose through (1) dissatisfaction
of the neoclassical explanation of persistent poverty
during good economic conditions; (2) failure of the
17/ The 3% figure is used for two reasons. First,
nothing in life is constant and problems arise on a
regular basis. In order for a family to overcome short-
run financial problems, it must have extra income to take
care of the situation. Otherwise, it will not be able to
meet its cost of living requirements and its temporary
crisis. Second, the average savings rate in the United
States is 3% of net income. By setting self-sufficiency
at this level, the family can also have extra money on
occasion to provide for "luxuries" or to be saved for
future consumption.

theory to satisfactorily resolve the question of
discrimination; (3) its disappointing explanation of
the existence and persistance of social class
structures; and (4) its deficient explanation of
unemployment. -3^/, -12/
The term "segmented labor market theory"
includes a variety of nonorthodox economic theories
which have been proposed to resolve these objections.
They are the dual labor market theory, the tripartite
labor theory, the stratified labor theory, the hier-
archial labor theory, the job competition theory, and
radical labor theory.^/ Each theory makes similar
objections and criticisms of the Neoclassical labor
model including (1) that the labor market is segmented
and ruled by forces other than competition; (2) that
investment in human capital i.e., education, does not
guarantee access to the primary labor market; and,
(3) that there is a shortage of primary jobs and an
18/ Glen G. Cain, "The Challenge of Segmented Labor
Market Theories to Orthodox Theory: A Survey," Journal of
Economic Literature. Volume 14, No. 4, December 1976,
pp. 1215-1121.
19/ Kathryn H. Anderson, John S. Butler, and Frank
A. Sloan, "Labor Market Segmentation: A Cluster Analysis
of Job Groupings and Barriers to Entry," Southern
Economic Journal. Volume 53, No. 3, January 1987, pp.
20/ Cain, Footnote, pp. 1215.

abundance of secondary j obs in the labor market. There
are differences among the theories such as: (1) the
number of segments which actually exist in the labor
market; (2) the reasons for the lack of occupational
mobility; and (3) the solution to the problems of labor
market segmentation.
Without denying the relevance of any particular
segmented labor market theory, the assumptions and
characteristics proposed by the dual labor hypothesis
is offered as the most complete theoretical foundation
for analysis of the labor market.^!/ Also, in keeping
with general usage, the term "segmented labor market
theory" is used throughout this paper. Basically, the
theory assumes: (1) that there are imperfections in the
labor market which prevent occupational mobility;
(2) that large companies operate in a labor market
environment composed of internal and external labor
21/ In general, this is the original theory
developed by Clark Kerr in "Labor Markets: Their
Character and Consequences," American Economic Review.
Vol. 40, No. 2, May 1950, pp. 278-291. He further
developed this theory in "The Balkinization of Labor
Markets," Labor Mobility and Economic Opportunity, edited
by E.W. Bakke, New York, Wiley; Cambridge, Mass:
Technology Press of M.I.T. 1954, pp. 92-110.
The theory was subsequently modified by Peter B.
Doeringer and Michael J. Piore, in Internal Labor Markets
and Manpower Analysis. Lexington: Lexington Books, 1971.

pools; and (3) that industrial dualism^/ is a charac-
teristic of the United States economy which promotes
segmentation of industries as well as occupations.
The first assumption of the segmented labor
theory states that the labor market is not a single
competitive market (as defined by traditional labor
theory). At least two labor markets exist within the
American economy and at any given time, a company has
access to two or more labor pools from which to choose
its employees.
This assumption of noncompetitive labor markets
is meant to explain some of the aberrations of the
labor market which traditional theory had inadequately
explained. For example, it offers an explanation for
the cause of discrimination and the development of
class structure because of the lack of occupational
mobility. Traditional labor theory, based on human
capital theory, argues that differences in wages result
22/ An economy in which there are two or more
industrial bases, each with distinct characteristics,
which influences economic organization and behavior in
different ways. John Kenneth Galbraith named two such
systems: as the Industrial System composed of "a few
hundred technically dynamic, massively capitalized and
highly organized corporations" and the industrial system
consisting of "the thousands of small and traditional
proprietors." John Kenneth Galbraith in The New
Industrial State. New American Library, New York, 1967,
at pp. 22-23.

from individual differences among people.^/ In
particular, these differences are centered around
education, work experience and individual competitive
drive. In contrast, segmented labor market theorists
postulate that differences in income distribution occur
as a result of differences among occupations;^/ there-
fore, any assumption of labor market competition can be
Also based on this assumption is the notion
that at least two labor markets exist within the United
States economy, i.e., a primary and a secondary labor
market.^5/ Table 2 differentiates the major charac-
teristics which each labor market exhibits:
23/ According to human capital theory, employers pay
wages based on an employee1s marginal productivity of
labor. The theory also assumes that the more education
or skill an employee has, the greater the productivity
and therefore the higher the wages. Daniel S. Hammermesh
and Albert Rees, The Economics of Work and Pav. Harper &
ROW, New York, 1984, pp. 89-98,115-116.
24/ Cain, p. 1222.
25/ The primary/secondary labor market scenario was
proposed in the original theory. However, as discussed
later in this paper, more than two labor markets may

Dual Labor Market Characteristics
Job Characteristics
Primary Labor Market: High wages
good working conditions
advancement opportunity
stable work
Secondary Labor Market: Low pay
bad working conditions
unstable employment
few advancement
It is further argued that occupational mobility
is nonexistent between these two segments. In other
words, a person employed at an occupation classified as
part of the secondary labor market would have very
little opportunity to find employment in the primary
market regardless of education, skill and experience.
Two possible explanations for this situation are that
discrimination prohibits certain segments of the labor
supply from obtaining or advancing into the primary
market and that a shortage of primary occupations has
prohibited many people from entering this segment.^/
The original segmented labor theorists also charac-
terized the primary sector as that part of the labor
market made up of large firms and/or unionized jobs
26/ William T. Dickens and Kevin Lang, "A Test of
Dual Labor Market Theory," American Economic Review. Vol
75, No. 4, September, 1985, pp. 792-793.

while the secondary sector included small companies
and/or nonunionized firms.22/
A second assumption of the segmented labor
market theory is the existence of an external and an
internal labor market in large companies. The external
labor market refers to the labor market which exists
outside a company. It includes all individuals seeking
occupations with anv company. On the other hand, the
internal labor market refers to the labor market which
is inside large companies. It consists of employees of
one particular company who are looking for a promotion,
more responsibility, new job duties, etc., and those
jobs which provide these opportunities within the
company. Although the primary/secondary charac-
teristics are pertinent mainly to the external
market, primary and secondary positions are found
within internal markets as well.
Based on the duality of a company's labor
source, at any time, the company has at least two
sources for its labor supply. In other words, the
company can find labor internally or externally within
either the primary or secondary segments depending upon
its needs. For example, if a supervisory job becomes
available within a company, the personnel department
27/ Doeringer and Piore, pp. 15-16.

has the option of filling the position from its
internal labor pool through the promotion of one of its
own employees or to fill the position from the external
labor pool by hiring someone from outside the company.
Depending upon the type of supervisory job open, i.e.,
accounting supervisor or janitorial supervisor, the
personnel department can go to the external primary or
secondary market or it can go to its own internal
primary or secondary market to fill the position.
The third assumption of the segmented labor
market theory relies on the idea of "industrial
dualism." This concept is based upon the belief that
technological progression as well as the concentration
of market power within certain industries are unevenly
distributed. As a result, two distinct industrial
groups have evolved in the United States.^/ One set
of terms used to describe these groups is core
industries and peripheral industries. The core
category is characterized by industries which are
powerful, concentrated, unionized, capital intensive,
and technologically progressive. Periphery industries,
on the other hand, are void of these
characteristics. ^9/
28/ Gerry Oster, "A Factor Analytic Test of the
Theory of the Dual Economy," The Review of Economics and
Statistics. Vol. 61, February 1979, p. 33.
29/ Ibid.

