THE USE OF MARKOV PROCESSES TO EXAMINE MOBILITY PATTERNS
IN THE LABOR MARKET
by
Dan S Bernstein
B.A. University'.67f Massachusetts
( 982)
SUBMITTED IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS OF THE DEGREE OF
MASTER OF CITY PLANNING
IN URBAN STUDIES AND PLANNING
at the
MASSACHUSETTS INSTITUTE OF TECHNOLOGY
June 1986
(D
Dan S Bernstein
1986
The author hereby grants to M.I.T. permission to reproduce and
to distribute copies of this thesis document in whole or in part.
Signature of Author.
DepartmerQ-Oban Studies and Planning 5/21/86
Certified by
Martin Rein, Tlesis Supervisor
Accepted by
Gary Hack, Department Chair
ROtch
MASSACHUSETTS INSTITUTE
OF TECHNOLOGY
JUL 10 1986
I IIQDAMRF5
M
ITU biran
Document Services
Room 14-0551
77 Massachusetts Avenue
Cambridge, MA 02139
Ph: 617.253.2800
Email: docs@mit.edu
http://Iibraries.mit.edu/docs
DISCLAIMER OF QUALITY
Due to the condition of the original material, there are unavoidable
flaws in this reproduction. We have made every effort possible to
provide you with the best copy available. If you are dissatisfied with
this product and find it unusable, please contact Document Services as
soon as possible.
Thank you.
Some pages in the original document contain pictures,
graphics, or text that is illegible.
Thank You
A special thank
Harvard
knew
you
University
and
presented.
to
Professor
Roderick
J.
Harrison
for the effort he put into assuring
thoroughly
understood
the
material
about
that
of
I
to
be
This generous effort, which took many, many hours
of
his time, included training in
the
necessary
mathematics,
computer system and statistical package at Harvard, and
all
the
the
help with successive drafts of the paper.
To Efrat F. Levy.
For being a sounding board for organizing my
ideas, and for becoming an expert in this
analysis
without
type
labor market
even understanding the mathematics.
help in typing and editing, and for support
the completion of
of
my
thesis
such
as
For
her
which contributed to
bringing
home ice cream
sundaes at midnight.
To my advisor, Professor Martin Rein, for his
help in locating
Professor Harrison and for making sure I understood the
material
and completed the work on time.
To Professor Joe Ferrera, for help in connecting the model with
the other ideas in
the thesis.
TABLE OF CONTENTS
Page
INTRODUCTION
THE MODEL
THE DATA SET
RESEARCH CONTEXT
THE Q MATRIX
CONCLUSION
APPENDICES
BIBLIOGRAPHY
INTRODUCTION
The purpose
into
of
my
changing demand for labor
how
available (jobs)
this data.
national
the
affects
in the labor market, and to
of an occupation by
income
thesis is to make an initial investigation
test the usefulness
model in analyzing
input-output
occupation
opportunities
Using this type of input-output model, I will look at
labor
markets in a new way.
and low status occupations;
I will
the
analyze
specifically, non-farm
and service occupations as defined by the U.S.
Census.
low
labor
The data
set for this study is the Panel Study on Income Dynamics (PSID).
The
model
I will use is a special form
model used in economic research.
The model
vacancy chains that arise in the labor
are methodological abstractions that
predict
than
movements
of
of
the
is used to trace the
market.
are
input-output
Vacancy
easier
individuals.
A
chains
to measure and
vacancy chain
is
initiated when an incumbent retires and the position is retained,
or when
a
new position is created.
Open positions in the labor
market are referred to as vacancies.
As long as employed persons
are recruited to fill each vacancy, a chain of job vacancies will
be created, because workers, when leaving jobs, must be
replaced
when they move into new jobs.
chain of
This process
moves which continues as long as
inside the
labor force.
employees
creates
are
a
recruited
from
Recruitment of an unemployed individual
will end the chain of opportunities.
In the
case
of
a vacancy
chain originating from a retirement there can be a whole chain of
- 1 -
opportunities without a single job creation.
Adequate analysis
of
absence
has been limited by the
opportunities
structure of occupational
appropriate quantitative methods.
of
by
this direction has been done
Roderick
University (see chapter on the model for a
a
I will take
the mathematics involved).
macro level methodology and then
the changeing
regarding
issues
policy
work
Initial
Harvard
of
Harrison
fuller explanation of
brief
at
look
It is
underlying
model
type
input-output
this
processes that will determine if
these
in
stability
of
absence
or
presence
this
an exploration of some of
make
the underlying processes taking place in the labor market.
the
in
will accurately predict the occupational mobility of labor.
It is
that
possible
a vacancy which arises in one occupation
creating
can affect the other occupations by
as good as new jobs; that is,
position exists.
job
To
with
make
These openings will appear as opportunities for
those
the
job
there is no difference as long as a
who
market
actually being created.
long
as
the
are currently employed,
person enters the labor market.
will
of
changers the new openings are
those seeking the positions, but as
filled
series
Each move will appear as an
moves for labor market participants.
opportunity to the mover.
a
appear
openings
no
are
unemployed
Movement within the labor market
to have more new jobs
than
are
This process is illustrated in the model
by the vacancy chain and job multiplier.
These chains of job vacancies might
-
2 -
start
in
the
managerial
someone
job for
occupations.
I will analyze these
a service occupation.
in
in
meaning
moves and their
The result could lead to a
anywhere.
end
could
occupation but
service
the
labor
unskilled
and
who
I will analyze who stays in these occupations,
leaves, and who arrives.
If regular patterns of vacancy chains can
be
predicted,
then
this chain of opportunities will reflect the standard recruitment
patterns
of
occupation.
occupations,
Assuming
and
be
should
these
different
by
stability, for each vacancy that arises it
will be possible to estimate the mean number of moves made in the
labor market.
of the
This is the mean passage time.
model
input-output
This configuration
will make it possible to observe the
way
demand for labor in occupations (rather than the traditional
of demand by industry), and
by
estimating
vacancy
to
chains,
observe the normal recruitment pattern of labor by occupation.
I will focus on the occupational groups of
service,
of
because
the
recent
changes
unskilled labor and
in
employment
occupational opportunities that have arisen in them
characterized
manufacturing
to
a
as
reflecting
a
"post-industrial"
transformation
service
tranformation has been of concern to both social
policy makers who would like to better understand
results of this ongoing process.
to
the
These structural changes have
structural changes in the economy.
been
due
and
from
economy.
scientists
the
a
This
and
potential
Occupationally, one result is a
decline in the number of traditionally unionized blue collar jobs
-
3 -
and an increase
the managerial, professional,
in
service occupations.
technical, and
Although I will not analyze this,
it
will
show up in the occupational demand for labor.
Occupations
that
are suffering from the economic
will
tranformation
have
fewer
a greater number leaving for other occupations.
recruitments and
Occupations that are expanding should have a
greater
number
of
recruitments.
My interest in this methodology stems
disagreement with micro-economic
work.
If
we
were
part
from
about
assumptions
a
why
basic
people
to believe those assumptions we would assume
work because wages are high enough to lure the
only
that people
in
lazy human being away from the pursuit of leisure activities.
actuality we work for many reasons and we
reasons
other
than
income.
There
In
choose occupations for
are
many
social
and
psychological reasons for working and choosing the occupations we
do.
Many
are
not
so
much
the
the
product
of
of
individual
macro
social
characteristics
as
they
processes.
is
it that most women enter the labor market as
Why
are
products
sales or clerical workers?
Why do people
backgrounds most often seek work
Why do
people
college?
categorize
Why
from
are
answers
gender, class
and
in
from
"working
class"
high paid blue collar jobs?
"middle class" backgrounds most often go to
blacks
to
race.
usually found in low status jobs?
these
questions
as
We
discrimination--by
The result is that the labor market is
stratified even before we start looking for our first jobs.
- 4 -
Politicians and
policy makers who make public statements about
employment often only concern themselves with the number of
created.
This
is
misunderstanding
of
an
oversimplification
the
labor
market.
that
Along
leads
jobs
to
with new
a
jobs
appearing and old ones disappearing, changes are constantly going
on.
There can never be a
fixed number of positions in the labor
market but rather an environment
in
destructions consistant
constant
with
a
flux with job creations and
size.
Policy makers
often talk about jobs but not about what kind or the needs of the
local population.
for them as
Jobs are not homogenous and people do not look
though
they
were.
Few people approach the job by
thinking, "there are x number of job vacancies, I think I'll find
one."
We think of occupations and possible careers, our previous
experience, and level of
education.
More
that some type of stratification exists,
place (level) where we
are
suited
Implicitly we think in terms
of
precisely, we assume
and
or
that
there
interested
in
is
working.
opportunities, not jobs.
is simply an opportunity for career
advancement.
non-career job is found in a specific occupation
A
a
A job
career
or
that is thought
of as having a general set of qualifications.
The usually static nature of the labor market assumed by policy
makers
can
potentially
economic development.
of
industries
with
lead
to
If policy
job
poorly
planned
policies
of
makers promote the introduction
mixes
that
educational and skill level of the
are
local
not
suited
population,
to
the
then
the
policy will promote employment for people outside, and will force
-
5 -
either unemployment or employment
in
lower
wage/skill work, or
migration upon the local market.
The methodology
gives a fuller
labor
potential
picture
results
will
the
market
of job creation upon
occupational mobility.
stable it
of
job
I
am
testing
by observing the
opportunities
and
If the vacancy chains and multiplier are
be possible to project the effect upon the labor
market of future changes in occupational demand for labor.
I hope to find that the methodology to be used in this analysis
will lend itself to use in policy analysis.
At the very least it
is
for
an
interesting
sociological
tool
analyzing
institutions and occupational stratification defines
mobility in the labor market.
- 6 -
and
how
limits
THE MODEL
The model I will use is based on one outlined by Harrison White
in his book Chains of Opportunity: Systems Models
Organizations.
but
The
of Mobility in
model is mathematically the same as White's,
has
different
assumptions,
Roderick
Harrison.
Harrison's
which have been
model
is
developed
referred
to
as
by
an
opportunity model while White's model is referred to as a vacancy
model.
models.
The difference is due to the intended
The vacancy model is
in organizations, while the
aimed
at analysis of job turnover
opportunity
analyses of national occupational
application of the
model
is
for
use
in
structures.
White defined four types of systems:
"1.
Tight
systems:
limbos
are
shorter
than
vacancies; men move with little or no interval from one
incumbancy to the next whereas some time is required to
fill vacancies.
2. Loose systems: vacancies
are shorter than limbos;
vacancies are
usually filled at once whereas men spend
some time floating in a limbo status between successive
jobs.
3. Coordinated systems: vacancies and limbos are both
negligible in length.
4. Matchmaking systems: limbos
and vacancies are (on
the average)
comparable
and
substantial
in
length
(note: some part of
the
defined
system of
jobs
if
examined
separately might
fall within
this
type
although the whole
system
is tight or loose)" (White,
p.
8).
The vacancy model assumes a tight system.
defined jobs seperately for
positions.
the
individuals
White identified and
who occupied these
He also followed the actual person in
-
7 -
the job.
This
is not necessary for calculating the model's equation but he used
this information to check the accuracy
lengths.
jobs.
Each job stratum is
For a system to be
in a state of limbo
in
the
system).
composed
are
go
type of situation
chain
of an homogeneous set of
no
limbos
as
soon
in
the
as
The
system
they
A "tight" system is likely to
only if there are no quits.
who
estimated
(limbo is defined to mean no job or position
individuals move into vacancies
are those
his
"tight" its members must never be left
There
current positions.
of
leave
be
when
their
maintained
only individuals who leave jobs
to other positions in the organization.
could
only
occur
in
the
This
highest levels of
bureaucracies found in large companies, governments or churches.
White's
assumption
is
that
"pulled" by vacancies
in
the organization, that movement around
individuals can only
the organization is between jobs and that these
in
different
organization.
strata
according
to
their
However
fill vacancies,
vacancies.
turn
creating
job
generates
openings
mobility
Individual chains
of
The organization would find
into the newly vacant position.
the
is
incumbent's
former
in
the
their
throughout
is
jobs
to
to
the system,
dependent
upon
(observed as vacancies)
created
a
sufficient
or
an
replacement
incumbent
who
moves
The vacancy then would move into
position.
-
is
leave
that
moves
would occur when the new position
retires.
jobs are located
"level"
this
create chains of moves, where individuals
in
when
Vacancies only enter the system by the retirements
of personnel or job creations.
which
move
8 -
Vacancies
are
conceptual
This
they move in the opposite direction as do people.
entities;
process is the path of vacancies throughout
their
from
"pulled"
when
Because
system.
a
models
method
the vacant positions, White's
into
positions
move
to
assumed
are
individuals
a
chain of vacancies.
White's vacancy
model
represents
using an embedded, discrete time,
vacancies.
this
stratified
Vacancies arrive in a system by occupying
and can move to
other
and/or to the outside.
in
jobs
Each job
stratum
some
job,
or different strata,
same
the
for
chain
markov
first order
system by
is
comprised of one type of (homogeneous) job.
considered
Each
to
vacancy
be
that
arrives in a stratum must move independently of the others and of
its
prior
the
history, to another level or to outside
with conditional probabilities q rc (row, column) and
outside) respectively.
total
number
of
makes
This
vacancies
system,
q ro (row,
it possible to determine the
arising
in
the system due to
the
initial requirements and job creations.
Vacancies are assumed
beginning of each time
counted
and
to
wave of
in
(usually
period
a
single
one
represented in a row vector F(t),
(number of strata) vector.
each cohort
arrive
sets
off
cohort at the
year).
They
are
that is a 1 by
The number of arriving
vacancies
waves of "moves" by vacancies.
S
in
The first
personnel redistribution would be represented as F(t)QI
the second
wave
as F(t)Q 2 , the hth as F(t)Qh.
The total number
of vacancies generated, represented in row vector M(t)
-
9 -
would
be
obtained
summing each wave from h = 0 tocsO.
by
The result
is
the fundamental equation for vacancy models:
1.
M(t)
Li
= F(t) (I-Q)~1
M(t) = [F(t)]
la.
I-Q
where
the
sum of
the
infinite
represented by the formula for
Above M(t)
F(t)
is
series
the
the
retirements and
initial
new
cohort
jobs.
vacancy
multiplier
stands for the total number
for
of
of
of
(I-Q)~
waves
(I-Q)~ 1.
effect
vacancies
vacancies
represents
generated.
arriving
first
cohort
of
arrivals.
via
the multiplier
effect or the subsequent waves of moves made in the system
result of the
is
as
a
The sum is the total
number of vacancies M(t) arriving in each stratum.
In equation 1, the mobility process is formally represented
generated
by
positions.
possible
vacancies
that
arrive
by
retirements
and
new
Once the transition matrix has been calculated it
to
estimate, by calculating mean
passage
as
times,
is
the
average number of vacancies created in the system by a retirement
or job creation arriving in each stratum.
The assumption in White's model is
stationary probability
which
stratum
it
each
vacancy
has
a
of moving to another stratum according to
currently
resides
independently of all other vacancies.
assumed time limit (i.e.
that
in.
Any
10
-
moves
Vacancies have a stated or
one year), after which
-
vacancy
the position is
assumed to be abolished.
Because vacancies arrive continuously,
the time limit set would
defining
cohorts
of
be
an
vacancies.
probability of moves to other
White
tested
artificially
this
The
created
diagonals
on
clergy
in
three
the
denominations and found it useful in predicting
the vacancies in the system.
He did not
tests because he did not use a
the system.
This made
compared the predicted
represent
the
jobs in the same stratum.
model
entire chains by following
boundary
Protestant
outcome
of
perform any statistical
random sample but instead sampled
their
paths
statistical
outcomes
of vacancies throughout
tests
against
difficult.
He simply
the observed outcomes.
This lack of statistical testing created some controversy.
Opportunity Models:
The opportunity
model as defined by Harrison is mathematically
the same as the White vacancy model.
occupational structures
identifiable jobs,
are
not
However,
"tight"
or
because
national
decomposable into
different assumptions must be made when using
opportunity models.
Clearly, due to quits, lay-offs, and periods
of unemployment between jobs, national labor markets do
White's definition of
"tight".
National
not
fit
labor markets contain
too many unstable jobs, many of which are tailored to individuals
or are simply self-employment.
characteristics
position
rather
depend
than
on
on
An
the
a
unstable
individual
defined
set of
job
who
is
occupies
11
-
whose
the
responsibilities.
Another (and more important) difference in the analysis
-
one
is
that
of
in the vacancy model, White identified the movement
vacancies and individuals between
identifiable
jobs
specific
as well as
between strata.
A
model
of
national
occupational
conceptualized as being similar
defined by White, but
this
the transition from vacancy
to
require
quits
and
a
opportunity
more
complicated
models.
limbos
In order to
use
between
jobs
concepts
from
so that the
the vacancy model new assumptions must be introduced
a
opportunity model represents
First, the
lost, and second, due to
is
layoffs, there are substantial
which complicates this process.
be
that must be solved in
problems
ability to identify individual jobs
might
"matchmaking" system as
the
would
There are two important
model.
to
structures
"tight" system and can therefore
make use of equation 1.
The reason for redefining
model is the
assumptions
problem that arises
distinguishable jobs.
in
calculated
the
because
in the opportunity
there
Only net changes in stratum
the opportunity model.
The
are
no
size
occupational
used for this study are the same as defined by the
U.S.
longer
can
strata
Census.
This assumes that the skills and qualifications needed for
jobs are more similar
the other occupations
occupations
"fixed."
rather
to
each
In the model we
the
these
other than to the job holders in
(occupational
than
be
strata are homogeneous),
individual
substitute
jobs
aggregate
are treated
change
in
and
as
the
strata for observation of the actual job creations and abolitions
-
12
-
that take place.
An
example
of
an
occupational
acquire labor in order to meet demand
for labor necessary for it
and services that is taken as
goods
net
The
some
represents
change
of job creations, abolitions
number
unknown
for
demand
predetermined.
exogenously
is
One assumes that this stratum must maintain and
service workers.
to satisfy the
stratum
terminations.
and
Because individual job creations, abolitions and lay-offs can not
identified
be
in
this
of
type
micro-processes) they are simply
stratum size.
labor.
taken
as
for
the
labor
form
the
within each stratum
of
occupational opportunities."
net
change
is
(Harrison, 1983 pg.
individual jobs, the formula used in
unseen
in
labor
the
exogenously
but indistinguishable
denumerable
To distinguish aggregate demand for
are
(they
occupational demand for
This is assumed to be the
Demand
generated, "in
analysis
9)
from
vacancies
opportunity
model
in
is
modified so that:
2.
M(t) = D(t)
(I-Q)-1
This differs from the vacancy model in two respects.
for net demands for additional labor in each
stands for
the
Q
individually
identifiable
D(t) stands
stratum, while F(t)
vacancies.
Additionally,
matrix in opportunity models is a zero diagonal matrix.
In
this model it is also possible to have a negative D(t) net demand
for labor.
Similar to White's vacancy model,
in each strata,
now
defined
as occupations, the net demand for labor is represented in a 1 by
-
13
-
S
row
vector,
D(t).
The
demand
outside the stratum, either from
for labor is recruited
inside
the labor force or from
outside the labor force, with probabilities
q
rc,
q
ro.
instead of abolishing unfilled positions they are netted
the calculation of stratum size.
=
2a. M(t)
In the
aggregate
model
ABOLITIONS)
the information
is
not
out
in
(I-Q)
it is only possible to identify the
movement of individuals between the strata.
although the White analysis
But
The calculation for M(t) is:
(NEWJOBS + RETIREMENTS -
opportunity
from
does
identify
However,
the individual jobs,
used in equation 1. (The information is
used to measure, identify and test chain lengths.)
Because
the
model
can
formula only models the aggregate movements, the same
be used for both vacancy and opportunity models.
Estimates of the demand for labor by
comparing the
Q
matrix
occupation can be made by
difference between results derived with the pooled
and
the
probabilities of
observed
data.
Q
The
matrix contains
the
recruiting labor from outside the labor market,
and from the other occupations inside the labor market.
2b-
b
q
=
rc
Each
row
cr
Mr
of the
r
Q
=
1 to s, c = 0 to s,
r =
c,
M =
rowsum
matrix has a probability of 1, each row
cell
contains the row probability of recruiting a new entrant from the
outside the labor market or from other occupations.
matrix or aggregate is derived by adding
-
14
-
The pooled
Q
all the elements of the
matrix is
equation 2 to estimate M(t).
in
used
The
matrix.
aggregate
yearly matrices to get an
between the estimated M(t) and the observed M(t)
aggregate
Q
The differences
calculated
are
and compared with the standard deviations.
The mean passage time is calculated by
Q)
(I
-
would give the mean expected
This
by a unity column vector.
post-multiplying
number of moves or opportunities generated in
each
per
stratum
opportunity initiated by a retirement or job creation.
just
Interestingly, opportunity models are
input-output
by Leontief,
developed
models
generally known as the Keynsian multiplier.
and
formally
all
models
input-output
a
special case of
where
is
(I-Q)~1
Vacancy, opportunity
contain
the
same
Therefore it is possible to conceive of
mathematical equations.
opportunity models as adaptions of input-output models.
linear functions over
organizations of
and
goods
the
of
produce a unit
time,
labor
inputs required to
services
demanded are fixed
the
Input-output models assume that
given the prevailing technology and
these
Given
production.
assumptions,
it is
reasonable to assume that aggregate demand for occupational labor
the production needed to meet
is also a fixed linear function of
the
exogenously
imposed
for goods and
demand
services.
The
treatment of the labor requirement of the economy in input-output
models is similar to
opportunity models.
that convert
the
treatment
output
-
occupational
industry-by-occupation
The use of
anticipated
of
15
into
-
demand
labor
in
matrices
for occupational
labor is a standard practice among economists.
In
order
to
test
the
validity
and
made
opportunity model, Harrison has
feasibility
some
of
the
preliminary tests on
data from the Special Labor Force Reports on job and occupational
and 1977. Preliminary calculations from
1972,
mobility in 1965,
the model suggest that recruitment
stable and
patterns
by
occupation
types of analysis can be made with this model.
other
One counted job
He tested two versions of the opportunity model.
the
changes on
the diagonal.
are
while the other zeroed out changes on
diagonal,
The first version includes a test
for job changes
make
The second version does not
within the occupational group.
this test and explicitly assumes that the occupation can not fill
increased demand for labor by the occupation
its own
The
members.
by
recruitment
of
tests that Harrison made resulted in the
diagonal movements not being in proportion to the total number of
He found that most of the movements
vacancies in the occupation.
diagonal
the
along
tested.
significantly
differed
The results suggested that the number
within occupations
This
for labor.
recruitment
more
of
job
could
be
the
effect
due to economic cycles.
on
upon
or
During economic
experience
that
is
directly
Thus with less demand for labor,
much
recruitment
from
16
-
more
relevant
the
to
work
the
result will be
outside the occupation,
-
slowdowns,
the labor force, employers can
occupation.
lower
changers
labor mobility or
selective when hiring in terms of requiring
experience
years
the
is not dependent upon the occupational demand
due to the increased pressure
be
across
but
the
possibility of increased
recruitment
results were more consistant
accurate
more
The
occupations.
for recruitment between occupations
The estimates of
with the diagonal zeroed out.
the jobs started
the three transition matrices were
of
as derived by the pooling
inside
of
for the
version
of
opportunity
the
zero
the
with
model
diagonal.
The strength
illustrate
the
the
all
in
movements
model
the
is
its
ability
The
market.
labor
to
opportunity model does not treat national occupational structures
systems
as aggregations of tight institutional
of
fixed
jobs.
This approach might be thought of as an aggregated internal labor
market
The
model.
model
relies upon assumptions
similar
input-output models, so if these assumptions and those of
chains are
valid
(and
to
markov
reasonable) then these can provide a new
and unique way to view the labor market.
The opportunity model has in effect abandoned
calculate individual
Aggregation is
able to
but
it
predict
a
events
potential
the
or
resulting
any
attempt
to
vacancy chains.
weakness in the model.
It might be
the movements in the labor force in some detail
does this by aggregation and therefore can
some of the forces at work that underly this process.
-
17
-
not
analyze
THE DATA SET
The data for this study is taken from the Panel Study on Income
Dynamics (PSID).
The PSID is conducted by
the
Survey
Research
Center, in the Institute for Social Research at the University of
Michigan, Ann Arbor, Michigan.
This
survey
has been conducted
yearly since 1968. I will use data from the years
Its distinguishing characteristic is the
tracking
families since the beginning of the survey.
the
fact
that
as
children
1969
to
of
Also
1983.
the
same
important
panel families form their
of
households, they are retained in the survey.
weighting, the study has continued to be
Through
is
own
the use of
of
representative
the
U.S. population and its labor force.
Due to the disproportionate sampling of low income families and
the
subsequent
necessary.
dropout
Three
rate,
basic
weight
probability of obtaining an SRC
obtaining
an
interview
weighting
of
of
measures
interview,
sample
were
the
reinterviewed
the
was
used--the
probability
families,
and
of
the
probability of obtaining an interview in the combined samples.
The questionnaire includes questions on housing, utility bills,
commuting,
housework,
health,
occupations,
industry,
wages,
employment status, size of household, size of house, relationship
of the household members, ages and birth dates.
In 1968, 4802
from
two
families
groups.
One
were surveyed.
was
from
-
18
a
-
Families selected came
cross-section
sample
of
from a subsample of families
selection
Census in 1966 and 1967. The
by
made
low income families for the OEO was
of
families
All of these
(OEO).
U.S.
the
by
by the U.S. Census for
interviewed
the Office of Economic Opportunity
had been interviewed
and the second
part of the U.S.,
contiguous
the
dwellings from
the
formula $2000 + N(family size) * $1000 or approximately twice the
as
line
federal poverty
It
1967.
in
defined
excludes
also
Statistical
families living outside Standard Metropolitan
Areas
(SMSA) in the Northeast, the North Central and the West.
By 1983
number of families in the survey had increased to
the
6852, due to children leaving the household and establishing
households, which
institutions
sample.
Of
college,
the
the
in
original families, 10,925 members remained
these, 325 were away at
In 1983, of the
in the sample.
retained
are
or
(school
Since 1968,
armed forces, hospitals, nursing homes and prisons).
been
5016 children have
members
to
born
new
and
4386
nonsample
individuals are living in sample families.
The respondents to
the
sample
are
from 49 of the 50 states.
Only Vermont is excluded, as is the District
is a low of one family from Montana, and
a
of Columbia.
high of 649 families
PSID families now also live in
from California.
There
Puerto Rico and
11 foreign countries.
Of an estimated
interviewed.
subtracted due
7050
families,
6852
or
If 55 of the respondents from
97.1
percent
previous
years
were
are
to death, illness or institutionalization, or the
-
19
-
reuniting of a couple, then the response rate is 98 percent.
Quality and Comparison:
The Survey Research Center considers the quality of its data to
be good,
due
assignments
the
to
response rate and the low number of
high
internal
by
(data corrections,
assumption) made by
the Center.
... there is very little year-to-year variation in the
quality
the
number of assignments we have to make, so
continues
of the data, according to our measure of it,
to be good (PSID, p. 3).
attrition
from the original
respondents in the survey, there might be some
question about it
Since there has
been
lot
a
possible
is
Population Survey (CPS).
PSID
to compare the PSID with the Current
The CPS allows the
Population Reports, Consumer
1981,
and the
between
PSID
the
comparison
refer to: Current
difference
and
CPS
in
the
surveys.
way
members, 14 years old and over, present as of
the
interview (March).
all
date
household
of
PSID includes in this figure any income
people who moved in and out during the year.
be
income
The CPS asks for
of
they
126, June
income is reported
sources for the entire previous calendar year
that
income
and
Series P-60, No.
Income,
the
1980 data set for the PSID.
minor
There is a
of
For purposes
characteristics.
of
evaluation
for demographic
1980, relative to the CPS,
in
the self evaluation of the
Besides
still being representative.
Institute, it
of
living
household
in the
-
20
-
at
the
from
It does not require
the
time
of
the
are made
comparisons
Income
interview.
by
this
subtracting
amount off the income of PSID responding families.
to get higher median income than does the CPS.
tends
The PSID
There are two important reasons that could
contribute
First, the participants in the PSID are "practiced,"
do the survey every
and are cooperative.
year
this.
to
that is they
are used to filling out the forms
and
affiliation
Second, there is no
U.S. government which could serve to
inhibit
with
the
to
the
responses
CPS.
The only major
greater
is the
For
the
demographic
of single person households in the PSID.
number
variables
difference between the two surveys
age,
education,
race,
distributions were all within two percent of each
and
region
the
There
other.
One, by
are two reasons there are more single person households.
following families over time it is easier to locate single person
households.
from
the
The panel nature of the survey means that
splitoffs
Two,
surveyed families are kept in the sample.
when
splitoffs return to their original families they are continued as
a separate household (by PSID convention).
Advantages and Disadvantages of the Data Set
The PSID
data
panel nature
of
set is interesting in this study because of the
the
By following the same families and
study.
many of the same individuals not only can occupational demand for
labor
be
estimated
and
vacancy chains estimated,
-
21
-
but
actual
It is not my intent
career paths of individuals can be observed.
Further work
to follow individuals in this study.
paths
career
individual
of
set could include an analysis
with the data
these could be compared with the estimated vacany
and
chains to test
how close vacancy chains resemble an individual's career.
The data set also includes many variables
that can be used for
a more detailed analysis into which types of individuals and what
types of family backgrounds are
These variables include questions
and higher status occupations.
on:
mother
father's
and
to lead to higher paying
likely
occupation,
mother
and
father's
of
income,
and
housing.
education, current family size, types
that
This allows an analysis into many background variables
may
play a role in people's job and occupational choices.
It
would
be possible to ask questions such as,
often work in the same occupational groups
as
children
"do
their
parents?".
Similar questions about how education passes down from generation
to generation could be
household work
exploration
of
the
in
or
asked
same occupation.
interesting
very
some
to check if people in the same
This could lead to the
-questions
that
are
sociological but affect the economic and occupational structure.
The disadvantages
of
the panel nature of the study have to do
with its ability to remain a representative sample of the current
U.S.
after
population
representativeness
of the sample.
16
Part
years.
of
this
is dealt with by the occaisional
problem
reweighting
But the survey still has a problem having
-
22
-
of
to
do
families,
newer
immigrants
are
there
with migration to the U.S. Because
potential problem is that the PSID follows families,
in
this paper about the concentration
surveyed
Another
while
in
one occupation of all
the
working individuals from a member family.
is
no
This may result in the
aggregate sizes of the occupations not being proportional to
actual sizes in the U.S.
-
23
-
this
analysis
There
study is concerned with individuals.
new
represented.
not
are
no
the
RESEARCH CONTEXT
In 1970 Harrison White's book
was
Models of Mobility in Organizations
Opportunity:
of
Chains
System
This
published.
book
operations
attempted to link sociology, institutional economics,
research, and management science in a new way in order to examine
mobility
within
organizations define and constrain mobility.
analyze
how
In the last fifteen
field
this
in
done
been
years additional research has
to
White wanted
organizations.
using
White's methodology.
vacancy
chains
and
developed a mathematical model to predict their mean
length
and
observing
and
In his book,
more
White
defined
fully
His methodology is also a tool for
distribution.
analyzing vacancy chains in concrete
model
his
White tested
by predicting the movement of clergy in
comparing
by
and
denominations,
Protestant
three
situations.
organizational
these
predictions with the actual vacancy chains.
developed his vacancy chain model, this methodolgy
Since White
has been used to analyze job
types of organizations.
the
of
analysis
with
organizations
service
organizations,
apprenticeships,
"Internal"
recruitment,
is
and
mobility
in
other
It has most prominently been applied
internal
an
and
vacancies
internal
labor
market
corporations,
professions
Examples
markets.
labor
trades
requiring
defined to mean administrative
would be
to
of
civil
requiring
accreditation.
rules
regulating
training, promotion, lateral movement, seniority and
-
24
-
retirement.
defined
Thus the internal labor market is
under control of, a firm or other centralized agent.
industrial
and
Analysis of
sociological,
the
analysis
organizational
or
relations,
in
seen
internal labor markets is generally
by,
literature.
Several versions of the model developed by White and incorporated
movement of
been used to predict the
have
into internal labor market models
jobs and vacancies in organizations.
In the fifteen years since the publication of White's book some
this field has been done by Shelby Stewman at
in
important work
Carnegie-Mellon University.
Administrative
Science
Quarterly) he reviews the
and presents his new findings.
in
promotions
bureaucracies.
in
work
in
the
His paper offers an overview of this research
field since White.
analyzing
"Demographic
(upcoming
Markets"
Labor
Organizational
of
Models
paper,
latest
In his
Stewman's
analysis,
has
and
forces
police
In this type of
work
focused
other
on
city
for each job level as
defined in the organization, a multiplier effect is found.
Stewman's research is directly related to White's
organizations.
it
is
possible
that
it
created within these
chains
vacancy
examines organizations and
in
As a different extension of White's work, I think
to
apply
methodology to examining
same
the
mathematical
labor
national
model
and
markets, as contrasted
with internal labor markets.
White's model assumes that
new vacancies in
an
retirements
or
job creations open
organization, starting vacancy chains which
-
25
-
job
assumption concerning vacancy chains and
place of
organizations,
models,
a
occupation/organizational
vacancy chains would
outside
labor
the
affects
multiplier
organization;
in
be
The
level.
has
Leontief's
the same
in
mobility
other
Harrison's model it affects
form
to
job
mobility
multiplier
effect
found
in
promotion,
lateral movements are straightforward, and jobs and
When analyzing labor markets,
their ranking are clearly defined.
unemployed
promotion, only mobility.
ordering the
in
In White's model, when analyzing
an organization the inside and outside are clear cut,
employed and
the
of
parts
the
In both models, mathematically,
the
as
input-output models.
demotion and
eliminate
model
White's
In
force/organization.
job
way
the
to exclusively recruit individuals from
other parts of the labor market.
this
only
in
In both
for
determined
is
multiplier
job
but
mobility,
occupations.
substitutes
it
same
the
makes
Harrison's model
in turn produce job mobility.
occupations.
or mean number of
moves
clear, but there is no demotion or
are
There is nothing to prevent
one
from
Finally, both have mean passage time,
the system as a result of the
in
made
original vacancy.
There are important reasons for examining the labor market this
way.
First of all, most economists do not
examine
many
of the
human factors other than income involved in what motivates people
to work.
Sociologists, on the other hand,
give
less emphasis to
income but examine social economic status (SES), mobility, family
Work, in sociological
background, and discrimination.
-
26
-
research,
rather than from a macro level view.
economic
this way, using
view
of
point
individual's
the
is usually examined only from
labor markets in
Examining
and sociological concepts,
techniques
fields
transcends some of the limitations imposed by using these
separately.
Assuming recruitment
patterns can be established as stable, it
is possible to examine other effects upon the
patterns can be broken down by region, gender,
or
these
observing
By
level.
educational
mobility of labor, we can develop
enable us to see
effects upon
The
market.
the
a new way to observe the labor
patterns
will
moves affect other parts of the labor
job
how
The
race, age cohort,
The ability to look at internal mobility
market.
am
I
methodology
using
will allow me to
Thus
recruitment patterns by any of these groups.
the
track
outcomes
labor market of different demographic characteristics can
in the
be followed.
The analysis I will use will permit a more thorough
look at the dynamics inside the labor
from
labor market.
analysis,
micro-economic
a
market than we usually get
or
from
a
sociological
perspective.
This model has only been used in this context by Harrison.
significance
of
this
model of the
markov
occupational
market is the ability to observe the structural limitations
makers
to
discuss
how
labor
upon
such as professional requirements
occupational and job mobility,
that are barriers to entry.
The
It is a
changes
-
27
in
-
common
the
practice
economy
by policy
affect
the
availibility of jobs,
but not to discuss effects of these changes
upon those already in the labor market.
Explicit
and
implicit
structures in the labor market affect the career choices of those
in
the
labor market and regulate the points of
might better
inform
view
This
unemployed who wish to enter.
of
the
entry
labor
for
the
market
the policy maker on how and where to create
jobs, as a result of the effects of mobility.
In order to analyze the normal recruitment pattern in the labor
market, so that differences over time or other variables (such as
a model must be able to observe
race or gender) can be observed,
these types
the
of movements.
observation
patterns over
of
time,
The Harrison opportunity model allows
recruitment patterns,
an
observation
of
the
ability to observe demand and supply of
labor, and the ability to calculate the multiplier for the entire
system.
-
28
-
Q
THE
MATRIX
ability of the
This paper represents a preliminary test of the
This will involve using a
Q
made
moves
each occupation, and the number of
into
entrants
job
of
opportunity model to estimate the number
in
system.
the
matrix (see chapter on the model) on
the PSID data set (see chapter on the data set).
In
Q
Harrison's test of the
matrix
CPS
on
all
data,
the
predicted number of entrants to each occupation were within 5% of
cells off the diagonal were insignificant at
differences between
That is, M(t) (predicted vacancies), was estimated
the 5% level.
by
the
model).