Several attempts have been made to empirically
test the existence of segmented labor markets.^/ In
30/ The empirical tests that have been done to test
the segmented labor market theories include the follow-
Kathryn H. Anderson, John S. Butler, Frank A.
Sloan,"Labor Market Segmentation: A Cluster Analysis of
Job Groupings and Barriers to Entry," Southern Economic
Journal, Vol. 53, No. 3, January 1987.
William T. Dickens and Kevin Lang, "A Test of
Dual Labor Market Theory," American Economic Review.
Vol. 75, No. 4, September 1985.
Gerry Oster, "A Factor Analytic Test of the
Theory of the Economy," The Review of Economics and
Statistics. Vol. 61, No. 1, February 1979.
Michael Reich, "Segmented Labor: Time Series
Hypothesis and Evidence," Cambridge Journal of
Economics. 1984.
P. Osterman, "An Empirical Study of Labor
Market Segmentation," Industrial and Labor Relations
Review. Vol. 28, July 1975.
P. Ryan, "The Empirical Analysis of Labor
Market Segmentation," The Dynamics of Labor Market
Segmentation. London Academic Press, 1981.
Russell W. Rumburger and Martin Carnoy,
"Segmentation in the US Labor Market: Its Effects on
the Mobility and Earnings of Whites and Blacks,
Cambridge Journal of Economics. Vol. 4, 1980.
David M. Gordon, "Class, Productivity and the
Ghetto," unpublished doctoral dissertation, Harvard
University, 1971.
Robert Buchele, "Jobs and Workers: A Labor
Market Segmentation Perspective on the Work Experience
of Young Men," unpublished doctoral dissertation,
Harvard University 1976.

general, they have concluded that the theory has some
validity. However, the methodology of each test has
been severely criticized by many economists. Some of
the empirical studies have included factor analysis,
cluster analysis and simple wage comparisons of
different occupations. However, no one test has been
developed which simultaneously tests all character-
istics of the segmented labor theory. Consequently,
many of the empirical tests have examined the labor
market based only on one or two characteristics or
Samuel Rosenberg, "An Empirical Test of the
Dual Labor Market Hypothesis," unpublished manuscript,
Williams College, 1976.
Eric 0. Wright, Class Structure and Income
Determination. New York: Academic Press, 1979.
Lynne Zucker and Carol Rosenstein, "Taxonomies
of Institutional Structure: Dual Economy Reconsidered,"
American Sociological Review. December 1981, Vol. 46,
pp. 869-83.
Robert Bibb and William H. Form, "The Effects
of Industrial Occupational Sex Stratification on Wages
in Blue-Collar Markets," Social Forces. June 1977, Vol.
55, pp. 974-96.
Bradley Shiller, "Relative Earnings Mobility in
the United States," American Economic Review December
1977, Vol. 67, pp. 926-41.
Barry Bluestone, "The Tripartite Economy: Labor
Markets and The Working Poor," Poverty and Human
Resources Abstracts. July/August 1970, Vol. 5, No. 4,
pp. 15-35.

The major Neoclassical criticism of each method
and the studies that have been conducted on segmented
labor markets is the high degree of subjectivity which
researchers have made in determining occupational and
industrial classifications. As a result of this
criticism, concrete proof of the existence of the
segmented labor theory has yet to be firmly es-
Even though empirical acceptance of the
segmented labor market theory is far from complete, it
is useful for purposes of this paper. And, at the very
least, it is believed that the possibility exists to
identify low paying, low skill jobs with essentially no
upward mobility where the job experience does not
qualify a person to move into a job in the primary
labor market.^/ Therefore, the assumption is made
that the external labor market does exhibit
primary/secondary characteristics. Assuming this
31/ Although segmented labor market theory covers
both external and internal markets, the focus of this
paper is only on the external, primary and secondary
sectors. The job placement evaluation system was
designed to help those people that are poor and
unemployed. Since the internal labor market affects only
those that are employed, this part of the theory is not
relevant for this project.

position, it became necessary to find a workable
procedure for classifying occupations according to at
least one segmented labor market characteristic.
After an extensive review of current literature
and empirical the tests available, the author
determined that an unpublished paper by David M. Gordon
which outlines a procedure for classifying occupations
would best serve the purposes of this thesis.-^/ In
his paper, Gordon outlined and developed a three-step
method for classifying occupations and industries.
These steps included the identification of jobs in a
three step process which identifies industry
categories, occupation categories and a classification
of jobs by combining both the industry and occupation
categories by degree of skill. (Chapter IV describes
Gordon's methodology in more detail.)
Given that an acceptable method for classifying
occupations was found, the next step in the development
of the job placement evaluation system was to determine
32/ David M. Gordon, Procedure for Allocating Jobs
into Labor Segments, unpublished paper, Department of
Economics, New School for Social Research, May 1986,
pp. 1-2.
Gordon is in the process of writing a paper which
justifies this allocation procedure. Consequently the
details of the rationalization for the development of
this classification system are known only to Gordon at
this time.

cost of living estimates for a variety of family
compositions based on a variety of family needs.
Chapter III outlines the cost of living factors and the
estimates used for the computer program.

In order to construct family profiles
describing family income needs, it is necessary to
estimate the cost of living for a family and the wage a
mother would have to earn to achieve the minimally
adequate living standard. This chapter describes the
categories which constitute a minimally adequate living
standard and the formulas which determine their
Cost of Living Estimates
The general procedure for determining the
categories which make up a minimally adequate living
standard came from three sources: studies conducted by
the federal government, two private studies, and the
author's inclusion of further categories to make the
cost of living estimates more realistic for the
Jefferson SDA. All cost estimates are based on 1988
dollars and reflect costs specific to either the
western region of the United States or to the Denver,
Boulder and Jefferson County urban areas. Tax rates
and credits are based on the 1988 tax schedules.
The initial source of information was the
United States Department of Agriculture (USDA). The

USDA has established that food, clothing, housing, and
transportation are major factors which determine living
standards.^3/ The USDA has also developed cost
estimates under these four categories for different age
and sex groups, by regions within the United States,
and whether the cost estimates are found within an
urban or rural area. Only cost estimates for food and
clothing within an urban area in the western region,
developed by the USDA are used in this thesis. Housing
and transportation cost estimates are specific to the
Denver metropolitan area and were found from other
The second source of information came from two
studies conducted by Suzanne W. Helburn and John R.
Morris.-=^/,^J>/ These studies incorporated other
categories which the USDA did not include such as
household furnishings, household furniture, household
supplies, laundry needs, child care costs, basic phone
33/ USDA Cost of raising a child
34/ Suzanne W. Helburn, John R. Morris and Daurie
Augostine, Denver Cost of Living Study 1986-1987. Working
Paper Series A-70405, Department of Economics, University
of Colorado at Denver, April 5, 1987.
35/ John R. Morris and Suzanne W. Helburn, Boulder
County 1988 Cost of Living Study for Low Income Families
Headed by Single Mothers. Working Paper Series A-81110,
Department of Economics, University of Colorado at
Denver, November 10, 1988.

service, state and federal income taxes and local sales
taxes. The Helburn and Morris studies reflected cost
estimates based on a normative market basket approach
as well as a survey of actual expenditures of 188
families who receive assistance from the Boulder Social
Service Department. Thus, their cost estimates are
current as of 1987-88 and are specific to Denver and
Boulder County, Colorado. The actual cost estimates
used from these studies were for household furnishings,
household furniture and household supplies. All other
cost categories were found from other sources and'
reflect costs specific to Jefferson County, Colorado.
Finally, after combining the USDA and the
Helburn and Morris categories, the author included
categories for an amortization of rental deposits, an
amortization of basic telephone connection costs, state
and federal income tax rate formulas, social security
tax formula, the child care tax credit formula, and the
earned income tax credit formula. The rental deposit
amortization and the basic telephone connection cost
used are specific to Jefferson County.
After all cost estimates had been determined,
the author compared them with the Helburn and Morris
studies to determine if they were comparable. This
comparison indicates cost estimates were comparable in

ail three counties from +2% to +7% for each category.
The major differences between the estimates occurred
within the categories of food, housing and child care.
Minor differences in costs occurred in the sales tax
and clothing categories. The following sections
describe each of the cost of living categories and the
sources of actual dollar estimates in greater detail.
Food estimate. Most of the information on food
costs were found in research conducted by the United
States Department of Agriculture.^/ The USDA
established costs for four food plans: (1) a thrifty
plan, (2) a low-cost plan, (3) a moderate cost plan,
and (4) a liberal food plan.-3^/ For this study,
36/ United States Department of Agriculture, Human
Nutrition Information Service, Cost of Food at Home
Estimated for Food Plans at Four Cost Levels. November
1988, U.S. Average, issued December 1988.
37/ For all food plans, the USDA assumes that all
food is purchased and prepared at home. The thrifty food
plan estimates were computed from quantities of food
published in Family Economics Review (1984 No. 1). All
other estimates were computed from quantities of foods
published in Family Economics Review (1983 No.2). The
costs of the foods plans were estimated by updating
prices paid by households surveyed in 1977-78 in USDA's
Nationwide Food Consumption Survey. Updated price
information came from a one page release published by the
Bureau of Labor Statistics: "CPI Detailed Report",
table 4, November, 1988. These costs reflect average
food expenditures for a four-person family living within
an urban area in the western region of the United States.
Other estimates for different family sizes are adjust-
ments based on the four-person family expenditure

however, only information from the monthly thrifty food
plan was used.
The following table from the USDA gives the
estimated figures used in the food calculation:
Monthly Estimated Food Costs
1-2 years
3-5 years
6-8 years
9-11 years
12-14 years
15-19 years
12-19 years
20-50 years
Monthly Cost
Although this table provided most of the
information needed for establishing food costs, it did
not include information on food costs for a child under
the age of one year. Information for a child under one
year was found in another USDA study.This study
provided that a child under the age of one year
consumed $461 worth of food per year at the economy
cost level. It also provided that a child aged one,
38/ United States Department of Agriculture, Annual
Cost of Raising a Child From Birth to Age 18 Economy Cost
Level. June 1988 Average Price Levels, Western Region,
Urban. From this table, only information on the category
"Under 1" and "1" were used for the Food at Home.

consumed food costing $576 per year. In order to
determine the estimated cost per month of feeding a
child under 1 year, the following formula was used:
Based on this formula, the estimated cost to feed a
child under 1 year was $43.21 per month.
As previously stated, these estimates were for
individuals. In order to establish family food costs,
further adjustments were necessary to take into con-
sideration economies of scale. The USDA also provided
estimates of these adjustments:
Familv Size Cost Adjustments
1 person +20%
2 persons + 10%
3 persons + 5%
4 persons 0%
5-6 persons - 5%
7 + persons -10%
From Tables 1 and 2, food costs for a variety of family
compositions were developed.^/
39/ Note that these food cost estimates do not
include males over the age of 20 years, nor are adult
males included in the adjustments for family size. The
primary reason is that a large majority of poverty
households are headed by females.

Clothing estimate. Estimates for clothing were
also based on a thrifty/economy plan.^/ The following
table breaks down the cost by age differences:
Family Clothing Estimates
Monthly Cost
Mother (16-50) $ 36.64
Child (0-12 months) 6.42
Child (1-2) 8.00
Child (3-5) 9.58
Child (6-11) 14.42
Male child (12-19) 22.42
Female child (12-19) 22.42
Laundry estimate. In calculating the laundry
estimate, the following assumptions were made:
40/ Clothing estimates for children came from the
USDA, (See Endnote no. 3). These estimates reflect both
new purchases and replacement purchases. An interpreta-
tion of the USDA study can be found in USDA Estimates of
the Cost of Raising a Child: A Guide to Their Use and
Interpretation. United States Department of Agriculture,
Agricultural Research Service, Miscellaneous Publication
Number 1411, Issued October 1981.
Clothing estimates for Mothers came from Boulder
Countv 1988 Cost of Living Study for Low Income Families
Headed bv Single Mothers. John R. Morris and Suzanne W,
Helburn, Working Paper Series A-81110, November 10, 1988,
Department of Economics, University of Colorado at
Denver. The clothing estimates are for a working mother
and include the purchase of both new and used clothing.
See Table C2 in Boulder Study Appendix.

1. Families did not have washing
facilities within the home
and had to go outside to a
coin-operated laundromat
(either commercial or
2. Each member of the family had
one load of wash per week;
3. One load per family was
allocated for general
household wash such as bed
sheets, blankets, towels,
etc. ;
4. The family wash would be done
every week (on a monthly
basis this averages 4.33
times per month)
Based on a survey of laundromat prices, the
cost to wash and dry one load was $1.50 in Jefferson
County, Colorado.-^/ From this price and these
assumptions, the following equation was developed to
determine total monthly laundry costs;
((Family size + 1) x $1.50) x 4.33
The laundry estimate does not include dry
cleaning costs. It represents only the costs incurred
by a family for ordinary household laundry.
41/ These assumptions were based on Helburn and
Morris' cost of living studies for Denver and Boulder
42/ The survey was based on a random sample of 23
commercial laundromats and 8 apartment complexes of 25 or
more units with laundry facilities located throughout
Jefferson County.

Household furniture, furnishings and Supplies
estimates. Cost estimates for household furniture,
furnishings and supplies were determined by the Boulder
County Cost of Living Study conducted in 1988.4-2/ The
calculation for household furnishings is based upon the
following assumptions: (1) base expenditure for house-
hold furnishings is $21.63 for a family of two; (2) an
additional increase in the family size to three results
in a base increase of the furnishings expenditure to
$23.26 which is a 7.5% increase; and (3) a family
composed of more than three persons the result is a
base increase of 11% (for the family of three) plus a
3% increase for each additional family member. In
other words, the increase in household expenditures
(after three persons) increases at a constant rate.
The household supplies estimate reflects a base
increase of 14% for a three-person family. For each
additional family member the expenditure increase by
1%. In this estimate, the supplies expenditure also
increases at a constant rate.
The furniture expenditure increases, becomes
constant and then decreases as the family size
43/ Tables B6 and B13 of the Boulder study. Because
this study did not extend beyond the four person per
family limit, it was necessary to derive an estimate for
larger family sizes.

increases. The following table shows these expenditure
variations by family size:
Percentacre Increases bv Familv Size For
Household Furnishinas. SuDDlies and Furniture
Percentage Increase
nilv Size Furnishinas SuDDlies Furniture
2-3 7.5% 14% 23%
3-4 11% 15% 22%
4-5 14% 16% 22%
5-6 17% 17% 18%
6-7 20% 18% 15%
Based on these percentage increases, dollar
figures for household furnishings, supplies and
furniture were determined. Table 5 shows how these
dollar amounts change and the difference each
additional family costs under these categories.
Monthly Cost Estimates
for Household Furnishings,
Furniture and Supplies
Family Size Furnish- inas Increase Sup- i olies Furn- Increase iture Increase
2 $21.63 $9.53 $12.25
3 23.26 $1.63 10.84 $1.31 15.08 $2.83
4 25.92 2.66 12.49 1.65 18.42 3.34
5 28.95 3.03 14.41 1.92 22.46 4.04
6 32.05 3.55 16.66 2.25 26.47 4.01
7 36.76 4.26 19.31 2.65 30.45 3.98