(the first cohort of arrivals)
equation D(t)
(I -
multiplier
If
most
that
found
With this data Harrison
the observed numbers.
Q)~
details
(for furthur
opportunity
making tools then the
Q
see
times
the
on
the
chapter
are to be widely used as policy
models
matrix must be stable in order to predict
changes in the labor market.
Stability can
be
against
the years surveyed
of stability
yearly
made at five
by comparing the
Q
matrices from all
their standard errors.
Another test
to compare an aggregate matrix with each of the
matrices.
advantages.
relative
is
tested
This
is
a
where
yearly
survey
With the estimates of movements in the
year
stability
Harrison
intervals,
in
of
most
the
found
cells.
that
Most
some
has
force
labor
there
was
differences
between the cells with larger variations were partly attributable
to
changes
in the definition of the occupations by
-
29
-
the
Census
between two of the periods surveyed
Due to the choice of
categories used by the PSID
occupational
for some of the early years, it will only be possible to test for
generally accepted occupational categories.
and
(5)
operatives and transport,
labor,
farm
military, and miscellaneous.
clerical,
and
(8)
From 1976
to
service,
protective
1980 PSID used a two
(2) digit code that lends more flexibility to the analysis.
1981 to 1983 the
Census.
the U.S.
some
Due to the relatively low number of people in
the occupations,
of
cells
relative
will not be accurate.
use
the
to
probabilities
transition
population as a whole, some of the
the
and
farming,
as
such
weighting to get the correct proportions
the sparse
From
the same three (3) digit codes as
used
survey
(4)
(6) service, non-farm labor
owners,
farm
(7)
more
They are as follows:
sales
(3)
(1) professional, (2) managerial,
craft,
combines
one (1) digit coding of occupations that
a
used
the years 1968 to 1975 the PSID
For
stability in a general way.
of
U.S.
in
This means that some of
the occupations will appear to have large changes in their yearly
recruitment patterns.
pattern of exiters from
have
It will
cells,
these
a
effect
greater
the
where
on
pattern
the
will
appear to jump around.
In order to test stability over the whole 15 years, I
model with the the same eight by eight
period.
matrices
From 1976 on I ran a more detailed
13
which it was easier to interpret the categories.
-
30
-
for
ran
the
the entire
by 13 matrix, in
The
categories
in the 13 by 13 matrix are:
sales,
(4) clerical, (5)
(8) unskilled labor,
craft,
appear
Opportunities will
by
vacancies filled
protective
(11)
I will
service.
household
labor and service occupations are
leaving.
workers are staying vs.
many
growing in size, and how
operatives, (7) transport,
farm labor,
private
unskilled
examine whether the
(6)
(10)
(9) farm,
service, (13)
service, (12)
(2) managerial, (3)
(1) professional,
tables
the
in
other
from
workers
workers
analyzing where labor and service
of
percent
the
as
occupations.
Besides
opportunites
find
I
will analyze the opportunities available in the labor and service
present
some
I
workers from other occupations.
for
occupations
descriptive statistics on occupational
also
groups
will
by
age, education, income and race.
Results of the Eight Category Matrix
In
this
section
I
analyze
the
stability
-of
matrices.
different
mobility in the labor force using two
occupational
The
reason for testing an 8 by 8 matrix is the possibility to observe
stability over a longer time period.
covers the
whole
Therefore the 8 by 8 matrix
time period involved while the 13 by 13 matrix
covers 1976-1983. What follows are the results from analyzing the
8 by 8 matrix.
The aggregate
Table
Q
matrix for the 8 by 8 matrices are presented in
1. The 15 matrices for each of the
appendix.
The calculation
of
Q
the
years
matrix
appears
is
31
-
the
made by first
calculating the vacancies that appear in the labor market.
-
in
This
calculation is made by surveying the working population
starts, and moves.
a
was
With each move it is fair to say
determined
are
these
Once
vacancy.
the
Next
diagonal is
probabilities of recruiting
that
so
zeroed
those
numbers are then summed across
rows
labor force who entered during
the
there
is
matrix
than
people.
the
the occupation.
The
outside
the
can
those
adding
year.
job
calculate
we
outside
that
the
transposed to get the movement of vacancies rather
for
Lastly, the row cells
are divided by the row marginal.
TABLE 1: The Aggregate 8 by 8
Outside
Prof.
Mngr.
Craft
0.189
Sales/
Cler.
0.223
P
0.330
0.000
M
S
Cr
Op
LS
F
M
0.123
0.419
0.140
0.243
0.477
0.373
0.203
0.210
0.111
0.070
0.021
0.051
0.057
0.125
0.000
0.190
0.1430.050
0.043
0.170
0.033
0.244
0.000
0.071
0.102
0.104
0.065
0.214
Q
Matrix
0.076
Oper./
Trans.
0.044
Labor
Service
0.106
Farm
0.001
Misc.
0.025
rowsum
74820
0.161
0.046
0.000
0.301
0.096
0.072
0.099
0.068
0.089
0.399
0.000
0.197
0.036
0.178
0.080
0.080
0.140
0.250
0.000
0.217
0.146
0.010
0.006
0.007
0.007
0.024
0.000
0.001
0.004
0.020
0.030
0.024
0.018
0.009
0.000
93704
140822
86789
108181
132687
6626
9250
1983, 33.1% of
In the professional occupation, between 1968 and
vacancies were
labor force
filled
by
(unemployed,
unofficially unemployed).
those who were defined as outside the
time
full
students,
housewives, and
In three of the years, recruitment of
this group makes up over 40% of all professional recruits, and it
is
always
occupations.
the
largest
cohort
of
recruits
to
professional
This would seem to confirm an intuitive notion that
-
32
-
more training and therefore
the professional occupations require
many
largest
second
The
graduates.
groups of recruits
in
recruited
The remaining occupations are
school
professional
from
come
(22.8%).
workers
sales/clerical
managerial workers (18.9%) and
7.6% by craft workers;
or
college
to
go
would
positions
much lower numbers:
10.6%
4.4% by operative/transport workers;
by labor/service/farm labor workers; 0.1%
by farmers; and 2.5% by
This would indicate that service workers have few
miscellaneous.
opportunities to move into the professional occupations.
the occupation that received the largest
Between 1968 and 1983,
for
possible to determine if most of the recruitment
positions
come
from sales or clerical
the vacancies were filled by
by
professional workers,
of
the
middle
12.3%
labor
the
of
market,
Although
labor market falls into a
occupations
managerial
recruitment,
A
is:
16.1% by craft workers.
recruitment of those from outside
midrange catagory
outside
those
be
managerial
occupations.
range of recruitment to fill managerial positions
21.1%
will
In the more detailed matrix it
(34.4%).
sales/clerical
was
occupation
managerial
the
in
opportunities
of
number
percent of vacancies filled, not the total number
of
(by
positions)
are the lowest recruiters of those outside the labor market.
The
category
of
8%
by
0.4%
by
remaining
occupations
recruitment,
6.8%
labor/service/farm
all
by
labor
fall
into
a
operative/transport
low
workers;
workers, 1% by farmers,
miscellaneous workers.
-
33
-
and
vacancies in sales/clerical (41.9%)
the labor market.
cohort of recruits to fill
largest
Between 1968 and 1983, the
were
by
those from outside
These occupations are where many women work or
start careers and it is not surprising that there are such a high
Since the role of women
occupation.
number of recruits in this
in our society makes them primarily responsible
for
market
families, many women leave and reenter the labor
occupations, this might be a
recruitment
rates
sales/clerical.
of
A
operative/transport
those
mid
professional workers
contributing
outside
range
workers
labor workers (11.8%).
and
(8.9%),
high
market
by
is
formed
by
(18.9%),
by
labor/service/farm
is made up of,
A low group
the
to
workers
by
several
labor
recruitment
of
managerial
(11.2%),
factor
the
for
in the clerical
Since women also tend to be concentrated
times.
caring
4.6% by craft
workers, 0.6% by farmers, and 2% by miscellaneous workers.
In craft occupations, between 1968 and 1983, the largest
of recuits was 39.9% by operative/transport workers.
to be a large amount of
these two
groups.
occupational
changes
that
group
There seems
go
between
A middle range of recruitment is made up of:
13.9%
by vacancies were filled by those outside the labor market,
14.3%
by
managerial workers, 7.2% by sales/clerical workers, and
14% by labor/service/farm labor workers.
of: 6.9%
by professional workers, 0.7%
A low range
by
farmers,
is made up
and
3%
by
miscellaneous workers.
In
operative/tranport occupations, between 1968 and
-
34
-
1983,
the
high range of recruitments to
vacancies was
The
24.3% those outside the labor market.
to
workers
craft
(but
seamstresses,
category.
But
not
due
the
to
some
occupational boundary
occupations
areas.
Sales/clerical
reciprocal
be
such as carpet
could
that
workers
and
makers
go
in
either
of
of making some type
are
bound
to continuously
closely
job changes in
related
10.1%, made up the middle range.
workers,
of
occupations
remaining
four
to
such as dress
necessity
making
by
switch
The
factory),
of
recruitment
occupations,
craft
occupations,
operative
and
layers,
high
operative/transport seems
There are some
recruitment.
filled by: 30%
by labor/service/farm labor workers, and
workers, 25.1%
by craft
fill
professional workers, farmers, and
workers,
managerial
miscellaneous workers made up
only 10.5% of the recruits to operative/transport occupations.
In labor/service/farm labor occupations, between 1968 and
47.7% of the available opportunities were filled
outside the labor market, by
A middle range is made up
(9.7%)
workers.
craft
workers,
far
and
those
from
the largest group of recruits.
(10.4%)
of,
by
1983,
(19.7%)
sales/clerical
by
workers,
operative/transport
A small number of vacancies are filled by
professional
workers, managerial workers, farmers, and miscellaneous workers.
In farming,
between
1968
and
1983,
the
recruits (37.3%) to fill vacancies were those
labor market.
largest
from
This large group of entering farm
children of farm owners who
inherit
-
35
-
farms,
group
outside
of
the
owners could be
otherwise it is hard
to understand how so many from outside the labor force can afford
A
farms.
buy
to
managerial workers, and
from
(21.7%)
recruits
operative/transport
workers,
craft
fill
workers)
miscellaneous
and
workers,
the
of
occupations
remaining
sales/clerical workers,
workers,
(professional
labor
labor/service/farm
The
recruits.
of
group
(17%)
from
is
It is not surprising to find farm labor in one
workers.
larger
of
range
middle
about 25% of the vacancies.
In
miscellaneous,
which
occupations,
service
1968
between
were
sales/clerical
by
filled
three
are
and 1983, there
occupational groups that make up most of the
vacancies
protective
of
comprised
is mostly
21.4% of
recruits.
20.4%
workers,
vacancies were filled by those outside the labor market,
of
17.8% by
operative/transport workers, 12.5% by professional workers,
9.9%
labor.
The
workers
and
by craft workers, and 14.6%
by
labor/service/farm
remainder of the vacancies were filled by mangerial
by farmers.
A Look at the Labor/Service/Farm Labor Workers
to
In the matrix it is possible
labor
in
two
ways
labor/service/farm labor.
occupational
observe
the
pattern
to
relative
the
observe
the
One method is
recruitment
category
occupational
to
of workers recruited.
observe
The
-
36
-
usual
the
second
opportunities for labor/service/farm labor
other occupations.
of
is
in
to
the
for the years 1969 to 1983.
All the calculations I made are
the table below all the recruitment
rates
almost
market.
farm
sales/clerical
(5.1%),
Recruits from professional
and
miscellaneous
(1.8%)
group
the
(10.4%),
from
and craft
(4.3%),
managerial
are
labor
comes
recruits
employed
The greatest percent of
(1.4%),
profession
half its demand for labor from outside
operative/transport (19.7%),
(9.6%).
each
occupational
This
into labor/service/farm labor appear.
fills
for
In
consistently
Recruits into labor/service/farm labor make up 20.3% of
all
low.
the
moves made.
It
is
difficult
to
draw conclusions about
the
recruitment
patterns of labor into this occupational category from the 8 by 8
matrix.
The
broad definition of unskilled
labor
and
service
occupations put together in one category seems to be the cause of
this
difficulty.
that
provides
market.
It
Nonetheless, this category is
many opportunities for those
seems
that
operative/transport workers
the
into
this
large
an
outside
occupation
labor
the
recruitment
category
might
also be
reflected in a large recruitment into unskilled labor, were
a separate category.
of
that
The recruitment of sales/clerical into this
category would seem to reflect the movement of these workers into
service occupations.
-
37
-
Table 2:
Yr.
82
Prof.
Outs.
Recruits to Labor/Service/Farm Labor
Mngr.
Sales/
Cler.
Craft
Oper./
Trans.
Farm
Misc.
# Vacancies
in labor/service/
farm labor
49.9%
2.5%
2.4%
4.3%
6.6%
26.6%
1.8%
5.8%
13,616
50.6%
3.8% 5.7%
10.6%
5.9%
21.7%
0.6%
1.0%
11,444
48.2%
7.1%
2.1%
10.5%
9.4%
21.5%
0.6% 0.6%
11,936
46~.4%
3.6%
4.7%
18.0% 10.8%
15.2%
0.8% 0.5%
8,952
37.9%
6.3% 4.8%
15.6%
6.0%
26.3%
1.2%
1.9%
11,956
47.2%
8.3%
2 .6%
9.2%
6.9%
22.1%
0.9%
2.8%
8,750
4 3.6%
2.8%
3.3%
8.4%
8.9%
22.6%
2.1%
8.3%
10,237
65.7%
2.3%
6.0% 9.2%
12 .7%
2.6%
0.0%
10,471
17 .6.
0.7%
0.0%
8,
18. 21
2.4%
0.0%
7.982
1b.
0.9%
0.0%
7,968
0.0%
8,148
8.1%
11 .2%
8.5%
8.1%
43.4%
5.1%
3.7%
12.8% 14.3%
53..'4
5.
3.9%
9.0% 11.4%
5 .%
b
7.8%
11.8% 21 .1%
19.7%
7.o%
7 .9%
-4.3%
4.6%
0.1%
9.3%
15.1%
1%
12.2%
7.4%
17.7%
1.4%
0.0%
6,296
11.3%
15.8%
2.5%
0.0%
o,396
Labor/service/farm labor workers made up
force in 1969, while in 1983
all the workers
who
of
the
labor
this number shrunk to 15.1%. Out of
changed
labor/service/farm labor
15.6%
occupations
from
1969
to
1983,
workers made 11.4% of the moves.
Using
the weighted N, during this period 132,687 workers were recruited
to fill vacancies, 63,355 from outside the labor force and 69,332
from
other
occupations;
occupation, 74,178
the workforce.
while
143,470
workers
for new opportunities (jobs),
The net is -10,783,
-4,846 from
-5,937 from/to outside the labor market.
-
38
left
the
and 69,292 left
job changes, and
All the figures indicate
disproportionately
a
receiving
are
that labor/service/farm labor workers
the
of
number
occupations,
changing
opportunities available for
small
relative
to
workers in other occupations.
Table 3. 1969-83 Net Arrivals & Departures From:
Labor/Service/Farm-labor)
NET
DEPARTURES
ARRIVALS
to/from other
occupations
69,332
74,178
-4,846
to/from
outside
63,355
69,292
-5,937
132,687
143,470
-10,783
total moves
range
from
labor/service/farm-labor
into
year
by
Labor market arrivals
a high of 6,882 in 1976 to a low of 2,863
1970.
in
Labor market exits from the occupation range from a high of 7,851
low
in 1977 to a
of
3,260 in 1976. Recruitment of workers from
to
a
low
of
546.
labor
vacancies
range
from
low
in the labor market range from a
35,846.
of
While
labor/service/farm
as a percent of vacancies in the
a high of 24.3%, to a low of 17.4%.
yearly breakdown of table 3 and the labor
above, see appendix).
-
by
range from a high of 13,616, to
labor/service/farm-labor workers
high of 64,355, to a
occupations
other
to
Departures
a low of 6,296. Total vacancies
from a high of 8,581
ranges
other occupations to fill vacancies
39
-
labor
(For a
market
complete
market moves described
Results of the Thirteen Category Matrix
In this section I analyze the 13 by 13
matrix,
by 8
recruitment of labor.
more
a
be
will
This analysis
occupational
of
stability
the
analyze
will
I
As with the 8
matrix.
thorough
8 matrix, because of the more detailed
analysis than in the 8 by
breakdown of the labor market;
I will
also
descriptive
include
statistics in this section.
The reason
for
observe a more
detailed
13
a
testing
by 13 matrix is the ability to
and logical occupational breakdown over
the time period from 1976 to 1983. From 1976
to
PSID
the
1980
listed 27 separate categories, and since 1981 the survey has used
the same occupational groupings as the U.S. Census.
the
Q
The test of
matrix using the PSID survey might be a precursor to larger
tests with yearly CPS data.
The aggregate
Q
matrix for the 13 by 13 matrixes is in table 4.
The 13 matrixes for each
appendix.
of
the
years
can
be
There is no difference in calculation of
found
the
in
Q
in the 13 by 13 matrix from that done in the 8 by 8 matrix.
-
40
-
the
matrix
Table
Matrix
Sale
Lier
craft
Gper
Tran
m 0.
p 0.195
0.252
0.
0.275
0.158
0.042
0.148
0.092
0.036
0. 1b5
0.047
0.032
0.011
0.023
0.027
0.15e
0.013
01.022
tto.016
1 0.011
f 0.122
0.035
0.021
0.067
0.012
u.033
flo.007
pSO.082
s 0.070
pho .001
o 0.068
0.021
o . 087
0.009
0.053
0 .053
0.013
0.002
0.027
Labor
Farff
0.009
0.019
0.016
0.181
0.040
0.168
0.000
FarmL
Irrober
Serv
0.001
0.001
0.
0.005
0.003
0.088
0.004
0.009
0.
0.001
0.042
0.095
0.005
0.055
0.060O
0.001
0.055
0.079
0.016
0.004
0.001
0.021
0 .e87
0.
0.001
0.000
0.002
0.002
0.001
0.008
0.017
0.002
0.013
0.022
0.026
0.153
0.
0.000
0.009
0.013
0.
0.001
0.
0.016
0.
0.00 6
0.003
u.002
professional
0.V58
0.009
0.065
u.442
0.117
0.059
0.046
0.
0.076
0.185
0.2b1
0.008
0.088
0.067
0.053
0.174
0.012
0.
0.0/5
0.0b7
0.024
0.
0.147
0.012
U.
0.
0.
0.
0.083
0.047
0.059
0.02b
0.011
0 .50
U.016
0.014
0.006
PHHW
outs
0.001
0.001
0.
0 . 008
0.001
market
9.2%
remaining
(31.2%)
0.