The household furnishings and supplies expenditures
includes those goods other than food and clothing
(i.e., towels, cleaning supplies, writing paper,
sheets, etc.) that are used up or need to be replaced
or repaired.
Sales tax estimate. The sales tax rate used
was 2.5% for food and 6.2% for clothing and household
Transportation estimates. The transportation
expenditure included costs for insurance, depreciation,
interest, licensing, gasoline, oil, tires and main-
tenance of a 5-9 year old car that has been purchased
44/ Sales tax rates vary in Jefferson County. Since
the major urban areas within Jefferson County have a
general tax rate of 6.5% and a food tax rate of 2.5%,
these figures were used. The following table shows this
Citv Food Tax General Tax
Arvada 3.5% 7.2%
Lakewood 2.5% 6.2%
Wheatridge 2.5% 6.2%
Golden 2.5% 6.2%
Since Lakewood, Wheatridge and Golden cover the
largest area and population, the 2.5% food tax and the
6.2% general tax rates were used. The tax rate for food
consumption in a restaurant was not used since it is
assumed that poverty families cannot afford this

for cash.4^/ This cost is constant for all family
Child care estimate. The child care estimate
is based on dollar average amounts charged per week per
child.The following table shows these dollar
amounts by age and facility:
Averaqe Weeklv Child Care Costs
Age Home Center Both
0-2 67
6wks-l year 88
lyr-2 1/2 yrs 75
Infants/Toddlers 67 82 75
2+yrs 62 66 64
Full-Time/All Ages 65 76 71
Before & After School 34 41 38
Only the average of the "Full-Time/All Ages" and
"Before & After School" categories were used for this
child care formula. Since a working mother would need
full time child care for all children under the ages of
6, $71.00 was used as the weekly rate. A child between
45/ Transportation estimates were taken from the
Denver Cost of Living Study 1986-1987. Suzanne W.
Helburn, John R. Morris and Daurie Augostine, Working
Paper Series A-70405, Department of Economics, University
of Colorado at Denver, April 5, 1987. See Table B5.
46/ Mile High United Way Information and Referral
Center Child Care Rate Report. Survey dated 7/88. This
survey is based on average cost of child care of all
licensed child care facilities in Jefferson County.

the age of 6 and 11 would need part-time child care
which averaged $38.00 per week.Also reflected in
the formula for the Before & after school category is a
full/time child care rate for those times when the
child is not in school.48/ The formula used for the
child care estimate is as follows:
Total children between 0-5 x ((71x52)/12) + Total
children between 6-11 x ((38x35.71)+(71x16.29)/12)
Housing estimate. Estimates for housing costs
include: rent, utilities, basic phone service and an
amortization of rental deposits.
The rental rates were found in the Denver Area
Apartment Vacancy Survey, Second Quarter, 1988 by
Gordon E. Von Stroh, Ph.D, Graduate School of Business
47/ There is a considerable difference of opinion as
to what age a child no longer needs child care. However,
in keeping with the Mile High survey, they used 11 as the
cut-off age.
48/ Full time child care for a school age child was
estimated to be four months per year:
Labor Day
Teacher's Workshop
President's Day
Spring Break
Memorial Day
Teacher Conferences
1/2 day/semester
3 months
1 day
1 day
2 weeks
1 day
1 day
1 week
1 day
1 day
This information came from the Jefferson County
Public School Administration office.

and Public Management. This survey reflects average
rental rates for various apartment sizes.These
figures were also cross checked against a random sample
of apartment rental advertisements found in the Rocky
Mountain News and the Denver Post for the major cities
in Jefferson County.
Utility rates were found from the Jefferson
County Housing Authority's average allocation of
utilities for 1, 2 and 3 bedroom multifamily dwellings.
Utility rates include gas and electricity not paid by
the landlord.
The telephone rate of $8.88 was based on
information provided by USWest and reflect the basic
phone service rate for two-party measured service for
the Jefferson County area. This service includes 50
"free" calls per month on a two-party line and the long
distance access charge. After 50 calls per month the
charge is $0.12 per local call. The telephone rate
estimate also includes an amortization of the initial
49/ The Denver Area Apartment Vacancy Survey surveys
Apartment Association of Metro Denver members and
nonmembers through a random sample process. The Survey
reflects approximately 200,000 rental units in the
Denver Metro area and has a response rate of one-third or
approximately 70,300 units. The nonmember sample comes
from public records, Yellow page advertising, etc. The
author of this survey has tried to be consistent in using
the same people for all of his surveys. The Jefferson
county response rate was 12,582 units.

phone hookup of $53.00. This amount is amortized over
a two year period at 18% per annum interest rate in
order to get the monthly cost of the phone hookup. The
telephone estimate does not reflect the cost of
telephone equipment rental or the purchase of telephone
equipment. It also does not reflect any charges for
long distance phone calls.
The rental deposits are based at the rate of
$100.00 per bedroom, and are amortized over a two year
period at 18% per annum.
In order to determine the number of bedrooms
for any given family composition, the following
guidelines were utilized:
Housing Guidelines
No. No.
Mother Children Bedrooms
1 1 (under 6 years) 1
1 1 (over 6 years) 2
1 2 (same sex) 2
1 2 (different sex) (either over 9 years of age) 3
1 2 (different sex) (under 9 years of age) 2
1 3 (one child under 6) 2
1 3 (all children over 6) (or 2 children of different sex over 9 years of age) 3
All other family compositions 3

This bedroom calculation reflects a maximum of
three children per bedroom. Although this standard is
not as strict as the Housing and Urban Development
(HUD) standards, it is more realistic for two reasons:
(1) multifamily dwellings generally do not exceed three
bedrooms per unit and (2) at poverty, people tend to
"make do" with whatever space is available.
Total Cost of Living. Based on the total
monthly cost for each category, a total monthly cost of
living estimate can be determined by summing the
figures of all cost of living categories.
Tax liability estimate. The total cost of
living estimate is a useful indicator of the amount
money a family needs buy those goods and services which
it requires to maintain a poverty level standard of
living. However, if an AFDC mother is required to
enter the work force, she must make a large enough
income not only to take care of her family cost of
living requirements but also her tax liabilities.
From this perspective federal income tax, state income
tax, and social security tax must also be included in
the family income requirements.
In order to determine the total tax
liabilities, the following formulas were used:

1. Total annual cost of living =
Total monthly cost of living
x 12
2. Taxable cost of living =
Total annual cost of living -
(single head of household
standard deduction ($4,400) +
(total dependent deduction
(including mother) x $1,950)
3. Total state income tax =
taxable cost of living x 0.05
4. Total social security tax =
Total annual cost of living +
total state income tax +
total Federal income tax
[Circular Problem]
5. Total Annual cost of living
with tax = Social security
tax + state tax + federal
income tax (after tax credits
or before tax credits
depending on the
circumstance) + taxable cost
of living
6. Total Federal income tax
before tax credits = taxable
cost of living x 0.15 (if
taxable cost of living is
less than 23,900) or (3,585
+0.28 x (taxable cost of
living 23,900))
7. Total Federal income tax
after tax credits = Total
Federal income tax before tax
credits (Child care credit
+ earned income credit)
8. Child care credit = Child
care cost x decimal amount
from the following table:
(based on total cost of

$0-10,000 .30 20,000-22,000 .24
10,000-12,000 .29 22,000-24,000 .23
12,000-14,000 .28 24,000-26,000 .22
14,000-16,000 .27 26,000-28,000 .21
16,000-18,000 .26 28,000 .20
18,000-20,000 .25
If there is only one child, the child care
credit must be less than or equal to $720 or less than
or equal to $1,440 if there are two or more children
receiving child care. (If the child care credit is
greater than the federal tax liability, only the amount
of the tax liability can be deducted and the federal
tax liability is $0.
9. Earned Income Tax Credit =
Total cost of living with
taxes (if < $25, 2; if <
6225, (total cost of living
with taxes $25)/25 x 3.5 +
2; if < 9850, $874; if <
18,576, 874 2.5 X (total
cost of living with taxes -
9850J/25 + 2
If the total cost of living with taxes is
greater than $18,576 no earned income tax credit is
applicable. If the earned income tax credit is greater
than the federal tax liability then the earned income
tax credit minus the federal tax liability equals
negative tax liability. In this situation the IRS will
issue a check for the negative tax liability.

10. Total annual cost of living
with taxes and credits =
total annual cost of living -
tax credits
11. Hourly cost of living with
taxes and credits = total
annual cost of living with
taxes and credits / 2080
The end result of the cost of living estimates
is an hourly cost of living wage needed to maintain a
poverty standard of living. From this cost of living
information and the Occupation and Wage Survey, the job
placement evaluation system was developed.

The Occupational Classification System
The Occupational Classification System is a
list of 2,700 occupations with their corresponding
minimum and maximum wages (at entry to the job) found
within the Denver Metro Area. These jobs and wages
represent the potential pool of job opportunities for
job training clients who are seeking employment. This
chapter describes the procedure used to classify these
occupations based on a method developed by Dr. David M.
Gordon.-^/ Also included in this chapter is an
explanation of the procedure used to develop the job
placement evaluation system and three family profiles
which show how the program works.
Gordon?s Classification
As previously stated, David Gordon relied on a
three-step process for classifying occupations and
industries. He began by gathering data on education,
skill level and industries from the 1967 edition of the
Dictionary of Occupational Titles (DOT). He also
determined whether the occupation and its industry were
50/ David M. Gordon, Procedure for Allocating Jobs
into Labor Segments, unpublished paper, Department of
Economics, New School for Social Research, May 1986.

goods-producing or non-goods producing. Based on this
information, Gordon settled on three guidelines which
served as the foundation for his classification system.
These guidelines are as follows:
1. As much as possible the
segment categories should
refer to data about the
characteristics of -jobs
excluding information about
the characteristics of the
workers who hold those jobs.
2. As much as possible, . .
segment categories should
build upon data which
excludes information about
final labor market outcomes,
such as wages and turnover
3. Given the importance of
industrial characteristics in
defining differences between
"core" and peripheral" firms
and given the strong
likelihood of job segmenta-
tion within core firms, it is
important to take both
industry and occupational
characteristics into
In guideline no. 1, Gordon establishes that job
characteristics rather than Worker characteristics are
the primary focus. The basis for Gordon's reasoning is
found within the segmented labor market theory since
the theory assumes that labor market segmentation
51/ Ibid., p. 1.

arises as a result of differences among occupations.
Gordon refers to the four classifications used in his
allocative process as "segment categories." These
classifications are (1) the Independent Primary
Professional/Technical Segment (IPPTS); (2) the
Independent Primary Craft Segment (IPCS); (3) the
Subordinate Primary Segment (SPS); and (4) the
Secondary Segment (SS).
The IPPTS category consists of those
occupations which require a college degree in a
specialized occupational field. Although this is a
necessary condition, the majority of occupations in
this category also required special licenses and
advanced education, experience or skill.
The IPCS category consists of those occupations
which do not require a college degree but do require
extensive vocational training, apprenticeship training
or other lengthy informal training not usually found
within a college/university environment. Many of these
occupations also require years of on-the-job training
before a license to work unsupervised can be obtained.
The SPS category consists of those occupations
where a college degree was not necessarily required for
the occupation but prior work experience or skill level

was extremely important. This particular category was
the catch-all for those occupations that did not
strictly meet the standards established for the
Independent Primary Professional/Technical, Independent
Primary Craft or Secondary segments.
Finally, the SS category consists of those
occupations that did not require any formal education,
specialized training, skill or prior work experience.
In many cases, not even a high school diploma is
needed. Guideline no. 2 sets the parameters for
the classification scheme. It is assumed that Gordon
included this guideline in order to avoid regional wage
differences and economic factors that could effect the
results of the occupational classification. Basically,
Gordon's allocation system is a categorization of
occupa-tions by the degree of skill or education
required for a particular job and the industry to which
the job belongs.
Finally, guideline no. 3 establishes the effect
industrial dualism has on occupational segmentation.
Although Gordon does not give his justification, he
seems to have included this guideline to reflect the
fact that the lack of occupational mobility within the
labor market occurs within or between industries as
well as occupations. Another consideration may be that

since numerous jobs have the same or similar
occupational characteristics, Gordon used industry
characteristics to clarify the distinctions between
them. In other words, semi-skilled and unskilled
occupations are classified as to whether they are
within a goods-producing industry or within any other
industry. Since goods-producing industries fall within
the core classification, semi-skilled and unskilled
occupations in these industries are ranked higher than
similar jobs in periphery industries.
With DOT data and these guidelines, Gordon then
developed a two by two matrix to classify occupations
and industries. Gordon's Table 1 and explanation is
quoted below:
Table 1: Industry, Occupation and Segment
All Other
& Unskilled All other
Occupations Occupations
industrv characteristics determine segment occuoation characteristics determine segment
occuoation characteristics determine segment occuoation characteristic determine segment
I have formalized this hypothesis in
the following way: in semi-skilled and
unskilled "blue-collar" occupations in
goods-producing sectors, I assume that
location in either the core or the

periphery is sufficient to determine
location in either the subordinate
primary or the secondary segment,
respectively. In this unique industry/
occupation combination, in other words,
industry characteristics alone
determine segment location. In all
other industry/occupation combinations,
conversely, I assume that occupational
characteristics alone are sufficient to
determine segment location.-^/
The first step in Gordon's allocative process
was the determination of industry categories based upon
an indepth factor analysis developed by Gerry Oster.
This study analyzed data for 83 3-digit 1960 Census
Code industries in order to classify them as either
core or peripheral industries.-^/ This list fulfilled
one requirement for the classification matrix.-^/
The second step involved the development of the
occupation classification into IPPTS, IPCS, SPS and SS.
Gordon explained the procedure as follows:
The allocation of occupations into
segments is more complicated and less
satisfactory conceptually. I had
originally hoped that a single,
relatively neat factor analysis of
occupational characteristics would
result in a three- or four-way segment
52/ Ibid, pp. 2
53/ Oster, p. 39
54/ Ibid, p. 3. See Appendix B for a list of these

categorization. No such luck: the
patterns of distinctions and divisions
among occupations turn out to be too
complex to sustain such a direct
I have instead resorted to a
recursive method for dividing
occupations into segments. This method
results in an exhaustive allocation of
all three digit occupations into the
four segment categories.
Gordon rescaled the scaling procedures for all
skill levels used from the Dictionary of Occupational
Titles to make finer distinctions to combine education
and skill level. The following are his specific
allocation rules:
1. Occupations are included
within the IPPTS if Whole
Activity > .250 .or. the sum
of General Education
Development values 5 and 6 >
.800. and. if neither
Repetitive Instruction nor
Specific Instruction > .250.
2. Remaining occupations are
included within the IPCS if
Frequent Change > .250 .and.
sum of Specific Vocational
Preparation (SVP) values 7,8,
and 9 > .800. and. neither
Repetitive Instructions nor
Specific Instructions > .250.
3. Remaining occupations are
included within the SPS if
the sum of SVP values 4
through 9 > .800.
4. Remaining occupations are
allocated to the secondary
segment. (This is a

sufficient condition; it
effectively designates
secondary occupations as
those with SVP requirements
with values 1 through 3.)
The substantive implications of
these rules are relatively simple:
IPPTS occupations are those in which
workers are allowed control of whole
activities or with high general
education preparation requirements and
are independent of detailed or repeti-
tive instructions. IPCS occupations
are similarly independent and involve
both frequent changes of activities----
which turns out to be one of the
distinguishing characteristics of the
DOT coding for craft occupationsand a
high probability of specific training
time of at least wo [sic] years. Given
the elimination of those two
independent primary categories, the
distinction between the SPS and SS then
depends on whether or not jobs require
more than three months training time
the dividing line between SVP category
3 and 4.
The third step in Gordon's allocative process
involved a procedure to combine the industry and
occupa-tion categories based on the classification
matrix for semi- and unskilled occupations in goods
producing industries, classify jobs as SPS for core
industries and as SS for industries in the periphery.
For skilled jobs and for jobs in nongoods producing
industries, jobs are allocated according to occupation
characteristics. This classification scheme is based
on the theory of segmented labor markets that