0.
from craft workers, and
workers,
by
by farm owners,
This
0.316
0.396
0.174
23729
45640
37920
33417
12473
0.260
0.547
0.49 C
U .480
0.059
0.
0.007
8.9%
filled
workers
and
14888
2447
3493
349b
37755
4525
782 7
and
1983,
outside
from
(25.2%).
from
clerical
from service
workers.
14.8%
by
transport workers,
by farm labor,
1976
those
and managerial
were
3o706
47004
0.283
level group of recruits were:
vacancies
0.315
0.140
between
occupation,
rowsurr
0.306
0.201
0.223
0.447
recruits to fill vacancies were
second middle
workers.
0.
0.290
0.049
0.057
0.291
0. 045
0.
0.054
0.163
0.
0.023
0.
0.007
operative
0.059
0.051
0.041
0.009
0.001
0.103
0.009
labor
0.182
0.
.0b2
0.167
0.105
workers,
0.14d
0.
0.100
0.07
T
0 . 004
0.001
the
workers,
0.091
U .043
0.018
0.084
0.119
majority of
The
Q
13
Prot
O. 109
cro.068
the
by
13
The Aggregate
Para
S 0.056
In
4:
sales
by
workers,
unskilled
labor
by protective service
would indicate that service workers
and unskilled
the
professional
move
laborers have few opportunities to
-
41
-
into
occupations.
that
was
recruitment
into
out
amongst
be
the
result
is
first
The
tendencie s.
spread
could
This
occupatio ns.
workers
in the managerial occupation, it appears
1976 and 1983,
Between
management
positions.
two
reinforcing
to hire
experienced
of
preference
a
second
The
of
range
wide
a
to
is
means
This
managers familiar within the area they will manage.
hire
sales wor kers become who become managers usually manage sales
vacancies were filled by those outside the labor market,
workers, and 16.6%
make
by
craft workers.
by
2.4%
workers,
3.2%
workers, 0.4%
operative
by
labor
unskilled
workers, 1.9% by
transport
19.5% by
The remaining occupations
a small percentage of the recruits:
up
the
sales workers, 20.4% by clerical
by
professio nal workers, 14.9%
of
14.0%
The recruitment pattern is:
sales rel ated areas.
in
by farm owners, 0.0% by farm labor workers, 0.3% by
protective service workers, and
would indicate that
service
6.5%
by
service workers.
This
workers and unskilled laborers have
few opportunities to move into managerial occupations.
Between 1976 and 1983, most recruits to
came
from
two
groups.
of
31.6%
the
The
into a
occupation
sales
the vacancies in the
occupation were filled by those from outside
and 27.5% by managerial workers.
sales
only
the
labor
occupation
market,
to
fall
middle
category is clerical workers who made up 18.2% of
The
remaining occupations were all recruited in small
recruits.
percentages to fill the available vacancies.
-
42
-
1983, 40.1%
of
vacancies were filled by those from outside the labor market.
No
In
occupations,
clerical
vacancies available.
filling- this
to
other category came close
and
1976
between
percentage
of
the
made
up
of:
group
There is a low-middle
11.0% by professional workers, 15.9% by managerial workers, 10.1%
by
4.2%
was:
occupations
workers,
1.2%
by
workers,
0.2%
by farm owners, 0.1% by farm labor
0.1%
service
by protective
the
outside
of those from
group of
operative
labor
unskilled
by
workers, 1.8%
transport
by
5.8%
workers,
craft
low
A
by sales workers, and 9.5% by service workers.
and
workers,
Again the high recruitment
workers.
might be due to the high
occupation
percent of women working in clerical occupations.
In craft occupations, between 1976 and 1983, the highest
group
A second
large
of recruits came from operative workers (29.1%).
group of recruits came from outside the labor
professional workers (8.8%),
workers
(5.5%)
low-middle range.
other occupations.
transport
The
workers
workers
clerical
and
laborers
Unskilled
(15.8%).
workers
managerial
remaining
market (17.4%) and
(5.8%),
(6.2%),
(8.4%),
service
represent
a
vacancies were filled from the
In this more detailed matrix it is clear that
there is some crossover between operative and craft workers.
1976
In operative occupations, between
group of
recruits
came
from two places:
and
1983, the largest
30.6% of its vacancies
were filled by those outside the labor market, and 29.0% by craft
workers.
A
middle group of recruits was made up
-
43
-
of
unskilled
labor
The.high recruitment
occupations filled the remaining vacancies.
of craft
an indication of a crossover between craft
is
workers
other
The
(8.0%).
(11.9%) and service workers
workers
and operative occupations.
Operative
recruit
occupations
more
from outside the labor market than do craft occupations.
In transport occupations, between 1976 and 1983, 20.1%
of
its
vacancies were filled by those outside the labor market, 18.6% by
operatives,
18.2% by craft workers.
and
represents
This
A second group of
majority of recruits to transport occupations.
and managerial
recruits came from unskilled labor workers (10.5%)
workers
managerial
of
The relatively high percentage
(9.1%).
the
recruits seems counterintuitive.
In
1983,
and
labor occupations, between 1976
unskilled
the
largest group of recruits came from operative workers (26.2%) and
from outside
the
recruitment
came
(7.8%),
labor
come
to
It appears as though
labor
unskilled
occupations
of
workers
recruitments
Remaining
(7.6%).
workers
came from the other occupations.
workers
clerical
from craft workers (16.7%),
and transport
level
middle
A
(22.3%).
market
in
operative
search
of
temporary employment.
1983,
In farming, between 1968 and
44.7%
of
vacancies
were
comprises by far
filled by those outside the labor market.
This
the largest group of recruits to farming.
A second group is made
up of farm labor workers (15.3%),
and by managerial workers
by professional workers (12.2%)
The
(10.3%).
-
44
-
other occupations have
of the remaining recruitment.
shares
small
of the outside recruitment to
most
matrix, it appears as though
8
by
8
As in the
farming would be inheritors of farms.
28.3% of vacancies in the
and 1983,
1976
Between
occupation were filled by those outside the labor market,
16.3% by unskilled labor
craft workers,
farm owners.
workers
(8.8%)
occupations
were
and
transport workers
by
14.7%
and
from
operative
(8.2%).
The
remaining
the
of
shares
small
fill
to
recruited
17.4 by
came
recruits
of
A middle group
workers,
labor
farm
opportunities available in farm labor.
In
protective
service
vacancies were filled by
outside the labor market.
and
1983,
28.7%
workers
and
26.0%
by
a
majority
This constitutes
in this occupation, no other
those
the
of
of workers even filled 10.0%
group
as
It appears
of
Of the opportunities available
recruits to protective services.
of the vacancies.
1976
between
services,
though there is some crossover
between service occupations and protective service occupations.
of
In service, between 1976 and
1983,
58.1%
the
labor
market.
filled by
those
outside
occupation to approach
clerical (9.2%).
This
a
share
10.0%
of
the
were
vacancies
The
only
other
recruitment
is
may be due to the fact that some service
occupations have a lot of clerical tasks.
-
45
-
A Look at the Labor and Service Workers
In
recruitment
matrix it is possible to observe the
this
and service.
observe
the
in
the
workers
unskilled labor and service
for
opportunities
to
is
second
The
pattern of workers recruited.
occupational
usual
the
observe
One method is to
labor
categories
labor in two ways relative to the occupational
of
other occupations.
the table below are
workers entering
fills 22.3%
of
the
its
(1.2%),
Labor
from operative (26.2%),
comes
recruits
(7.8%),
transport
Recruits
from
clerical
and service (7.9%) are
what
might be called the middle.
and craft
Recruits
occupation.
vacancies from outside of the labor market.
The greatest percent of
(7.6%),
labor
unskilled
for
occupation
by
rates
recruitment
the
into
1976 to 1983. In
for the years
All the calculations I made are
(16.7%).
from
professional
farm owners (2.4%),
(1.0%),
labor
farm
managerial
(2.6%),
(4.4%),
and
sales
protective
service (0%) were consistantly low.
Private household workers made up such a
small
proportion
the labor market that they had no effect upon most of
occupational groups.
-
46
-
the
of
other
Recruits to Unskilled Labor
Table 5:
Year
Outs.
Prof.
Mngr.
Sales
Cler.
Craft
Oper.
Trans.
Farm
Farm
La bor
Prot.
Serv.
# Vacancies
in Unskilled Labr
22.3%
0.2%
0.2%
1.8%
5.6%
16.9%
26.4%
6.4%
5.4%
0.8%
0.0%
12.2%
2,806
23.9%
5.7%
1.4%
0.7%
5.0%
12.8%
35.6% 10.1%
1.6%
1.0%
0.0%
2.7%
2.192
15.1%
0.0%
3.1%
0.1%
7.7%
19.8%
26.1%
12.1%
0.9%
5.7%
0.0%
9.3%
2,119
31.3%
0.2%
0.0%
0.0%
5.3%
20.5%
23.A%
9.3%
3.9%
2.8%
0.0%
3.2%
1.875
26.4%
0.0%
7.0%
0.1%
10.9%
12.3%
23.4%
5.8%
3.1%
1.8%
0.0%
4.
16.7%
0.0%
11.3%
0.1%
7.1%
16.6%
24.2%
2.9%
0.0%
5.8%
0.4%
14.9%
19.6%
1.2%
7.8%
5.7%
14.0%
19.0%
23.0%
'5.6%
0.0%
1.2%
0.0%
2.7%
Out of all
5.7% were in
opportunities
1
708
1.844
workers
N, during this
Using the weighted
unskilled labor.
to
left
the
vacancies,
fill
from outside the labor force and 11,901 from
15,966
2.328
the vacancies in the labor market from 1976 to 1983,
period 14,872 workers were recruited
while
1%
other
occupation,
-1,094 from job changes,
from/to
-548
and
market.
-
47
-
occupations;
for
12,447
(jobs), and 3,519 left the workforce.
outside
2,971
The
the
new
net
is
labor
Table 6. 1976-83 Net Arrivals & Departures From:
Unskilled Labor
ARRIVALS
DEPARTURES
NET
11,901
12,447
-538
2,971
3,519
-548
14,872
15,966
-1,094
to/from other
occupations
to/from
outside
total moves
Labor market arrivals by year into unskilled labor range from a
from a high of 797 in 1981 to a low of
from the occupation range
335 in
of 239 in 1979. Labor market exits
low
high of 627 in 1976 to a
1977. Recruitment of workers
from
other
fill vacancies ranges from a high of 2,179 to
a
occupations
low
of
to
1,422.
Departures to other occupations by unskilled workers range from a
high
of
low
a
to
2,136,
of
1,388.
breakdown of Table 6 and the labor market
(For a complete
yearly
moves described above,
see appendix).
All the calculations I made are
the
the table below are
workers entering into
fills 58.1%
of
its
the
for the years 1976 to 1983. In
recruitment
unskilled
rates
by
occupation
labor occupation.
for
Service
vacancies from outside of the labor market.
The greatest percent of occupational recruits comes from clerical
(9.0%).
Recruits
from all other occupations
low.
-
48
-
are
consistently
Table 7:
Year
Recruits to Service
Outs.
Prof.
Mngr.
Sales
Cler.
Craft
Oper.
Trans.
Farm
Farm
Labor
Prot.
Unisk.
Labor
# Vacancies
in Service
78.3%
3.3%
1.5%
0.9%
5.0%
5.1%
1.5%
2.5%
0.0%
0.0%
0.2%
1.9%
7,004
48.7%
10.2%
11.2%
1.9%
7.8%
7.6%
7.7%
0.1%
0.0%
1.0%
0.0%
3.5%
3,841
51.3%
7.1%
4.1%
0.0%
14.3%
7.3%
9.6%
0.0%
0.0%
0.0%
4.1%
I.
5,.667
54.5%
7.7%
5.4%
1.8%
8.9%
3.6%
7.8%
2.7%
0.0%
0.2%
4.0%
3. 47
5,72 1
55.7%
7.9%
8.5%
1.8%
9.8%
7.8%
3.3%
2.0%
0.0%
0.1%
(.5%
2.47
1, 925
58.7%
10.9%
4.0%
1.3%
7.1%
4.8%
7.0%
2.9%
0.0%
0.0%
0.6%
2.9%
3,621
52.8%
6.6%
5.4%
1.9%
13.7%
7.4%
5.9%
2.8%
0.0%
0.0%
1.0%
2.
%
3.718
Of all the
vacancies
in
the
labor market from 1969 to 1983,
13.5% were in the service occupations.
during
this
period
35,517
workers
Using
were
the
recruited
vacancies, 20,645 from outside the labor force
other occupations;
19,352
for
workforce.
new
while
40,634
opportunities
workers
(jobs),
weighted
left
and
The net is -5,117, -4,480 from job
and
N,
to
fill
14,872
from
the occupation,
21,282
left
changes, and -637
from/to outside the labor market.
Table 8. 1976-83 Net Arrivals & Departures From:
Service Workers
ARRIVALS
DEPARTURES
to/from other
occupations
14,872
19,352
to/from
outside
20,645
21,282
-637
total moves
35,517
40,634
-5,117
-
49
-
the
NET
-4,480
Labor market arrivals by
from
year
into
service occupations range
high of 7,004 in 1976 to a low of 3,621 in
a
high
market exits from the occupation range from a
Labor
1981.
8,673 in
of
1978 to a low of 3,718 in 1982. Recruitment of workers from other
2,994
of
occupations to fill vacancies ranges from a high
workers
low of 1,495. Departures to other occupations by service
range
complete
a high of 3,666, to a low of 1,640. (For a
from
yearly breakdown of Table 8 and the labor
to a
market moves described
above, see appendix).
Descriptive Statistics
in
65.6% of the men
work.
PSID survey, and 48.9% of the women,
the
the inclusion
is
of
people
women
$2814.
but this is in part due to
low
seem
These income figures might
for women
for
mean income for men is $7729 and
The
who are not working.
The lower
figure
mostly due to the prevalance of women working part
time.
The occupational breakdown
higher
also partly responsible for the
is
The
income of the male labor force.
mean
population is
concentrated in the higher paying occupations.
In
group
is
the population of working men the largest occupational
managerial
(17.1%).
male
working
(19.6%),
followed by craft (18.6%),
paying
low
Few men are found in
none in private household service, and 0.9%
of
Working women make up 48.9%
-
50
the
-
and
professional
occupations,
almost
in service.
female
population.
The
can
largest concentration of working women
(23.9%),
clerical
Surprisingly,
(14.6%),
19.2%
of
occupations.
Women make up
managerial,
protective
occupations.
Women make up
and
working
a
be
(16.5%).
are
operative
in
disproportionately
a
sales
professional
women
farm,
service,
in
found
small share of
and
labor
disproportionatly
craft
large share of
private household service, service, and sales occupations.
Women
do not constitute a majority of the clerical occupations.
Women
might be disproportionately represented
as secretaries but there
are many male dominated occupations that fall under
clerical.
Some
of
these
are: mail handlers,
the
mail
heading
carriers,
dispatchers, estimators, real estate appraisers, etc.
The only occupation in which median income for
than for men is in
farm owners.
This is probably due to the low
number of women in this category in
distortion is
due
to
women is higher
the PSID survey.
weighting.
A possible
The highest mean salary is in
the protective service occupation and the lowest in farm owner.
For education the categories are 0
years
plus
non
academic
schooling (any college
to 11 years,
training, and 13
through
or
12 years or 12
more
advanced degrees).
population 35.5% are in the lowest category, 33.1%
category, and 30.6% in the
highest
category.
For
years
of
For the male
in the middle
the
female
population, 50.1% are in the lowest category, 34.3% in the middle
category, and 15.2% in the highest category.
-
51
-
CONCLUSION
One
of
the
clearest
conclusions
difficulty of doing research in new
from
this
areas.
study
Firstly,
little literature or past written history to refer to.
the computer and
experiential
or is something to be
rewards
for
intellectual tools,
how
there
the
is
Secondly,
knowledge is either not available
learned
learning
is
in small steps.
to
apply
Nonetheless, the
the methods, and the
new
make new research very exciting and valuable
both professionally and personally.
For
all
the
reasons
stated
above
I
could
investigate in one semester all the areas of
That is, to conclude a full
full
analysis
occupation.
of
the
However,
not
personal
hope
to
interest.
test of the opportunity model, and a
underlying
patterns
of recruitment
by
I can conclude that this methodology could
provide a rich area for future research.
I
have
matrix.
made
The
some
Q
initial
matrix
is
tests on the stability of
the
multiplier, where the stability
observed.
the
Without
yearly
aggregate
Q
essential
of
part
recruitment
of
the
the
Q
job
patterns can be
doing any statistical tests, and by observing
recruitment
matrix,
there
patterns
seems
and comparing them
to
be
a
with
the
relative pattern of
stability in the recruitment patterns.
Certain trends
are clear from observing the normal recruitment
patterns of occupational labor.
-
One is that all the occupations,
52
-
except for craft and managerial, fill a substantial percentage of
their
vacancies
with
recruits
from outside the
Additionally, according to the data from the
PSID
labor
force.
survey,
both
unskilled labor and service occupations are suffering a net
loss
of workers over the period being studied.
their way out of these occupations into
More people are making
other areas of the labor
market.
Further pursuit of
research
analysis would require the
the
Q
the
Q
matrix.
test
this
development
of
or
on
the
compares
matrix.
If the
than it
would
whole is
of
labor
market
statistical tests on
individual
A statistical test on the whole
that
type
It is possible to do statistical tests either on
matrix as whole
matrix.
in
each
stable.
were
possible
Testing
matrix
to
in
would
year's differences with
differences
be
Q
cells
Q
the
be the
Q
aggregate
not statistically signifigant
conclude that the
Q
matrix as a
of the cells would allow analysis of
stability of recruitment for each of the occupations
within
the
labor market.
The
Q
matrix I analysed is
the matrix
only
the
the one with a zero diagonal.
moves into an occupation are analyxed, not
the moves within the occupation.
recruitment to the
occupation
This allows the observation
from outside the occupation.
figures given are probabilities based on
matrix for every 1000 new recruits to
to come from
outside
With
the
labor
-
53
In the
The
13 by 13
Q
labor 447 will be expected
market,
-
one.
of
122- from
managerial
workers,
103 from
matrices can be read this
Q
so
Q.
by dividing the row sum by the
matrix cell.
After statistical testing for stability, the next
calculate the multiplier.
number of moves made
Q)~1 .
-
(I
in
taking
step
is
the
subtracting
the
to
inverse
Q
the
of
The mean passage time, or average
system
the
by
done
This is
matrix from the inverse matrix and
resulting matrix,
All the
on.
The actual cell numbers (except
way.
for the diagonal) can be obtained
percentage in the
and
workers
profesional
for
each
vacancy, can be
vector
calculated by post multiplying the multiplier by a column
of l's.
By multiplying the multiplier above by the demand
for labor by
we get M(t),
the number of
occupation, which is
a
row
If we use the aggregate
opportunities created.
this calculation then
vector,
we
get
estimated M(t).
an
collect information on the exact
Q
matrix
to
make
If we wait to
matrix (this could take years)
then it would be possible to get the actual M(t).
From this it should be clear that the more stable the
the better are the estimates made of M(t),
opportunities generated in the system.
to run tests using estimated D(t),
it could
be
matrix,
number of
total
the
Q
It would also be possible
This means
labor.
demand for
used as a policy analysis tool.
When attempting to
develop an area economically, the projected mix of new industries
would have an estimated demand for
estimated by occupation.
labor
D(t)
This
-
54
-
could
which
be
could
run
also
be
through
an
Q
opportunity model with an aggregate
effect upon the labor market.
regional
Q
matrix to get a look at the
This would require
matrices, multipliers and
estimation
The
times.
passage
mean
of
main idea is to test if regions have the labor force necessary to
the projected future economy or if it will be
in
fill positions
necessary to attract a different labor force or provide
training
for those already present.
It
is
investigation
into
how
differs
opportunity
demographic groups that explains where stability in the
make
This model may
comes from.
accurate
mobility but it can give nothing other than
accross
Q
matrix
of
estimates
labor
an aggregate picture
that leaves unobserved the underlying processes.
In order to perform this type of analysis
information
collect
Q
element is the
look
at
more
on the mobility of
matrix.
But with most surveys it is possible to
than
information
This
just
permits
a
more
differentiate
about who stays and who
Linked
with
look
demographic
at
different
areas
research.
-
55
-
of
and
these characteristics
occupational
the above analysis this more detailed look
labor force can provide large
career
An interesting analysis
to
according
leaves
of
but also how education
age can affect the recruitment of labor.
to
Therefore
occupation.
detailed
opportunities of women, blacks, etc.,
would be
essential
The
labor.
seperate analysis can be made on different types
groups.
necessary to
is
it
untested
labor
groups.
at
the
market
In this study I have used the
that it follows
study.
This
compare
their
mobility.
the
same
could
PSID survey.
families
permit
the
The advantages are
for the whole period of the
extraction
careeers with the aggregate
of individuals
estimates
of
The drawback of the study is that it has not
to
labor
drawn
a
fresh sample of families since 1968. Therefore it might no longer
be
representative
of
the
U.S.
population.
Elements of
population that are difficult to follow have
probably
sample, and there can be no
immigrants.
sample
of
groups may be small but we have lost
them.
The survey used
weighting
new
any
to
left
opportunity
make
up
for
the
the
These
to follow
some of its
deficiencies, however relatively low numbers of people in some of
the occupational cells could lead to distortions in
when people move.
When one
cell this appears as many
person
the Q matrix
moves from a highly weighted
moves.
The
more
stationary
people
remain in the PSID survey the more accurate is the weight applied
to them.
In
conclusion,
this
study
is
presented
as
an
initial
undertaking in the pursuit of new methodological tools which
improve our understanding of the
labor
improve policy making.
-
56
-
can
market, and of models to
APPENDIX
-57-
Yearly figures for unskilled and service workers
Unskilled Labor
in movers
year
outside
1976
1977
1978
1979
1980
1981
1982
total
627
523
320
239
614
286
362
2971
out movers
inside
Total
2179
1669
1799
1714
1422
1482
1636
11901
2800
2192
1875
1875
2328
1708
1844
14872
outside inside
443
335
484,
460
364
797
636
3519
1787
2136
2112
1643
1845
1636
1388
12447
total
2230
2471
2496
2103
2209
2433
2024
15966
Service Workers
in movers
year
1976
1977
1978
1979
1980
1981
1982
total
out movers
outside
inside
total
outside
inside
total
5485
2847
2908
3117
2188
2126
1974
20645
1519
2994
2759
2604
1737
1495
1764
14872
7004
5841
5667
5721
3925
3621
3738
35517
2490
3574
5131
2723
2664
2727
1973
21282
3079
3666
3542
1640
3356
2324
1745
19352
5569
7240
8673
4363
6020
5051
3718
40634
-
58
-
1969-70,
IRAASFCSL
Frct
Prot
Mana
baC1
Craft
up1 I
LSF1
Farr
Aisc
Outs
3L7
2331
2Lb
232
348
0
335
131 U
Q
Frot
Mana
.207
SaC1
Craft
CpIr
LSk i
Farff
Aisc
.101
.028
.019
.02(o
mana
SatI
'125
0
177
1100
u32
329
1862
232o
084
0
doo
1320
5b2
104
4u3
510 0
Craft
U ir
Lsv1
1-441
1 iU
627
1725
287
0
c32
d40'
1035
28 0
103
444
2919
0
137
164
5473
Craft
p'ir
3971
0
3b17
34
48
14
781
farm
0
148
92
44
203
25b
0
U
422
Misc cuts
337
60
550
102
1089
792
0
0
58C
2746
653
6777
575
2712
6793
74
580
0
Matrix 1969-70
0.110
0.
U.12
0. 158
(.052
0.024
0.437
0.
0.207
0.
0.117
U.10d
0.022
0. 192
0.021
0.
0.2
0.043
0.148
00
0.24 b
0.008
u.o5
0.008
0.021
0.073
0. 10
0 *b38
0.
0 /
0.046
LSF1
I'armii
0.095
(.U96
0.070
0.0t0
0.238
0.
0.194
0.1 UU
0.
0.017
0.007
0.006
0.011
0.019
0.
0.
hisc
v.051
0.007
0.041
0.014
0.089
0.058
U.
U.
Out
0.416
0.075
0.503
0.07i8
0.221
0.499
0. 105
0.357
rowsuff
6577
8715
13481
7376
1221
13616
705
1624
HESULI OF E.ACD
Cut
Prof
32813
1151
1503
6
257
1585
27 135
2622
1949
392
1156
5b1
0
305
854
1839
414
59
1047
1190
1970-71
.ana
SaCI
1817
2597
40120
245
6b4
714
2b49b
3664
121
1091
2397
56
594
958
6559
Craft
4
47
573
50
88
724
592
1007
3815
2b097
2792
41
225
1566
14d9
UpTr
LSF1
430
653
0
195
92
28
144
88
155
216
1218
670
2489
26091
72
118
1012
5794
Farm
Misc
2059
462
908
163
1199
b695
1455
44
3802
b68
b109
1416
0
265
51
430
1348
8370
2057 118857
223
95
10
102
5703
0
86
184
Outs
4
3102
7 51
41
THANSPUSE
rof
Mana
0
1585
1151
0
1817
2597
6b4
47
653
92
28
908
245
4
430
0
195
2059
SatI
1 503
2 622
U
714
1 007
1 218
144
155
5 b95
Craft
Upir
LSF1
6
1949
721
0
3815
670
88
21 b
1455
257
392
1091
38b4
0
2-%89
10
223
3602
1156
561
2391
!373
2792
0
102
913
6109
QMatrix 1970-71
Prot
Mana
Prof
Mana
SaLl
Craft
CpIr
LSF 1
farm
Misc
0.
0.181
0.115
0.036
0.000
0.038
0.
0.204
Farm
Misc
Cuts
0
414
56
50
47
72
0
4
265
305
59
594
88
225
118
0
1839
1190
6559
592
1489
5794
184
41
1839
0.
430
SaC1
Craft
CpIr
0.185
0.242
0.
U. 104
0.098
0.005
0.057
0.148
0.029
0.299
0.001
0.222
0.046
0.
0.405
0.059
0.142
0.226
0.041
0.045
0.0b9
0.569
0.
0.217
0.016
0.233
0.
0.105
0.107
0. Ob
0.232
0. 162
*
label
Prof
mana
SaC1
Craft
OpIr
LSFi
Far rr
Misc
Outs
LSEl
0.189
0.064
0.151
0.084
0.296
0.
0.1b4
0.099
Farm
Misc
0.
0.047
.0.U3
0.007
0.005
0.00b
0.
0.0U4
U.U49
0.007
0.036
0.013
0.024
0.010
U.
0.
Cuts
0.041
0.136
0.414
0.087
0.158
0.506
0.297
0.043
rowsum
6217
8772
15832
b790
9426
11444
620
957
kESUL1 0i
Cut
E.AUL
Prot
120899
2025
100)
7273
1031
3433
495b
105
110
1944
1971-72
Mana
2429
33799
544
2289
238
218$
255
0
234
1349
670
557
29380
1905
562
b52
431
184
64
331
Missing
SaCI
Craft
Cpir
LSEI
farm
Misc
cata
4195
1514
1537
41310
689
1970
0
913
776
3061
24981
3170
96
369
2012
5750
848
2454
1256
1118
2566
25187
70
74
1529
488
0
234
0
0
390
0
140
358
119
314
300
10
3024
1270
1985
317
ibbb
36b
229
279b
1017
407
714
185
25025
2733
1731
0
43
1587
Craft
OpTr
LSF1
Earfr
Misc
1031
23b
502
3433
218
b52
968
2733
0
2566
0
314
4955
255
431
18b
1731
3170
0
92
300
105
0
184
368
0
96
70
0
10
110
234
64
229
43
369
74
46
0
0
92
5209
4b
92
393
1608
97
160
457
0
8
8867
IpAhSPOSE
Out
Prof
0
2429
670
4195
1017
1970
5750
488
390
Mana
2025
0
557
1512
407
0
848
0
U
1007
544
0
1537
714
913
254
234
140
RESULI Ul. MUCIF Y
SaC1
7273
2289
1905
0
185
776
1256
0
356
bb9
0
3061
1118
0
119
Prot
Mana
SaCl
0.391
0.133
0.
0.110
0.133
0.059
0.
0.071
0.
0.
0.087
0.
0.135
0. 10
0.368
0.148
0.190
0.481
0.507
0.239
Out
Prof
Mana
SaCi
Craft
Upir
LSF I
F arm
Misc
Q Matrix 1971-72
Out
0.369
row
label
0.021
0.272
0.085
Cratt
Lplr
LSF I
k arm
Misc
0.038
0.116
0.035
0.130
0.041
0.0bb
0.
0.379
(0.038
U.
0.03o
0.013
0.060
0.04b0
0.164
0.032
0.027
0.074
0.105
0.
0.219
0.
0.297
0.090
0.
0.073
0.400
0.
0.215
0.
0.192
0.253
0.306
0.
U.
0.0u5
0.107
0.1b4
0.
0.009
0.006
label
Prof
Mana
0.020
SaCi
0.U006
Craft
0.036
upir
0.00/1 LSE I
Farfr
0.053
Misc
U.
ro5sun6207
5025
11364
6830
10355
11936
860
1631,
RESUL1
Lk
L.ACL
1972-73
SaC1
Cratt
Cp1r
LSF1
idrm, Misc
453
3574
4157
260
111
2040
14s5
503
391
2629
1010
24746
2545
594
378
k62
0
29
748
327
323
208
1042
560
859
418
1b09
969
1357
19606
76
40
155
80
0
0
247
4491
0
267
Prot
Cut
1171.7
1509
mana
1511
29b 0
640
108 ;
485
214
10.7
5b4
806
2187
5571
476
833
U
396
91
24,5
1859
72
HESUL1 Or 6.ALL
j/0
20200
2510
d19
92
105
1160
425
1344
1 162
4u
bb4
2e46
4112
22679
3095
60
43
204
2155
1564
USING mrzb.72 AMC tzb.72
PAGE
4
2b2
81
109
401
0
3032
833
Missing
Cata
2464
423
136
1115
67
243
676
0
46
7516
2
IHANbFOUS
row
Cut
irot
Q. 1509
1517
353
3573
583
2629
4157
260
111
0
1010
2020
391
327
323
40
20d
RESULI U
hana
1047J
60
0
1485
560
418
135
saC1
Cratt
bOb
552.
10
5
LSE1
farn
Misc
21871
214
5571
833
282
1342
1102
247
476
0
0
40
92
t00
76
0
401
0
398
97
29
583
105
204
43
0
0
2545
0
425
370
U
1609
4112
99
80
0
252
MOLi
upir
819
0
1357
0
109
309 5
U
mana
SaCl
Crait
Cpr
0.311
0.
0.131
0.222
0.099
0.068
0.336
0.098
0.2 ki
0.464
0.332
0.095
0.194
0.
0.139
0.176
0 .047
U. 04
0. 196
0 .U3
0.4i90
0.119
0.035
0.190
0.06
0.027
0.03b
0 .17
c
Out
Prot
Mana
SaC1
Craft
OpTr
LSFI
Farm
Misc
Q Matrix 1972-73
Prot
Cut
label
U.
u.002
0.0/2
LSF 1
Farr
0.043
0.171
0.072
0.054
0.109
0.
0.
0.0o0
0.010
0.179
0.102
0. 1b6
U.
0.126
0.424
0.
0.151
U.
0.21 a
0
0.093
0.
0.341
.0U8
0.13d
0.261
0.
0.315
0.343
Misc
0.001
0.oO
0.
0.
*
lakel
rowsur
0.015
Prot
0.002
0.059
0.017
Mana
SaC1
Craft
0.013
CpIr
0.006
0.
LS1
karn
5191
10630
5912
11846
8952
782
0.
Misc
116
4871
HtSUL
1973-74
OF L.ALU
73
Missing
Cut
119473
172o
10*78
4355
517
2779
5036
541
37o
2358
saCi
Prof
hana
1306
3002b
1007
850
979
22776
1232
1818
533
354
513
0
340
1610
4177
193u
3Ai94q
62o
109b
951
49
783
514
0
57
559
Craft
Cplr
LSel
Farm
490
192
1208
437
1431
26o
271
140
2 0i175
1561
23039
1436
32
12 1
2037
4530
749
51 o
18 t6
722
3145
19370
141
22)
1761
3959
678
64
36
1b3
284v
1310
Misc
428
362
101
595
468
521
195
0
3629
740
0
0
0
36
101
40
4335
1b
80
Cata
1900
314
162
904
111
402
801
156
9
7170
'3
R S U LI
01
130 6
b5
417
49 0
143 I0
453 1C
14
42 8
7zb
0
979
bIb
492
26b
749
0
302
ht.0LL
Ut*
Cut
0.247
0.160
0.425
0.00b6
0.222
0.378
0.42U
0.160
1078
1007
U
1930
120b
271
b7o
0
101
M0L.IE
4355
1232
1818
0
437
1310
1866
0
0
1561
595
722
36b
'on
354
514
1096
3959
u
3145
101
521
5036
513
294
951
541
67d
1436
U
40
195
0
49
64
32
141
0
376
340
57
163
36
121
227
1b
C
Out
Prot
Mana
Craft
Upir
LSF I
Farm
Misc
Q Matrix 1973-74
SaCk
Cratt
0.190
0.
0.196
0.1b4
0.04d
0 .0 48
0.233
0.343
0. 10
0.144
U.
0. 066
U.059
0.203
0.156
0.
u.248
0.060
G.
U.
U . 10 d
U.u37
0.222
0.1/ 5
Prct
Mand
0.
0.184
0.083
0.041
0.062
0.
0 .135
510
533
783
Cpilr
0.0t6
0.097
0.111
0.b37
0.
0.263
0.303
0.195
LEF1
0.097
0.055
u.096
0.092
0.223
0.
0. 120
0.073
Farm
0.
0.
0.004
0.0ob
0.004
0.011
U.
0.
Misc
label
0.064
0.010
0.016
0.004
0.018
0.018
0.04d
Prot
mana
SaCl
Craft
CpTr
LSFI
Farff
0.
Misc
rowsum
5285
5295
9810
7364
6428
11956
333
2b70
HESULI UF E.A00
Cut
1450
3u78u
14 4 7
5166
512
2565
5bb 4
693
1859
74
Craft
CpIr
LSF1
Earm
503
322
1231
234
4131
730
k27
801
605
1935
20413
78
243
1500
54
23
84
12b3
111
i
1571
3 o210
2 2798
1/31
705
b52
421
762
635
274
329
17
427
1561
2560
saCi
Mdna
Prof
122198
1974-75
407
178
19793
1933
1215
217
3d5
d09
4eb0 2
133b
737
2423
21679
203w
162
412
2021
SaC1
Cratt
LpIr
LSF1
2500
2505
2/4
3i9
773
1044
b
4e
0
0
0
65
7b
123
3883
0
252
Misc
333
54
33213
22
3b2
195
0
3907
666
Missirg
Cata
2809
295
173
1017
144
154
593
0
100
8022
1i6AhbyGSL
Lut
Erot
0
1450
512
32d3
503
1231
4141
5,
333
1447
FkSUL
ti
uI
row
Mana
v9 6
384
1bU2
L)
1131
322
234
731
2s
bi
Uut
035
762
1731
7us
5 bU
U
S/5
421
1215
2423
19.33
U
737
,0u
S13
2e7
b4
JJ
2O
US
24
21-4
0jiY
Q
Prot
7o
352
saC1
mand
0.
tJ.*089
0.1/l
0.091
U.2 b4
0.
0.30/
0.351
0.094
0. 10
0.4727
0.127
0.2 /
0.121
0.060
0.031
U.163
0.
O .054
0.u9e
0..
0.044
17
217
2C 3b
Q
123
195
7b
0
U
U
Out
Prot
Mana
SacI
Craft
Op1r
LSFI
Farfr
Misc
Matrix 1974-75
0.33
U.0b3
693
427
0
555
385
412
243
0
1b71
b52
7 73
152/
4 i
label
fdrn
Cratt
LE1 1
Cpir
0.148
u.122
0.104
0.0o4
0.115
0.0
0.324
0.363
0.
0.076
0.074
0.111
0.228
0.271
0.091
0.421
G0.
o.157)
0 .05/
'.
0.17 b
li.*047
0.11/
0.029
0.087
0.036
0.025
0.082
U .
0.292
b9
0. 1b2
Farm
0.003
0.
0.005
0.040
0.041
0.u0b
0.
0.
Aisc
0.099
0.
0.059
latel
Erot
Mana
0.054
0.021
SaC1
Cratt
Oplr
LSF1
U.
Farm.
0.