production jobs in core industries are mostly in the
primary labor market.-^/
Gordon makes the assumption that industrial and
occupational characteristics are stable over time.
Hence his categorization of I960, 1970 and 1980
industries.-^/ However, a comparison of the 1967
edition of the DOT and the 1977 edition (as well as the
1986 Supplement) show that some industry and
occupational characteristics have changed over time.
Consequently, for purposes of this paper, Gordon's 1980
list was modified to reflect these changes.
The occupation list developed by Gordon and the
subsequent modifications to the list served as a
starting point for the development of a job placement
evaluation system. Gordon's occupational classifica-
tion list was used as the basis for classifying
occupations that were found within the Denver Metro
Area. However, since Gordon's list is a general
occupational classification, many specific job titles
were not included. For example, Gordon classified
welders within the IPCS category; however, he did not
categorize different type of welders such as certified,
55/ See attached Appendix B for a list of occupations
and the segment in which they belong.
56/ Ibid, Footnote 4, p. 3 and Footnote 5, p. 5.

noncertified, arc, gas or combination welder. Also, he
did not distinguish between levels of occupations,
i.e., civil engineer A, civil engineer B, etc.
Consequently, different types of occupations and their
different grades had to be incorporated into Gordon's
list. The following section discusses the modifica-
tions of Gordon's list and the procedures used for the
Occupation and Wage survey.
Occupation and Wage Survey
With this occupational classification system
in place, the next step was to find current maximum and
minimum wages for each job classification. The Wage
Survey developed by combining several wage surveys
found within the Denver Metro Area. Four of the major
surveys used were: (1) the Area Wage Survey-^/; (2) Job
Bank Data;-^/ (3) Colorado Department of Personnel;-^/
and (4) a private industry survey whose authors have
requested anonymity.
The raw list of occupations and their wages
totaled approximately 3,200. However, after deleting
57/ This survey is developed by the U.S. Department
of Labor.
58/ Employment data compiled by the Colorado Depart-
ment of Labor, Employment and Training Division.
59/ Surveys all state occupations.

duplicates and redundant job titles, the final list
totaled approximately 2,700 different occupations which
included maximum and minimum wages (at entry to the
job) for each occupation.^/
The Occupation and Wage Survey included only
those occupations found within the state of Colorado.
It does not include occupations such as long shoremen
or automobile factory workers which may be found
elsewhere within the United States. Included in the
survey are several occupational titles that are
specific only to the State of Colorado i.e., state
government occupa- tions or instances where only one
position may exist.However, these occupations were
kept in the survey because many were ranked according
to years of service, experience or skill level (i.e.
Revenue Agent A, B, C, Senior, Principal, etc.).^/
The Occupation and Wage Survey along with the
cost of living estimates forms the basis for the job
placement evaluation system. The following chapter
discusses the development of the computerized program
60/ See Appendix A for a complete list of all
occupation titles and the minimum and maximum wages.
61/ For example, Eisenhower Tunnel Supervisor or
Director of the Colorado Lottery.
62/ See Appendix A for a complete list of the
occupations, wages and their classifications.

for the job placement evaluation system and its
connection to segmented labor market theory.

Job Placement Evaluation System
The components of the job placement evaluation
system were developed in previous chapters. In this
chapter these components are combined to explain why
characteristics of the segmented labor market theory
were used to develop this system and how the job
placement evaluation system works. This chapter also
includes cost of living profiles for three families
generated by the computer program, the results for each
family when the mother is placed in a hypothetical job,
and the evaluation of the job training program based on
these three family situations.
The first step in the development of a job
placement evaluation system based on the segmented
labor market theory and Gordon's classification method,
was to determine those occupational characteristics
which an employee would find the most valuable and
which were quantitatively measurable. Since this
thesis deals with income-related poverty, the
assumption was made that income and those factors that
relate to income were the most significant as well as

the best for quantitative analysis.^2/ These occupa-
tional characteristics are as follows:
1. job earnings
2. wage security
3. future opportunity
4. skill/education requirement
Job earnings. The measurement for this charac-
teristic relied on the actual market wages for occupa-
tions. Specifically, the market wages included both
the minimum and maximum wages at entry to an occupa-
tion. However, for this study, job earnings were not
ranked by occupation; rather, job earnings were ranked
according to their ability to provide for a family's
cost of living requirements. The reasons for this
method include:
1. empirical studies for this
thesis do not conclusively
prove that all primary
occupations are high paying
and all secondary occupations
are low paying;
63/ This assumption is not without merit. A survey
conducted by the Piton Foundation, Breaking the
Cycle of Dependency: A survey of Colorado's
Welfare Clients. June 1988, showed that when the
respondents were asked "what kinds of things
would you worry about if you were trying to get
off AFDC?" 83% said they worried a lot about
earning enough money to support their family
(Table 6, p. 13). When asked about job concerns,
80% agreed that "most jobs that I have had, or
could get, don't pay enough" (Table 14, p. 20).

"high" or "low" pay is an
arbitrary and subjective
notion; and
3. different families have
unique cost of living needs
and therefore different
income requirements.
Based on this survey of occupations and wages,
jobs in the primary segment are not all "high paying."
Nor does it make sense to take a "low paying job" in
the primary sector when an occupation in the
subordinate or secondary market would provide higher
earnings that would meet a family's cost of living
requirements. This conclusion would be especially true
if the wages for a particular primary occupation do not
have the future potential to satisfy a family's cost of
living needs. Of course, if the primary sector job
also has more long-run potential for promotion and
therefore a prospect of higher wages, it may be
advisable to take the primary job. However, the long-
run and the short-run must be defined on an individual
basis. For example, if a single mother with a sixteen
year old child is able to obtain a low-paying job in
the primary sector, long-run potential does not satisfy
short-run need. In this case, the need is short-run
because the child will be an adult within two years
and, theoretically, on his own. The cost of living for

this family would decrease as soon as the child moved
away from the household. On the other hand, if a
single mother with an infant child found a low-paying,
primary sector job with long-run potential, it may be
advisable to take the primary job since the child will
be with the mother for eighteen years. Although the
evaluation of earnings should focus on both
occupational characteristics and income requirements in
long-run and short-run time frames, this may be an
unaffordable luxury for the family. In other words,
the immediate goal is to achieve income self-
sufficiency with the best possible job regardless of
the occupation's classification. If this goal can be
met by an occupation within the primary category, so
much the better.
Wage security. The dual labor theory states
that job stability differs among labor markets, i.e.,
occupations in the secondary market are more unstable
than those in the primary market.-^/ However, given
the restructuring of industries which the U.S. economy
has experienced, many industries in both the core and
peripheral sectors have not exhibited job stability.
64/ Job stability is defined as permanent, full-time
work where employees have little fear in losing their
occupations and/or where there is opportunity to find
comparable work at comparable wages in the job market.