Misc
0.072
rowsur
4278
5t23
9340
5320
7512
8750
425
1202
Ht%,OL1
Ck
L.ALD
1975-76
75
Missing
Cut
125052
17 b7
Cratt
Lp'r
LSF1
Farn
ba0
2102
35u
484
4463
769
176
17
32471
4/U
842
291
337
861
2259
909
131
15813
130
3b10
Prot
Mana
.aCi
1675
2bbu I
445
604
249
4763
131b
173
121
35'1l
lob
571I
2 b1
43/19
4db6
ob9
9.1 b
u
/.,
1042
12b
1534
3d9e
1'u
17/
142
2035
1312
Cratt
Lpjr
111
4541
1708
338
111
0
0
2355
1193
026
50
Cata
Misc
1122
124
567
565
0
56
7321
0
2031
Q
214
24U
1.iG3
b53
0
1623
194
row
G
6aCi
Mara
i-rot
Cut
.1391
2b$ u
0
2499
7b9
lbb
7
64
4783
13b0
2102
445
35u
291
44 o3
33~
17
0U
1Jt
kitSLti Lk
938
i11
.cub3
,Cb4
560
2p
i rot
69
u
u
0
Cut
2309
0
u
130
O
563
t59
2355
842
0
I.9
MLLlti
u
511
628
1193
b9b
60
2031
lob
137 0
Q
LSt1
u
barn
Fisc
label
218
2267
128
Uut
79
0
142
17
164b
240
214
853
0
U
0)
u
SaCl,
ia
Cratt
Lpir
0.
U .095
0.293
0.u25
v .045
0.j09
0.4o3
0
0.
0.210
0.334
0.
0.02b
0 .uto9
U
*Ubb
0.029
0.1db
0.044
0.
0.253
C .4i5
0
307
0.C42
.026
-. U
mana
SaCI
Cratt
Upir
LSF 1
e'arrr
Misc
Matrix 1975-76
O.400
0.062
* .7 7
0.067
Frof
U I*
U .032
u.
v
0.271
0.084
O .Uob
u0.00
2
0.
0.225
0.
tarn,
LSE1
0.122
0.077
U.12 U
0.104
0.244
0.
0.28 b
lacel
rowsun
U.
0.02 I
Prot
0.000
0.009
10237
454
0.02
U .003
Mana
SaC1
Craft
UpIr
LSfl1
-. ;
U.
U.
- .ut
Farfr
0.U14
0.21o
0.02o
Misc
4b79
6079
9895
854 1
8306
0
1976-77
Prot
,ana
22b2
It,19
2 ,1'1
1218
1u31
Lut
131983
1375
1361
3216
z/lb
1
0
815b
24
151i
- EbULL1
Uk
170
v22b
1770U
3b)3 I
22838
2870
1028
563
791
0
0
607
284
23o3
3260
241
baC1
Cp'ir
LSEi
farm
1717
149 -
3097
115
318
27b
2169
20b!2
1197
0
bbb2
236
158
630
962
1331
18555
272
395
31
66
0
41
0
195
U
0
U
106
91
1345
20o49
24d2
100)
b93
41
92b9
U
1428
1dc
US1NG inZb.7o Ai'.ItZc..6
E.ACi
Misc
Cratt
PAGE
Missing
cata
2277
169
8b
389
119
221
864
0
0
6672
3584
2
IbA5USPOt
Lut
krot
0
1519
Fana
1375
0
1031
SaC1
34 1I
1 7u
0
±e4
703
0o
V
U
5d
1770
1' 17I
309'/
3S5
0
115
230
31
0
0.399
V0194
U 1 541
0.214
jISb
630
0
0
tb
U
O .657
0.542
0.
0.132
U .215
0.0 bt.
U.1job
0.11$
0.107
0. U44
U.0 lI
0.02 1
U 042
2bb5
U
9Q4
U
upIr
LSfl
Farn
2363
3260
bib
247
24
191
U
53
IU 7
2462
V
1331
0
1197
0
195
272
0
0
U
0
693
92t)
41
Misc
row
label
Out
Prof
Mana
SaC 1
Crdtt
OpTr
LSF1
Va rm
Misc
Q Matrix 1975-77
Maria
prot
0.016b
1 02b
276
It
Cut
Craft
0.
(. 090
6acl
V .3671
U.
.U /2
0. U1
U.01U
U.
Craft
00050
0.131
u.U23
0.
0. 30
0.091
0 . 05 b
Lo
r
0.049
0.094
0.309
0.
0.127
0.
LS I
Earn
0.143
0.101
0. 115
0.166
0.
U
67
0.
0.003
0.0v0
0.
0.U25
0.
F-isc
laLe1
i-ro f
mana
SaCL
Craft
OpIr
LSFc
Farm
Misc
rowsum
5bb4
7U2
10b74
8008
7172
10471
72b
0
Ok
RLStIj
E.ALL
'7
1977-78
Missing
Cut
125736
1927?
53d7
1 1/3
Lata
Pana
SaCk
Craft
UpIr
LSFl
Farm
1447
/o4
b29
701
b72
225b
3911
204
724
372
668
691
954
1503
98
0
52
0
2068
312
587
853
3
U
91
3164
181
377
1497
2351
3206i
349
195
702
1151
449
195U1
52U
52
1459
1510
L.ACL
22b25
IldO
U
-I Uj
1539
112
0
179b
USif%6
riZb.il
ANG fz8.77
SaC1
CraLtU
Upir
1173
5o2
2351
702
19o5
1151
441
321
kESULI Lr
929
1812
420L
1916
Misc
Prof
15
0
U
U
1659
1 U:59
172 b 5
64
0
1443
0
0
71 8b
0
159
FAG?
2
16 AkbkUSL
Cut
&rot
Mana
1927
0
1447
764
3829
672
U
1q 9
It
4bl
3911
9,
0
HtS0L1 01
929
52
U
0U
MULlri
Prot
Cut
U.295
0.141
0.414
0.14 1
0.304
(i.45b
U . 3U 1
0
0
11 d
U
2225
4
1501
0
91
0
0
95q
Ob
'wana
Sati
U.30us
0.
0.143
0.0g2
.223
0.
0.054
i.
124
UI
'..
C5
U
)
U
U
earm
row
ldel
Nisc
321
Out
16
72
112
15
U
Prof
mana
SaC1
Cratt
OpTr
LSFI
Farm
Misc
64
0
0
Q Natrix 1977-78
0. 162
0.
252
1539
349
U
1959
U
249
4b1
202
202
2256
5367
LSFI
LI*.i6/
U SUb'
u
.049
U.6 0
U .21 0
Cratt
0.114
0.214
0.037
0.
U. 19t
U. I 11
.
0.U12
CpIr
0.042
0.061
0.063
0.363
0.
0 .1~i6
0.
LS1i
Farm
0.095
0.003
0
Prof
4901
0.040
0.013
0.166
0.258
0.296
0.
0.372
0.012
0
6
u
Mana
SaC1
Craft
5413
9239
4561
/443
bb3b
.-
a
0.003
0.
Misc
label
0.007
u
UpIr
0
LSE i
0.
U
-
earn
1i, sC
rowsufr
244
0
RESUL1
Uk
Cut
130538
2490
L.ACU
6W/ u
1141
2954
b399
338
0
3782
mana
SaCl
Cratt
Lpir
LSF1
farm
1727
1288
4399
140
1470
346b8
189
168
0
RESULI Gk
1288
0
1d85
189
0
ESLI
Lut
1033
1321
22
0
0
U
878
27b2
380
0
297
1021
1144
1453
14568
191
c
192b
1b39
0b6
1043
181u1
14b5
4c
USING Mzt./d AiL
681
461
SaCI
1192
1409
2260
12 02
fzd.7o
4Ua
16b
0
297
U
1021
U
0
Prct
0.
u.243
0.197
0.00c
0.011
0.0 1
0 U (0b
- Jb
Upir
LbFl
1'11
495
1b b
2954
393
443
1033
6399
764
obb
1321
1 IIh
145o
0
40
U
2295
0
504
O00
UkF MuLik Y
Crart
067 u
380
0.277
0.1l6
0.379
v.259
0.434
0.395
23i3
688
Mana
249o
3399
1407
1470
3468
115$
443
408
'aol
1202
504
1792u
274 -j
1118
lb
0
202b
PAGk
Misc
Cata
1511
128
129
1010
61
46
0
20
40
2715
0
174
188
835
0
0
7341
2
-
Prot
0
172 i
obi
226U
2/142
16873
E.ALD
1RANblOSE
Cut
Missing
Prot
2 39 Jo
140
1529
395
3931
7o4
0
U
1345
1192
1978-79
Q
233
0
10434
19538
30
U
1144
U
U
20
0
Farw
338
0
U
22
18
46
191
0
0
Pisc
row
label
Cut
Prof
mana
SaCl
Craft
Upir
LSFI
Farm
MiSC
atrix 1978-79
mana
CpIr
SaC1
Crat t
0.2o
0.245
0.063
O .Obj
0.122
0.
0.295
0.149
00035 7
0.252
U. 169
000 I 9
o.037
0.
0.026
0.
0.289
0.143
0.
0.115
0.334
0.
Ubb
. 14 7
0.157
0.
0.00
0.25 b
0.
0.Ub3
0.00d
0.023
0 . u3
u.U70
0. 106
0.127
0.127
-.
U.162
0.041
Lbk l
Farm
(.
0.00
0.
Misc
label
rowsum
Prof
6217
Mana
7157
SaCI
Craf
upTr
8949
LS I
ar
7982
i sc
7108
5b09
418
0
RESULI UF L.ALD
1979-80
79
Missing
Cut
Prot
136d57
20 7
11~15
Craft
Lpir
LSFl
Farm
b82
1597
4237
19 4
137
442
7k0
4U6
c47
2465
174ob
1098
311
717
90t;
1281
14b71
74
0
1767
239
0
101
0
60
0
36
2746
0
3 -
mviana
baC1
lbbl
40 7 3
dI
12 to
1 b253
o72
403
2027
81 I
349b
1308
1333
2t215
'481
'175
11916
420
790
'0 2
5291
111 b
23b4
3c20
135
k19
26b
,437
0
k0LO
43
Q
309b
Ui
3425
bb
RkSUL1 LE
4U
U
1916
UsiSG Rzb.'19 ANC
E.ALL,
1337
z2f6.79
Misc
Data
1729
259
143
835
.58
211
540
0
0
7318
FAGE 2
I HAiSEUSE
Cut
Irot
0
Mana
6dat1
CraiL
1175
2027
U
1u7
071
611
1297
0
2027
3495
1304
194
1597
137
4237
239
0
kiLLI
1333
U42
Le
aslU
40
o47
4 db
31
711
U
erot
0.442
0.144
0.44.j
0.149
U.
0 .249O
0.021
0.055
0.
0.042
U
1481
0
6
0. 19
0.159
0 .03,
U
I
185
Ib
0
mara
U.
3c20
437,
790
702
1098
0
30
1)2
Q
0.240
Farn
Matrix
0.15z
0 .0,1
0.044
U O 4
U.
0.145
0.081
V.
Prot
0.053
U.
0.3b9
0.38)
0.
0.113
0.14
Mana
SaC1
Craft
Opilr
LSE I
Fd r m
Mis c
N
1979-80
Cpir
0.
- *089ut
row
latei
Out
0
43
U
40
74
0
0
Lratt
0.31
loisc
U
172
Jb4
90 U
MUClkY
LSkl
19'a
bli I
41
475
Cut
0.531
0.548
- .
111
-*04j
72c
s0l
V
plr
U.
-m
L
i
Farn
Misc
lacel
0.000
0.
Erof
U.077/
0. l0
0.154
- 010.171
0.
0.
Mana
0.005
bal
U.
Craft
Upir
LSF I
0.0t2
0.008
0.009
0.
Eari
1Mis c
rowsuir
4405
4551
6410
7968
43o
0
t.6Ull Uk L.ALU
1980-81
do
Missir9
Cut
I-ro1
Sali
Mana
13997 c
58b
Craft
Cpir
LSO1
Farm
099
1260
3250
191
457
354
633
965
2273
291
332
1495
161 5
1'120
108
0
0)
2,4
0
996
u
307
1461
15455
127
2590
1343
2,
4145
17 I
1d as
157 /7
2t91 9 2bibu
1481
e
900
544
13
3216
IkANb&6t
Cut
Mara
1622
buLI
1t'54
i b
3250
1e
U
bESULI
0I
0
0
'44
0
0
0
t;33
ut* mufir ',
0.
0.
900
2o91
354
u.240
/2-
17t,1
191
45/
etc f
U.)
Cratt
110 t~
tb
Lut
06.39"
0d
10Ud t
154
ni4b.00 AiL tz.80
bat1
2019
u
1424j
259c
699
,12
879
25
253
19
2413
0(
183
0(
1183
Cata
0
0
7355
FAGE 2
b0
-rot
b
14G3
40
(b
USING
L.ALL,
bO
30
11 U
710b
0
1191
134b
b
1b02
62
Lk
144
252
252
173
RESULi
2e31
Pisc
'i 1
LSFI
45934
35443
1)3
524
452
252
/10
1902
996
1J'
0
u
~/ u
0. 1ub
u. 109
u.034
0.L b
0.
farn
row
Label
P.isc
uut
174
46
30
42
erof
M ana
SaC1
Cratt
Upir
G6
62
127
LSk 1
19
b
0
Farr
0
Misc
Lratt
upir
Matrix 1980-81
SaC1
Aana
0 .20
19
1 178
bt
1461
Li
9c5
Q
upir
0.
0. 24 t
U.404u
U.
0 .150
U .045
0134
I 11
0.L .0c 7
I.14/
0.154
0.222
0.050
0.
u. 255
0.211
0.
0 .049
0.03 /
0.093
G.3/ b
0.
0.122
U.
arm
L SE1
0 .ULk9
. L 4?
.u
mi1s c
ro * s UP
0.
0.155
U. 009
0.UOb
0.171
0.005
U.20
0.00/
0.
0. u15
o.i27
Q.
u
6L49
v
u
u
0
7557
5027
5455
bI4b
64
0
0
-
149
0
hESULI Uk
-rci
1336
L.ALD
1981-82
Missing
mana
aaC1
Crart
1031
1404c
223-3
827
.198
38J
354
0
LI
2259
1439
121-1
d6I
107
2,5S
431
5123
453
i/9Y
1t 137
413
760
292
11I
120
643
10
0
20
U
2368
Farn.
478
b52
17
69b
469
1117
15354
23
0
Cut s
Lata
202 6
4
7
9
4
438
IOU
923
51
299
392 3
553
160
3/3
164
299
107
2273
C
101
65
1550
3126
1502
Misc
20
0
0
587
1541
1259
jt6 3
LSk1
14 6
0
0
275 1
0
70G2
14330 6
1742
IHASiPUSL
Frct
Pana
0
1031
1346
90/
354
U
2233
1U37
453
827
383
187
478
52
0
203io
I
bat
2e19b
Lratt
3b'i
27 b
43]
49b
1/
507
1604
U
1280
0
1117
4b9
U
0
16b4 9
Prof
Mana
3563
SaCl
Craft
IpTr
LSl1
Farm
886
15U2
3126
65
U
U
0
0
140
Misc
Uuts
Q Matrix 1981-82
Cut
Prot
Mana
0.4319
0.
0.
0.135
(1.433
0.180
0.214
0.1it24
0.351
0.1(00
0.
0.2t /
0.0'7
0.1/ )~
0.0/9
0.078
0.496
0.d49
row
label
860
23
3923
,99
q'
1449
80
U
u
Cuts
20
10/
U
misc
0
u
10
b
3
7I
tj4Q
rESULI Lk MU~iki
0.034
0.0/5
0. I
karf
268
537
413
1 /79
27
3737
LbF1
up1r
baCi
0 . Uth
LSFl
0.161
0.055
0.002
0.084
0.050
0.
0.341
0.324
0.0/4
0.
G.
0.u57
U . 10 4
0.u92
0.130
0.233
0.
0.409
0.07
bb
0.U78
Lplr
Craft
0.
U .093
U Ulo
0. 117
0.
Farm
Misc
rowsum
0.
4656
U.
6360
8224
0.001
0.016
0.003
U.
4921
S 474
6296
261
U
Prot
Mana
SaCl
Cratt
Opir
LSF I
k arm
Misc
HLSIjL1 Uf
1982 - 83
E.ALU
MiSS Ing
krot
Cut
14677 1
Mana
SaC1
Cratt
763
29b1
891
1528
2663
0
1744
1733
13881
2491
347
835
48
292
48
299
182
390
lo
527
1235
964
0
533
11b4
164
928
52
104u
2262
2427b
314
77
521
0
721
12743
698
5
0
147
U
5
205
1009
14817
157
0
2231
1430
22147
1105
1149
bib
ibo
b76
0
0
2035
2u48
2064
4105
1338
3053
3468
15b
0
37 57
Sbo
124911
1438
403
0
U
0
1069
131b
LS1
Lpir
0
1716
Misc
1farm
85
30
82
398
0
0
7375
2097
0
92
IHANSi-USE
Cut
Prot
0
1430
763
2981
891
1528
6aCl
Mana
204b
2064
0
17 14
110 5
U
3!0
1020
341
d
b
1b2
390
292
4d
0
L
RESLL
Cut
0.102
0.198
0.447
0.
1t
0
U
Li
MCL1k
OpIr
LSE1
farm
14.38
3053
160
204
787
143o
3468
158
576
0
Prot
928
527
403
52
U
Mana
SaC1
0
Craft
U
OpTr
LSF1
4105
1149
1b
IIb 4
2491
0
bb6
527
0
U
7 I
9b4
c
0
0
U
0
Q
0 4
0.07 1
0.010
0.045
0.317
sand
U.,19
0.
Matrix
Farm
Misc
U
1992
19 3
label
baCl
craft
Lpir
LSk I
earm
Misc
0.228
0.122
0.0317
0.035b
0.114
0.129
Prot
0.0989
0.00b
0
0
Mana
SaC1
0.3209
U.
0.1571
0.
0.089
0
Craft
0.203
0.
0.
0.u07
0.
0.
0.
0.024
0
U.33t)
0
OpIr
0.543
U.
0
U
LSk 1
Farfr
-
Misc
0.
~1
0.126
U.
0.041
0.119
0.150
0.
0.279
0.112
U . v O0
0.105
Out
0
82
0
0.159
0.047
0.10
Misc
157
k
Prct
row
label
Cratt
0.033
rowsun
5035
7415
7951
44b0
4418
0396
151
C
w
RESUL T OF MODIFY
1976-77
o ut
PFrof
0.399364
0.194694
0.397844
0.488532
0 . 2015-5 8
0.430564
0.218105
0.223450
0. :54 2582
0.439271
0.678049
0.707468
0.509863
0.552000
0.
0.132145604
0.055151667
0.082081732
0.120577341
0.012026940
0.021.175225
0.001781896
0.042582418
0.
0.024390241
0.029794918
0.
0.040969697
c raft
0.050141243
0.131761087
0.007520682
0.026196298
0.
0.292334830
0.183165696
0.169280114
0.05631861.
0.112348178
0.121.951220
0.046304656
017696970
0.215042373
0.