For example, occupations in the oil and gas, auto,
airline, agriculture, real estate, securities, banking
and mining industries have been highly unstable in
recent years. Given the variety of industries which
have experienced occupational instability, the idea
that job stability exists within any labor market is
debatable. Consequently, job stability, as defined by
the theory, is an unsatisfactory evaluation component
of the segmented theory.
Since alleviating income-related poverty is the
purpose of developing a job placement evaluation
system, the author determined that an evaluation of
wage security would best satisfy this particular
problem.^/ Although the segmented labor market theory
does not focus on wage security, per se, wage security
is a iob characteristic and falls within its
theoretical boundaries.
Two reasons for looking at wage security from
this perspective include the historical pattern of wage
fluctuation and the different wages offered by
employers. Fist, empirical studies have shown that
wages are downwardly rigid and only under immense
65/ Wage security is defined as a constant or
increasing level of income received for a particular
occupation which meets or exceeds a given family's cost
of living requirements.

economic pressure will they fall.-^/ On the other
hand, increases in wages occur regularly because of
inflationary pressures and because employers generally
give merit raises to their employees in order to retain
them. As long as the increases in wages keep pace with
increases in cost of living, wage security can be main-
Second, within any real-world labor market,
different wages are offered by different employers for
the same type of occupation.^/ The Wage and
Occupation Survey included in this study shows minimum
and maximum wages paid for particular occupations.
These occupational wage ranges provide information
which promotes wage security because they show
potential wages which could be earned for a particular
job. This information is important because once an
66/ Daniel S. Hammermesh and Albert Rees, The
Economics of Work and Pav. Harper & Row, New York, 1984,
pp. 376-377, 379-380. Although wage rigidity has been
shown to exist, the pervasiveness of this phenomenon may
be questionable since wages have fallen in some in-
dustries and occupations over the past few years.
However, the decrease in wages may be caused by struc-
tural changes, i.e., economic pressure, either within an
industry or geographic area rather than some other
unexplained factor.
67/ This method further validates the theory that the
labor market is noncompetitive. If the labor market were
competitive, all wages for a particular occupation would
be the same because information is constant for everyone
and market forces would keep wages the same.

employee acquires skill in a particular occupation,
he/she may be able to use these same skills to acquire
a different job that pays higher wages or provides more
long-run potential. The job training program also
benefits from this information because the probability
that the client will remain employed and self-
sufficient rises as wages increase.
The wage security category was measured by
(1) averaging maximum and minimum wages for a given
occupation (based on the Occupation and Wage Survey);
(2) comparing the average wage to the hourly cost of
living wage needed for a family;^/ and (3) comparing
the actual market wage range and a job training
placement wage.
The average wage for a given occupation is the
minimum and maximum wage divided by two. This wage
shows the average that can be expected from a given
occupation. By comparing the average wage with the
cost of living wage necessary for a given family, one
part of the wage security can be determined. If the
cost of living wage is lower than the average wage for
the occupation, the family will have a type of
stability indicator which shows a probable income that
68/ See Chapter II Cost of Living Estimates for ah
explanation of the hourly cost of living wage.

can be earned. The second part of the wage security
indicator involves the comparison of the actual job
placement wage and the average wage. This comparison
gives the job training client an idea where he/she
stands with respect to the market wage for the same
type of occupation.
Given both of these comparisons, a determina-
tion of wage security is possible if a family's cost of
living needs are currently being met and if a family's
cost of living needs have the probability to be met in
the near future based on the average wage for an
employee's occupation.
Although wage security does not guarantee that
a low-income employee will stay on the job, it does
provide some incentive to stay since there is "hope"
for the future. Given that a lack of money is a
significant concern for the poor, some type of wage
security would seem to be a major motivating factor to
continue working.
Future opportunity. The dual theory looks at
future opportunity in terms of advancement potential.
However, because the lack of money is officially
defined as the single most important determinant of
poverty, future opportunity is defined, in this
thesis, in terms of wages. Specifically, future

opportunity is the potential of future earnings (based
on maximum wages from the Occupation and Wage Survey)
for a given occupation to meet a given family's cost of
living needs. In other words, the current cost of
living wage is compared to the maximum market wage paid
for the employee's chosen occupation. This comparison
shows whether the cost of living wage required can be
maintained in the future.
Skill/education requirement. Although the
segmented labor market theory emphasizes job
differences rather than human capital differences, it
cannot ignore the importance employers place on
education, work experience and skill level. Hence,
Gordon's dependence on classifying occupations is based
to some degree on human capital factors. If a person
lacks skill and education, the odds that he/she will
find employment in the primary sector are extremely
remote. However, education and skill are not
sufficient conditions to guarantee a primary labor
market job.
Overall Job Rank
The job placement evaluation system is a five-
part evaluation of each labor market characteristic
described in the segmented labor market theory. In

other words, the evaluation system includes the
individual evaluation of job earnings, wage security,
future wage opportunity and skill/education required
for a particular job; and the collective evaluation of
all four of the factors as a group.
Table 10 explains how each factor was
Factor Characteristic Evaluation
J ob Rank
istics 1 2 3 4
A. Job Earnings Earnings Earnings Earnings
earnings: greater than 103% COL needs greater 100% but less than 103% COL needs greater 75% but less than 100% COL needs less than 75% of COL needs
B. Wage Probable Probable Probable Probable
security: Earnings greater than 103% COL needs Earnings greater than 103% but less than 100% COL needs Earnings greater than 75% but less than 100% COL needs Earnings less than 75% Of COL needs
69/ Cost of Living Estimates for more detail.
"Earnings" is defined as wages paid for a
specific occupation.
"Probable Earnings" is defined as probable
average earnings based on minimum and maximum wages paid
for a specific occupation.
"Potential Earnings" is defined as the maximum
entry level wages paid for a given occupation.

TABLE 10 (continued)
Factor Characteristic Evaluation
Job Rank
istics l 2 3 4
C. Future Potential Potential Potential Potential
Potential :Earnings Earnings Earnings Earnings
greater less than greater less than
than 103% 100% but 75% but 75% of COL
COL needs less than less than needs
103% COL 100% COL
needs needs
D. Skill/ Primary Primary Subord. Secondary
Education Profess. Craft Primary
Required: (College (Vocation (some (no skill
degree +) training college or experi-
plus some plus some ence
experi- experi- required)
ence) ence)
Job Place- Sum A-D Sum A-D Sum A-D Sum A-D
ment Rank: less than less than less than less than
or equal or equal or equal or equal
to 6 to 11 to 14 to 16
The following section describes three family
profiles based on the computerized job placement
evaluation system.
Family Profiles
The following examples provide cost of living
estimates and job placement evaluation system profiles
for three different family compositions. These
profiles are based on the computerized program and do
not represent the entire range of family composition
profiles that can be generated from this program.

Mother's Acre Sex and Age of Children
Family A 35 8 year old 12 year old male
Family B 19 Under 1 year
Family C 44 5 year old 14 year old male 16 year old female
Table 11 shows the cost estimates for each
family on a monthly basis and provides an annual total
with and without taxes and credits.

Total Monthly Estimate Family A Family B Family C
Food: Mother: Child: Child: Child: $ 79.40 63.90 79.30 $ 79.00 43.21 $ 79.40 52.20 79.30 79.00
Adj ustment $210.90 X 1.05 $122.21 x 1.10 $289.90 x 1.00
Total Food $233.73 $134.43 $289.90
Clothing Mother: Child: Child: Child: Total Clothing $ 36.64 14.42 22.42 $ 73.48 $ 36.64 6.42 $ 43.06 $ 36.64 9.58 22.42 22.42 $ 91.06
Laundry: $ 25.80 $ 19.35 $ 32.25
Household Furniture, Furnishings and Supplies $ 49.18 $43.41 $ 56.73
Sales Tax $ 14.62 $ 9.39 $ 17.86
Transportation: $148.67 $148.67 $148.67
Child Care: $209.46 $307.67 $307.67
Housing Estimate: 3 bdrm $573.58 1 bdrm $424.39 3 bdrm $573.58
COL before taxes: $1,328.52 $1,130.57 $1,517.71
Total Annual COL before taxes: $15,942.20 $13,564.46 $18,212.55
Total Annual COL
with taxes and
credits: $17,926.45 $14,961.10 $20,569.71
Hourly COL wage: $8.62
(Based on Annual COL
with taxes and credits)