0.169967410
0.127139364
0.171797164
0.028864657
0.073054526
c leT
0.0280720.-54
0.110169492
0.209177134
0..232138:381
0.158677262
0.
0.113866574
0.01136799 1
0.062587815
0.008819756
0 - 0. 0 1.1646 A
.01 7318794
0. 039703:. 4
0.01781.8959
0.090659341
0.
0 0 18 5 3 1 *z1,8
0.
U.
0.
0.008899781
0.
0.022303030
0.013930092
0.
0.020848485
oper
1ab
t rans
0.000353107
0.028197898
0.
0.017929910
0.048728814
0.043963086
0.068438205
0.067528234
0.277910989
0.
0. i34039174
0.264076978
0.
0.002024291
0.
0.013801109
0.
0.053575758
0.064504633
0.
0. 128542510
0.
0.022313943
0.
0.
Proser
s e r v
fl ab
.560 79 12
0.
0.
902 44-
0.0243
0.045272 '98
0.
S0 7200000
f a rmfl
II
0 . 0005296 6 1
0.008843886.
0.
0.010278265
0.
0.023(63424
0 .020374898
0.039085452
0.062332.354
0. 0006:38 6 51
0.051635664
0.086:273052
0497?7 1 779
0.
0.
0.054iA9F26
0 . 1.04. 05-,
0.104395t04
0.
0.
211!
74907
0.
0.
0.
0.0311 4162
0.030303030
PHHS
!11::7
d
row lbel
c, W
0.143361582
0.092540374
0 . 025570318
0.
0.0110027576
0.042379788
0.
0001127730
0.001119572
0.033515074
0.028057173
0.008196721
0.163461538
0.
0.
0.
0.
0.011393939
Sale
Mana
0.002192700
0.
0.0111515
0.037344772
0.00080R17'96
0. 013/68
0.121881682
0.
0.
0.046558704
0. 151219512
0.
15
C. 04%7
T' . f
0.0508366733
01i
0.
0.
0.
-
73
36
1266
5
1.03 49 099
-
cr a Ft
"IS
92
0.
0.490136571
0.13248485
1394
4
443
91
1
:sr
se rv
PH HL
MI
1
1976-77
outs
Mana
131963
1375
Prot
15 87
22b2
1031
2283d
2 9~111
13o1
1302
1914
1Ob5
1671
1218
159
6:24
123 d
1632
1028
284
27b
343
220
692
2
b92
443
3
24
i47
10
85
2490
sales
b9
490
to18
259'j d
425
560
30
273
0
92
'1
134
115
0
0
40
317
1 00
0
807
1b
0
455
0
15 8
b631
741
181
0
41
0
475
3471
2195
9
2U9
1
39b
152
23
0
236
U
3o
5
1394
0
1543
1if/4
52
75
346
538
4b
4a
344
120
102
5
50
30
474
3584
119
0
0
0
91
'6A5EubL
Oi
ardctIli.dio
cuts
Mana
rot
6ale
CIer
cratL
Uper
1375
0
13b I
1210
S3u4
100b
443
123t
1671
276
343
t92
2
U
1914
a24
ib3i
220
273
bbO
1)
t)9
92
175
'ibb
536
94
0
U
22b2
1519
15 b7
1031
220
419 t
705
1576
926
07 d
10924
1 j45
1tG
20b5
15
414
40
3U0
5
5S
e) 9
0
621
395
4J4
139
54b!)
33c
22 7'
153iei
31
0
1
0
0
6
U
231
I9
, Jii,
U
U
i9u/a
10!
75
11oI5
30
2256
0
2176
1823
0
3 dd 0
a31
U
0
U
497
342i
41
111
25
3b9
I54
306
322
0
2
0
0
107
173
0
3
14b
U
25
0
0
0
0
0
1588
31
0
36
5
serv
PHHS
5485
231
108
69
351
359
107
173
336
5
0
0
209
53
152
23
0
119
120
(I
0
125
0
24841
2292b
5648
4425
4173
0
0
0
U
41
'
990
0
0
0
0
0
1691
Serv
row label
0
0
1
2
3
4
5
6
7
125
8
0
47
46
548
144
6672
323
578
5
1266
10
0
40
0
9
24
0
41
0
2277
169
86
92
297
73
221
0
0
0
17
13520
749
FrSer
Missing
Data
0
0
0
104
0
0
247
421
104
111
2
127
148
120
b27
0
0
FaLab
13
0
5
0
0
92
0
0
0
(3
139
0
0
0
0
46
irans Labor Farm
181
U
127
74 1
1Vd
os
'4
264
102t
434
0
76
342
0(
443
PrS
FLa
395
31
66
627
3,1
55
108
1823
16128
322
19035
2176
306
Irans LaLor FEar
414
40
133
180
134b
92o
Upers
2b85
75
944
7U5
0
722
u12
Craft
1578
4 19et
4s
6786
0
80
0
Lierc
9
10
11
12
13
Misdata
Mdata
PHHW
2490
317
812
0
722
102
364
39b
236
26
342
0
46
31
0
323
548
0
0
120
0
19412
1805
5
36
0
0
0
0
749
0
144
1620
1516
807
455
1973
1543
1394
474
443
91
92
36
1266
5
0
11715
row label
outs
Mana
Prof
Sales
Clerc
Craft
Opers
Trans
Labor
Farm
FLa
PrS
serv
PHHS
Misdata
colsum
1977-78
RESULT OF MODIFY
Mana
P rof
out
0.295246
0.141142
0.31410
0.377128
0.132667
0,287131
0.284099
0.236973
0.401639
0.301486
0.351916
0.462701
0.435590
0.472362
craft
105488676
208571956
023922365
037604457
194401244
145149526
127322157
012295082
350318471
01/421603
072484967
1ab
0.
0.
0.+
0 0 174256 s t
0.016329705
0.028811087
0.009968283
0.372950820
0.
V.
0.000487567
0.
0.021014162
0O
0.162202106
0.015346423
0.107242340
0.102462825
0.027410575
0.022246535
0.057091074
0.
0.
0.139372822
0,097188363
0.
0.071265418
0.305()447868
0.
0.298803882
0.114825132
0.071096654
0.032076205
0.075492341
0,013593113
0.21311474
0.004246285
0.
0.106614660
0.
0.134079488
0.041420118
0.029743211
0.044685173
0.036985453
0.317611524
0.
0.1 0211 5244
0.351155415
0.
0.036093418
0.08710801 4
0.072647489
0.
0.041114664
Proser
0.009181800
0.004064290
0.
0.
0.0058085,50
0.008748056
0.0058351 7
0.
0.
0.
0.
0.010076386
o.
0.
0.044480718
0.128764086
0.0987,-j5356
0.23425)878
0.192733017
0.123800681.
0.004646840
0.051821351
0.083880379
0.006796556
0.
0.006369427
0.
0.01.8527548
0.
0.035861124
trans
oper
cler
Sale
l
abo
0.019189106
0.038'366057
0.008356546
0,061802974
0.031687403
0.
0.100135931
0.
0.
0.
0.001137656
0.039301310
0.000228415
0.109200743
0.149494557
0.084245077
0.
0.
0.233545648
0.
0, .03%3A42
126
0.
0.001 142074
serv
-
75
PHHS
-
farm
T
0.003468680
0.019951967
0.001020200
0.031405875
0.004965019
0.092226
0.026603
0. 052358
0.167595
0. 133829
0.126361
0.148796
0 .027186
0.
0.-004246
0.205575
0.
0.525109
0.063271
0.039962825
0.070762053
0 . 01.9328957
0. 047575895
0.
0.
0.198606272
0.073785:145
0.
0.158976702
issd
C.
0.
0.
0.
0.000 777 6 0 r
0.
0.00679655
0.
0.
0.
0,050706972
0.
0. 000685244
0.003264640
0.013301312
0.014669375
0.007273290
0.003485130
0.
0.
0.015
(05528
063694268
row
1.
51
703
419
1347
1508
1435
4
2)
4:)
301
159
66
1049
27
0
label]
C)ut
rowsum
491
'413
443
sale
cler
44
trsn
t
ra -
flab
serv
PH HL
Miss
'41
REbULI
Of
E.ALL
USl~i
tzL.77
mzt.i7 ANL
1977-78
outs
12573a
Mana
144'
irot
7b4
189o
4bo45
1927
1b47
3740
1110
2135
216b
335
1497
2
484
517
20j
5
17
20401
11;
321
I17Ci
4
3574
174
1910
45
42
0
1510
Saies
13i92
bb
091
b93
1324
/42
obuio
b4
170
19ts
22
lob
170
72
0
0)
2
144
Clerc
duo
23934
243
139
124
54
U
U
U
2.2
U
I03
1 .14 7
449
Craft
Opers Irans Labor Farg FLa
1417
779
f61
2C7
523
126
30
98
0
230
53
39d
15
0
0
3
571
441
306
20
172
18119
1367
26b
4'? 0
1000
14399
163
769
1b
75
b4
79
22
45
16
0
141
282
364
0
25
57t
0
1Ub
050
4
1435
2b0
3662
231
0
105
2a1
775
221
2143
34
60
15
301
40
2k4
142
0
2
3
0
165
17
0
110
30
b22
0
2
0
66
52
0
0
0
3164
91
0
0
0
159
PrS
serv
PHHS
101
40
0
0
57
5
25
0
0
0
0
1352
59
0
151
2847
598
656
114
454
446
447
7
207
399
0
0
0
0
0
0
36
0
0
0
0
481
783
27
0
3
62
12505
312
1049
Missing
Data
row label
2068
312
587
157
696
0
1
2
3
4
0
5
180
1
5
0
92
6
7
6
9
0
277
3
7186
10
11
12
13
Misdata
1iAAbELSE OF nactili.bil
cuts
mana
irot
)ale
Cler
cratt
Oper
rans Laoor Fan
1927
1491
0
It4 I
:1740
4d4
117/u
51)
1129
2135
203
216
335
ibe
i0
l
5
170
11
b!~
1124
0
22
170
54
1
1447
7b4
132
i
2447
571
1477
779
441
306
141
105
01
0
U
40
1Ij715
10u
job
207
3v
52
2
0
k (
'
.3l
315e.,
172
U
0
14U
0
0
U
31
371
U
"54
U
o0
JH7
2/30'
244
I UU (I
396b
3
105
05-0
284/
399
20c
( r
1 400
1424
98
142
097
Iv52U
J1i8b
44b
U
23tS2
161
196
239
1367
0
260
775
17
25
447
0
160
2U246
124
2b6
163
0
221
0
0
0
470
7b9
231
0
0
110
0
207
U
1
5112
:I
4614
FaLab
321
16
72
0
0
0
47
15
0
0
34
0
0
75
84
79
22
91
0
0
3
0
0
0
0
94
3764
-1093
Serv
3574
452
144
232
1083
576
650
408
60
0
2
59
0
17
65
30
0
PrSer
1570
PHHW
Mdata
174
0
0
481
0
277
3
1916
1510
703
449
1347
1506
1435
224
301
159
66
151
1049
27
0
20226
128B
10d45
0
0
0
4
0
15
0
0
0
312
row label
outs
Mana
Prof
Sales
Clerc
Craft
Opers
Trans
Labor
Farm
FLa
PrS
serv
PHHS
Misdata
colsum
RESULT OF MODIFY
out
0.277786714
0.166043574
0.258259046
0.379977963
0.199393151
0.293614220
0.070967742
0.150943396
0.395397490
0.121212121
0.111111111
0.464017871
0.272997033
0.392773590
craft
0.039996783
0.145417043
0.006554798
0.032740438
0.
0.244459074
0.262903226
0.197641509
0.
0.173160173
0.102880658
0.066060316
0.
0.011957369
flab
0.
0.
0.
0.
0.007368877
0.021066491
0.001612903
0.056603774
0.
0.
0.
0.008297431
0.
0.006498570
1978-79
Mana
Prof
0.
0.243006317
0.042737284
0.081536282
0.069498627
0.009436032
0.013440860
0.
0.351464435
0.006493506
0.
0.064624222
0.
0.033272680
0.226636641
0.
0.326953330
0.159452227
0.169050715
0.032038622
0.125806452
0.030660377
0.
0.
0.131687243
0.037019307
0.
0.033532623
oper
0.030561364
0.040092819
0.104614578
0.072721549
0.285363387
0.
0.187096774
0.261320755
0.041841004
0.170995671
0.
0.087282591
0.
0.042890564
Proser
0.003538684
0.003867475
0.
0.
0.004479122
0.
0.021505376
0.
0.
0.
0.
0.036859742
0.
0.
Sale
0.049059032
0.118602553
0.
0.111758224
0.01603814b
0.012727672
0.0446236b6
0.000943396
0.127615063
0.
0.
0.
0,
0.005458799
trans
0.032652405
0.017016888
0.043786051
0.000787030
0.057650629
0.087777046
0.
0.121226415
0.
0.034632035
0.016460905
0.
0.
0.005978685
0.104230
0.084827
0,081542
0.129230
0.031787
0.111477
0.046774
0.093396
0.083682
0.
0.209877
0.
0.663205
0.210294
-
77
-
0.196879524
0.177259250
0.133193498
0.
0.041756972
0.094140882
0.019354839
0.077358491
0.
0.
0.427983539
0.129567576
0.063798220
0.257083442
labor
farm
0.018658517
0.003867475
0.002359727
0.028018259
0.115012281
0.092824226
0.181182796
0.
0.
0.123376623
0.
0.010531355
0.
0.
PHHS
serv
cler
mis.sd
0.
0.
0.
0.000314812
0.
0.000438885
0.
0.000471698
0.
0.
0.
0.095739588
0.
0.000259943
0.
0.
0.
0.003462931
0.002600780
0.
0.024731183
0.009433962
0.
0.370129870
0.
0.
0.
0.
row label
1345
878
674
2078
1977
1.427
498
495
174
60
49
1082
2
0
rowsum
6217
7757
3814
6353
6921
4557
1860
out
mana
1rof
sa] e
cl e r
craft
tran
f a rm
flab
Psr
serv
PH HL
Mi S's i
21
47 c
24
626./
i
674
3847
HESLL1 I*
opts
13053
2490
1192
1698
4972
1081
25 e
392
484
33 ts
199
0V
5131
5 b6
378
E.ACL
uSIwG
s
Mana
1-rot
1727
23930
1409
30b
1224
373
190
203
116
6
12u8
920
1375
1128
311
132
30
0
0
22
04b
0
1345
i.Zi .Ib
bales
25
399
107
1170
111
269
16733
1915
399
796
18
96
0
821
2
0
31
220
0(
1977
2078
6/4
Opers Irans Laoor Farr fLha
1338
43
146
58
429
1114
12960
400
423
0
0
31
0
1380
481
1013
710
19793
20b
402
5
17 ti
22
30
b it
1 97'-1-- 7 9
Cratt
51
508
0
Clerc
2414
985
163
1247
0131
9
0
0
fZB.76
AisC
132
25
234
83
36
409
348
4453
337
46
3
0
40
508
2
1427
0
320
0
65
2
164
419
554
257
1735
20
120
0
19b
1
189
168
0
61
0
0
20
0
0
2715
0
0
40
0
174
49 b
419b
56
3
0
0
0
80
79
16
57
171
678
0
0
0
60
PrS
27
0
32
0
104
25
0
4
0
0
0
1131
51
0
49
serv
2908
405
232
0
812
414
547
0
66
0
52
231
10130
600
1082
PHHS
184
0
0
0
43
0
0
0
0
0
0
0
447
484
2
Missing
Data
1511
128
129
21
989
46
165
23
0
0
25
0
809
1
7341
row label
0
1
2
3
4
5
6
7
8
9
10
11
12
13
Misdata
Transpose
c;t
i-ana
0
2490
0
1727
12bb
965
1885
2414
13bu
1338
bib
481
43
132
2b
320
189
U
1ib
11Ile
14uy
0
1247
11u
1170
14 t
290b
184
1511
0
405
0
1:6
Craft u e r
109
1081
3/3
3ub
0
137r
bua
710
0
1114
399
462
1975
0
Jo
k'9
1:3
3 t.
46!2
34 8
2
104
419
0
$0
554
20
7 1u
0
ci
U
'34
0
129
2562
190
311
2b
zub
20
16b
0
21
cr
Pret
0
0
i
q0
25
At14
43
0
9Ot9
4c
79
0
b47
0
165
T rail
392
2U3
132
167
b
399
400
0
257
Sb
4
0
0
234
,a
4a4
116
30
9
17 b
v' r.
r
338
199
0
0
0
0
22
796
423
337
0
57
c
66
0
0
46
20
0
171
u
0
0
0
0
0
0
51
96
3
120
0
0
(j
52
0
25
Prd*SirScrv
60
5131
22
648
30
658
0
311
0
821
31
220
0
508
87
40
0
198
0
0
0
231
O
0
40
0
51
0
447
809
585
0
0
2
0
2
0
1
0
0
0
600
0
1
3782
1345
878
674
2078
1977
1427
498
495
174
60
49
10d2
2
0
LO
Li
L2
L3
L4
L5
L6
L7
L8
L9
L1O
L 11
L12
L13
Missing.Data
4
1979-80
RESULT OF MODIFY
P r of
out
0.422474
0.144615
0.291912
0.415538
0.137455
0.262025
0.119976
0.313067
0.548165
0.171533
0.421348
0.528126
0.534066
0.458013
Sale
Mana
0.
0.231276748
0.061764706
0.175384615
0.040307971
0.020962006
0.015225335
0.002133333
0.
0.
0.089887640
0.074212131
0.
0.068609272
0.228603859
0.
0.214705882
0.115230769
0.163496377
0.050907730
0.081607795
0.
0.231651376
0.
0.022471910
0.052355134
0.002197802
0.037880795
c Ie r
0.135754824
0.221825963
0.27279411.8
0.
0.016799092
0.139621969
0.
0.092615385
0.015172101
0.035186225
0.096833130
0.
0.
0.
0.
0.017621145
0.
0.009006623
0.092391304
0.026763990
0.09561.5104
0.052800000
0.
0.007299270
0.
0.086750254
0.
0.212185430
RESULT OF MODIFY
craft
0.091486947
0.141226819
0.054044118
0.052307692
0.
0.382930938
0.200974421.
0.205333333
0.137614679
0.321167883
0.
0.034903423
0.
0.015364238
RESULT
oPe r
0.029284904
0.032453638
0.014705882
0.034923077
0.318614130
0.
0.176004872
0.234133333
0.
0.222627737
0.106711573
0.075906472
0.
0.044768212
0.014755959
0.006597718
0.056250000
0.
0.073369565
0.053527981
0.
0.093333333
0.
0.018248175
0.123595506
0.025923416
0.