Table 12 shows the outcome for each family when
the mother is employed in a hypothetical occupation.
(Minimum and maximum wages used in the table are actual
wages found within the Occupation and Wage Survey for
those occupations.)
Mother Wage
Family A:
Secretary III $8.75
Family B:
Bank Teller $3.35
Family C:
General Duty
Nurse $12.00
Minimum Wage Maximum Wage Average Wage
$7.77 $12.21 $9.99
$3.35 $6.37 $4.86
$6.00 $18.00 $12.00
Table 13 shows how the results of the ranking
system for each family based on the market and
placement wages and the family's individual cost of
living needs.
Current Probable Potential Job
Familv Rank Rank Rank Rank
A 2 1 1 3
B 4 4 3 4
C 1 1 1 3

The mother in Family C is the only person that
has achieved income self-sufficiency at the present
time. Based on market wages, this mother has the best
probability to maintain her income self-sufficiency
The mother in Family A does not quite meet her
income self-sufficiency requirements. However, within
a very short time, this mother will probably meet and
maintain her income self-sufficiency needs.
The mother in Family B does not now meet her
cost of living needs and does not have the potential to
meet her family needs in the future. This mother
should be directed toward a different occupation that
pays higher wages or should enroll in some type of
vocational training program.
Finally, the computer program counts the number
of participants the job training program has helped and
determines where the percentage of participants fall
with respect to the ranking system. For our three
family example, 1 person (33.33%) became self
sufficient; 1 person (33.33%) has the probability of
becoming self-sufficient within a relatively short
period of time; and 1 person (33.33%) does not meet and
probably will never meet her COL needs from this

occupation. The following chart shows how the
computerized program evaluates the job training agency.
Probable Potential No
No. Self Self Self Self
Assisted Sufficient Sufficient Sufficient Sufficient
3 33.3% 33.3% 00.0% 33.3%
If these profiles had included more than three
families, the computerized program would have ranked
them and determined whether the job training program
met its internal self-sufficiency placement goals.

Conclusion and Policy Implications
The job placement evaluation system has the
potential to provide job training agencies with a
valuable tool for administrative decision-making. On
the one hand it can be used to guide its job training
participants toward occupations that will alleviate
their income-related poverty. On the other hand it can
provide a substantial cost-saving substitute for
internal evaluation.
By continually monitoring where its clients
fall with respect to the agency's internal goals, the
job training agency will know how well its self-
sufficiency placement goals are being met. The larger
the number of clients that are directed toward self-
sufficiency, the greater the probability that these
clients will remain self-sufficient. Although a long-
term analysis of client job tenure after job placement
would serve the same purpose and may provide more
evaluative accuracy, it is very costly to undertake.
Therefore, the next best alternative is this job
placement evaluation system.
Since funds are limited to social service
departments and job training agencies in general, one

possible policy implication that can result from the
implementation of this job placement evaluation system
is its ability to serve the greatest number of people
at the least cost.
The job placement evaluation system is also a
very flexible program. Although the system is set up
to include only single mothers, the cost of living
formulas can be expanded to include males. In
addition, other cost of living information can also be
included in the program as it is needed. Finally, the
program can also be linked with the Consumer Price
Index. This connection would alleviate the need to
conduct yearly or bi-yearly cost of living studies,
yet still keep the estimates in line with current
The occupation list is easily expandable to
include other job titles. Minimum and maximum wages,
on the other hand, will change on a regular basis.
However, assuming that wages generally keep pace with
inflation, they too can be linked to the Consumer Price
Index and thereby alleviate the need for gathering
continuous wage survey information on a yearly basis.
There is an indirect measurement of job
satisfaction inherent within the primary/secondary
classification, i.e. primary jobs generate more job

satisfaction through increased responsibility and
better working conditions than secondary jobs. This
indirect measurement could serve as a proxie for
determining job satisfaction and could be used now as a
counseling tool. However, additional research for the
job placement evaluation system could include a
quantitative measurement for job satisfaction based on
the interest level generated for a job. With the
inclusion of this factor in the job placement
evaluation system, all of the characteristics that
identify the differences between occupations (from the
segmented labor market theory) will have a procedure by
which they can be measured.
Finally, the implementation of this program
will provide the most benefit to the job training
client or AFDC mother and her children. If she is able
to find a job that provides both a spiritual future as
well a financial future, then a major victory will have
been won in the war on poverty in the United States.

Anderson, Kathryn H., John S. Butler, and Frank A.
Sloan. "Labor Market Segmentation: A Cluster
Analysis of Job Groupings and Barriers to
Entry." Southern Economic Journal. Vol. 53.
No. 3. January 1987. pp. 571-572.
Ashenfelter 0. and Card, D. "Using the Longitudinal
Structure of Earnings to Estimate the of
Training Programs." Review of Economic
Statistics. November 1985. Vol. 67. No. 4.
pp. 3-9.
Barnow, B.S. "The Impact of CETA Programs on Earnings-A
Review of the Literature." Journal of Human
Resources. Vol. 22. No. 2. pp. 157-193.
Barnow, B.S. "Evaluation Employment and Training
Programs." Evaluation and Program Planning.
Vol. 9. No. 1. pp. 63-72.
Bassi, Lauri J. "The of CETA on Post Program
Earnings of Participants." Journal of Human
Resources. 1983. Vol. 18. pp. 539-556.
Bassi, L.J. "Estimating the of Job-Training
Programs, using Longitudinal Data," Journal of
Human Resources. Vol. 22. No. 2. pp. 300-303.
Bassi, Lauri J. "Evaluating Alternative Job Creation
Strategies." Econometric Inquiry. October 1985.
Vol. 23. No. 4. pp. 671-690.
Becker E. M., Patrick M. Horan, and Charles Tolbert,
"Stratification in a Dual Economy: A Sectorial
Model of Earnings Determination." American
Sociological Review. 1978. Vol. 43.
pp. 704-739.
Berlin, Gordon and Andrew Sum. Toward A More Perfect
Union: Basic Skills. Poor Families, and Our
Economic Future. Occasional Paper No. 3. Ford
Foundation Project on Social Welfare and the
American Future. 1988.
Berlin, Sharon B. and Linda E. Jones. "Life After
Welfare: AFDC Termination Among Long-Term
Recipients." Social Service Review. September
1983. pp. 378-402.

Berman, Elain Gants, Grant Jones. Breaking the Cycle of
Dependency; The View of Colorado Social
Services Staff. The Piton Foundation. June
Bibb, Robert and William H. Form. "The s of
Industrial Occupational Sex Stratification on
Wages in Blue-Collar Markets." Social Forces.
June 1977. Vol. 55. pp. 974-96.
Bishop, J. H. and Montgomery, M. "Evidence on Firm
Participation in Employment Subsidy Programs."
Industrial Relations. Winter 1986. Vol. 25. No.
1. pp. 5664.
Bishop, John, George Farkas, Michael C. Keeley, C. Eric
Munson and Philip K. Robins. A Research Design
to Study the Labor Market s of the
Employment Opportunity Pilot Projects.
Institute for Research on Poverty. 1980.
Bjorklund, Anders and Robert Moffitt. "The Estimation
of Wage Gains and Welfare Gains in Self-
Selection Models." Review of Economics and
Statistics. Vol LXXIX. No. 1 (February 1987)
pp. 49-49.
Blau, D.M. and Robins, P.K. "Training Programs and
Wages; A General Equilibrium Analysis of the
s of Program Size." Journal of Human
Resources. Winter 1987. Vol. 22. No. 2.
pp. 113-125.
Bloom, Howard. "Estimating the of Job Training
Programs Using Longitudinal Data: Ashenfelter1s
Findings Reconsidered." Journal of Human
Resources. 1984. Vol 19. pp. 544-556.
Bluestone, Barry. "The Tripartite Economy: Labor
Markets and the Working Poor." Poverty and the
Human Resources. 1970. Vol 5. pp. 15-35.
Borus, Michael E. Measuring the Impact of Employment
Related Social Programs. W.E. Upjohn Institute
for Employment Research. 1979. Kalamazoo,