0.011125828
fa rm
labor
trans
0.
0.037446505
0.000735291
0.013538462
0.066802536
0.118847090
0.091352010
0.
0.
0.244525547
0.
0.033209082
0.
0.008741722
0.
0.
0.015808824
0.
0.
0.003743215
04 01218026
0.039466667
0.
0.
0.
0.
0.
0.
OF MODIFY
f lab
002923077
01607/899
003368894
0:0450670
027733333
082568807
0.
0.001863775
0.
0.004457917
0,.
0.
0.009737319
0.000935804
0.063337393
0.
0.
0.
V.
0.038461538
0.
PHHS
serv
,= rose r
w i SS c;
0.051078320
0.040477889
0 009761635
0.017279412
0.
0 .021230769
0.076307692
0.066576087
0.040801048
0.016443362
0.032000000
0.
0.014598540
0.235955056
0.
0.463736264
0.132715232
-
79
-
0.
0.
0.
0.
0.
0.
0.030667570
0.
0.001589404
rcjw laiiel
2061
668
904
out
C' W'
4- M
T
4405
560
sale
1857
932
405
635
f
19
61
1092
41
0
.4
cr
trn
9 11L
r
Pst
fse sosi
rv
P' HHL
18
btSUL1
GF
E.Acu
USIiG
mfzo.79
AND
tzd.19
Mana
Prot
sales
Clerc
19T9-80
Missing
outs
136257
2(27
1175
1200
1bb1
24073
1007
74
40b3
1072
598
19bb
403
12,i
3 96t
05
4o0
1b5
0
to
96
0
46
0c
2723
341
34k5
22b
43
40bi
d 11
1297
'94
iob
1b253
783
1444
792
oo75
742
1H2
37
210
0
0
25
'227
0
2701
1140
749
602
1d196
340
227
Cratt
b07
7Z2
0
67
40u
16971
140 1
324
2
dd
295
0
0
47
19
71
43
294
0
147
40
153
0
0
bOob
0
49o
138
2192
1857
Upers
1400
197
112
25
134
159
272
188
143
2046
12415
286
b35
20
18
5
218
157
33.0
2b9
4495
Ib
20
0
0
101
0
99
0
0
60
0
0
0
2746
36
0
0
0
38
385
439
175
1822
74
52
104
0
60
0
6,5
0
932
4
50
27
405
47
0
239
57
0
0
2
88
61
5
67
0
666
0
4
0
19
Data
serv
PHHS
486
0
2
0
0
0
3117
438
309
104
512
206
0
58
19
448
0
169
22
0
0
0
153
196
0
42
33
PrS
Irans Labor Fartr FLa
75
16
4
904
42
0
61
0
0
801
0
0
0
0
0
11
227
11059
181
1092
1729
259
143
34
row label
0
1
2
3
4
5
6
7
8
0
0
0
501
422
331
41
6
7318
9
10
11
12
13
Misdata
0
O
16AN, tkLSEU
r actill.V79
outs
mana
Prof
Sale
Cler
cratt
Cper
0
2027
11,5
1;0 b
4003
1072
1988
396
403
129
182
40
227
1407
0
65
37
153
U
811
794
2701
607
1400
197
587
239
47
75
3117
4bo
179
149179
1297
Ibb
1140k
17 b
11
25
4
0
100)
(I
*,49
122
27i
134
101
0
16
k59
4
309
2
143
4947b
23414
438
U
74
792
b 0,
02
188
159
0
0
0
104
0
34
9eUU
7i2
0
40$
143
151
99
141
340
0
2u4b
330
289
0
60
439
0
2
88
01
0
U
512
0
801
206
0
td
19
448
'bbid
22btu
0
1o9
17b44
FaLab
irans Labor Farn
0
324
286
0
175
0
5
22
153
0
42
bi1
460
0
210
46
185
0
0
2.
88
295
43
635
150
0
0
67
0
19b
0
20
0
25
0
0
43
5
104
0
20
74
0
0
0
0
0
0
0
0
33
0
0
3925
3088.
PrSer
227
0)
0(
1019
1354
Serv
Mdata
PHHw
2723
225
227
47
496
294
218
27
60
0
4
42
0
422
501
341
43
0
6
3425
2064
668
904
2192
1857
932
405
635
38
19
61
1092
41
0
15b44
1034
14333
0
138
0
0
0
0
0
0
0
181
0
row label
outs
Mana
Prot
Sales
Clerc
Craft
Opers
lrans
Labor
Farm
FLa
PrS
serv
PHHS
Misaata
coisum
RESULT OF MODIFY
PAGE
1
1980-81
Mana
Pr of
o ut
0.240734
0.088434
0.352747
0.297770
0.138443
0.230787
0.183047
0.263746
0.724832
0.191489
0. 1214t$2
0.529526
0.426752
0.521330
craft
0.140871654
0.222740262
0.019000413
0.057118353
0.
0.282726204
0.117936118
0.123281787
0.
0.273556231
0.097689769
0.074056147
0.
0.000458716
flab
0O
0.006932066
0.028907168
0.027641278
0.018041237
0.127516779
0.
0.002640264
0.000484027
0.
0.
0.
0.198977290
0.086327964
0.101372213
0.108536344
0.044653349
Sale
0.
0.
0.
0.095709571
0.075508228
0.
0.066743119
0.285175017
0.
0.315985130
0.186792453
0.149534561
0.02820211f5
0.143734644
0.070446735
0.
0.063829787
0.075907591
0.080590513
0.
0.076146789
oper
trans
0.000614251
0.025566232
0.030079711
0.041305246
0.072898799
0.306199247
0.
0.154/91155
0.234536082
0.
0.
0.031023102
0.031703775
0.021231423
0.057568807
0.004118051
0.007820725
0.009913259
0.027615780
0.070509012
0.040423032
0,
0.058419244
0.
0.085106383
0.
0.019361084
0.038216561
0.051132464
0.157918484
0.
0,085934820
0,009308774
0.013866040
0.135749386
0.000859107
0.
0.
0.045544554
0.148936170
0.246804031
0.079719124
0.
0.
0,036442860
0.090011751
0.044840295
0.109106529
0.147651007
0.
0.030363036
0.093417231
0.292993631
0.069036697
0.132568807
0.016940949
labor
f a rm
0.016129032
0.004662355
0.021891780
0.029331046
0.102594573
0.133490012
0.137592138
0.
0.
0.276595745
0.
0.022991288
0.
0.
0.009324710
0.019000413
0.
0.005941771
I.
0.025798526
0.030927835
0.
0,109422492
0.012541254
0.
0.
0.
0.012844037
PHHS
serv
'roser
0.013 555213
0.
0.
0.
0.004357298
0.
0.
0.
0.
0,
0.
0.005324298
0.
0.005275229
cler
0.073781743
0.03323808
() .054109872
0.141166381
0.061200238
0.105992949
0 . 028255528
0.090635739
0.
0.
0.484488449
0.
0.220806794
0.056880734
-
81
-
i
000940071
0.
0.
0.
0.002640264
0.050096805
0.
0.001146789
sd
row
1348
1191
699
1357
1086
938
586
448
183
58
label
ou t
maa
,al e
ro
ws u m
5828
6649
2421
5 83 0
5049
cra:ft
42 5
tr'a r 'T
-S T'
665
0
0
s (I r V
4
tFa
MI .Ti
43c'.
RESUL1 Gk
outs
13997b
2019
1622
L.ALU
UbING Wz.6b0 ANL
Mana
Prof
5d88
1323
15777
1050
1403
22934
b68
3337
1756
1923
670
3c4
174
135
2:
2664
380O
3216
IRAkbsGSL
29b
bb
821
149
24
94
0
U
0
1403
982
2019
0
1323
209
591
546
190
2 ib
1
184
2188
201
2213
0
0
14b
312
0
291
149b94
2b272
03
184
19460
1 b1
356
171
*0
518
4b
24
b3
298
1
190
120
59
47
234
383
1203
11657
172
5b68
1546
30
0
123
221
73
192
252
4104
224
42
45
0
46
0
586
614
0
164
2
254
287
546
136
1673
72
42
108
0
0
0
22
0
0
0
2413
19
0
211
0
184
145
115
69
46
148
47
0
63
0
21
0
0
90
0
28
91
36
517
0
0
0
68
0
0
0
19
4
867
734
4
0
0
0
U
35
131
t23
0
135 7
309
1 Ob6
Sale
Cier
crait
Uper
806
3337
1156
1923
670
364
174
135
19 b
808
1041
193
0
1d4
821
149
24
94
0
200
100
425
1546
0
252
546
52
24
161
356
172
0
136
0
28
31
62
46
0
30
0
42
72
0
0
0
0
0
Q3
0
0
obs
22
0(
4
936
0
448
0
1823
serv
PrS
2
2188
312
333
70
386
306
131
80
95
0
2
22
11119
207
665
PHHS5
Missing
Data
row label
2273
291
332
301
578
2
251
56
0
0
0
23
248
5
7355
201
0
0
0
138
0
10
18
0
0
0
0
104
652
0
0
1
2
3
4
5
6
7
8
9
10
11
12
13
Misdata
OF ffacti11.80b
mana
614
108
18717
333
425
62
1191
9b2
755
100
1348
699
548
10b9
b uI
bbbS.
200
52
221
Upers Irans Labor Farr FLa
7lo
591
1980-81'
Craft
1 136
193
31
Clerc
854
209
1b41
1481
79
43u
0
cuts
bb
854
173t
b99
sales
tZb.b0
irot
765
755
120
234
164
0
115
21
1050
lu
U
1481
333
Jo3
0
i203
221
13
192
2
254
47
b9
U
22
0
46
333
70
3bo
0
332
3u1
dzoS7
O994
(I
138
2b7
0
9U
30e
U
2
2c2a42
21123
b
0
47
0
131
10
251
80
16986
FaLao
Irans Labor Farr
1d
5b
5825
53
171
516
568
224
0
0
91
0
95
0
0
3882
0
36
19
0
0
0
2894
PrSer
35
123
45
42
19
0
4
2
0
0
922
Wli'
1015
di33
Serv
Mdata
PHHW
380
2664
430
221
131
823
309
451
46
211
0
0
734
0
104
248
4
207
0
. 5
17243
1247
14061
0
0
0
0
0
4
0
0
0
0
10bb
3216
1348
1191
699
1357
1086
938
586
448
183
68
2
665
0
.0
11787
13043
row label
outs
Nana
Prof
Sales
Clerc
Craft
Opers
Irans
Labor
Farm
FLa
PrS
serv
PHHS
Misdato
colsum
colsum
RESULT OF MODIFY
out
PAGE 1
Prof
0.309064
0.135220
0.362668
0.399104
0.180045
0.247666
0.274766
0,167447
0.249042
0.072386
0.288235
0.573974
0.657296
0.424257
craft
1981-82
Mana
Sale
cler
0.
0.286941581
0.047680412
0.162106918
0.079862199
0.094922005
0.
0.244910742
0.234981746
0.174558017
0.165566038
0.185534591
0.
0.181020983
0.
0.078439342
0.038982260
0.067289720
0.009570495
0,077829709
0.031746032
0.031775701
0.
0.199233716
0.056300268
0.086764706
0.106641469
0.004016064
0.106673161
. ope r
0.076030928
0.163050314
0.053554651
0.046797212
0.043478261
0.
0.352007470
0.168847352
0.
0,164485981
0.053019828
0.039937107
0.026307548
0.285917496
0.069200133
0*000609632
0.112997658
0.161764706
0.038876890
0.173529412
0.012958963
0.070843091
0.
0.
0.051470588
0.069384449
0.
0.024354603
0.009370817
0.
0.033365806
0.191427180
0.065134100
0*131367292
0.
0.009664948
0.044496855
0.008455997
0,006637902
0*242388759
0.061373600
fI ab
Proser
se rv
0.
0.
0.
0.000331895
0.000609632
0.
0.
0.057962529
0.183908046
0.
0.
0.
0.
0.000730638
0.001503436
0.000471698
0.
0.010122801
0.009550904
0.005602241
0.01869159
0.003512881
0,084291188
0.
0.
0.005669546
0.
0.
0.047036082
0.081761006
0.025367992
0.078659144
0.038813249
0.102474323
o.027414330
0.038235294
0*053547523
0.
0.053619303
0.120588235
0.
0.275769746
0.133463225
83
0.009020619
0.021855346
0.014719699
0.015765018
0.078439342
-
0O
0.
0.003131851
0.
0.016256858
0.125816993
0.004668534
0.128348910
0.
0.
0.
0.061662198
0.
0.
0.
0.
0.
0.141762452
0.107238606
0.
0.028347732
0.
0.000487092
PHHS
0.148711944
-
farm
labor
trans
0.166276347
0.
0.008042895
0.013235294
0.046706263
0.
0.012420848
0.066414687
0.042483660
0.089096573
0.040498442
0.000585480
0.0/6628352
0.055679740
0.035247432
0.
0.029274005
0.
0.300268097
0.066176471
0.028617711
0.
0.011446663
0.
0.209115282
0.160008591
missd
0.
0.
0.
0.
0 .0032513/2
0.003734827
0.005607477
0.
0.
0.
0.
0.022408207
0.
0.
row label
221,j 9
out
1214
571
1797
1259
1120
421
660
101
38
it
1
729
12
0
mana
Prof
sa I e
cler
C ra ft
t ran
l abo
farm
f Ab
Psr
serv
PHHL
Missi
rowsum
4656
6360
3193
6026
4921
4284
1605
1708
261
373
680
3704
747
4106
RE .:1
U
E .ALD
USN1G RzE.81 Ai)
fzU.81
1981-82
cuts
14330 b
2', 2c
909
490
/9,
1c 9
I
Mana
1439
22729
1 33b
222
745
354
247
45
42
860
1031
14846
1053
1180
1037
254
283
139
0
U
tic
119
272 1
1/4
27b1
7
219
G
2259
IhAN.SE.USE UP
outs
Prof
3
520
0
1214
Sales
1158
255
182
5782
578
171
84
27
47
10
0
0
81
0
571
Clerc
2405
512
'17
1b3'16
282
262
40
95
0
2
81
414
0
1797
Cratt Opers Irans Labor Farm FLa
886
363
859
3
386
13386
1407
274
386
80
3
47
191
16
1259
1061
136
167
41
182
1508
9758
151
539
20
0
24
439
16
1120
286
0
193
1
121
284
414
50
1547
0
99
6
254
0
600
441
51
108
65
143
271
264
3964
206
20
3
41
9
421
Prof
Sale
Cler
craft
Oper
1604
1336
0
182
1416
859
167
108
193
17
49
110
144
0
100
909
222
1053
0
417
3
41
2828
745
1180
57d
0
1649
354
1037
171
282
0
1508
271
284
27
196
212b
491
1742
202o
0
1031
2 55
572
383
13b
51
0
52
21
59
3#5
3
438
3
9
173
0
51
2504
247
254
84
262
1407
0
264
414
0
78
2et
246
40
252
490
45
2b3
27
40
274
151
0
50
(
0
112
45
106
154747
17713
21b31
19127
15564
5587
1t U
00b
4'i1
27
21
49
0
0
3
78
112
40
23
599
0
20
0
38
serv
PHHS
196
59
110
118
35
9
26
45
0
0
0
1920
82
0
111
2126
395
144
48
257
173
246
106
105
0
0
21
9604
83
729
491
3
0
7
0
0
40
0
0
0
0
0
206
768
12
Missing
Data
1742
436
100
137
786
51
252
47
2
0
3
0
548
0
7002
row label
0
1
2
3
4
5
6
7
8
9
10
11
12
13
Misd ata
nactill.Bbi
Mana
U
65
52
17
20
0
0
0
0
37
2273
48
22
0
0
101
PrS
1
20
U
11
48
7
137
bo8b
162
143
121
0
0
35
257
0
ido
24601
0
Trans Labor Farm
0
47
797
42
139
47
95
386
539
20b
0
37
40
0
105
0
2
146
0
0
10
0
80
20
0
C
0
23
0
0
0
0
3980
2552
FaLab
106
0
0
0
2
3
0
0
99
48
0
0
0
0J
3
857
PrSer
Serv
PHHW
Mdata
119
7
3
0
61
47
24
3
6
22
0
0
21
0
0
2727
219
520
81
474
191
439
44
254
0
20
82
0
206
548
174
0
0
0
0
16
16
9
0
0
0
0
83
0
0
2751
2259
1214
571
1797
1259
1120
421
660
101
38
111
729
12
0
2233
14o61
1066
13043
row label
.outs
Mana
Prof
Sales
Clerc
Craft
Opers
Trans
Labor
Farm
FLa
PrS
serv
FHHS
Misdata
colsum
RESULT OF MODIFY
ou t
1982-83
PAGE 1
Prof
Mana
0.284012
0.102900
0.227776
0.384524
0.198884
0.356865
0.220416
0.196312
0.
0.365772
0,481959
0.513528
0.824324
0.538937
craft
0.122144985
0.159676332
0.233715442
0.079405252
0.130782313
0.077455357
0.012784880
0.001806685
0.012472885
0.317880795
0.092398022
0.026266996
oper
0.036064537
0.044217687
0.
0.278488049
0.210478771
0.189804772
0.033112583
0.119127517.
0.054123711
0.072320499
0.006756157
0.009270705
flab
Proser
0.
0.
0.000340136
0.010044643
0,012506948
0.044263776
0.011930586
0.410596026
0.
0.
0,
0.008341609
0.008361430
0.
0.
0.003348214
0.
0.002710027
0.
0.
0.078859060
0.
0.010145682
0.
0.008652658
0.
0.
0.
0.113402062
0.
0.005253399
0.
0.036082474
0. 133454735
0.
0.159456119
labor
farm
0.
0.018730489
trans
0.005163853
0.029669589
0.077190762
0.009523810
0.063392857
0,012506948
0.
0.056399132
0.
0.
0.067010309
0.027315297
0.06081081.1
0.001545117
serv
0,091757696
0.073364801
0.03163555,i8
0.043027211
0.023437500
0.049749861
0.044263776
0.026572668
0,
0,
0.010309278
0.
0,108108108
0.097960445
85
0.
0.090848214
0. 103946637
0.093044264
0.140455531
0.035491071
0.002501390
0.037037037
0.057483731.
0.052289282
0.
-
0.177357100
0.168846932
0.165453970
0.
0.096428571
0.190136054
0.186383929
0.029738744
0.067750678
0.078633406
0.105960265
0.
cler
0.050645482
0.167093729
0.219463754
0.
0.361910788
0.
0.054123711
0.064516129
0.
0.026613704
0.005933918
0.019614046
0.072278912
0.257589286
0.
0,218608853
0.229934924
0,
0.119127517
0.123711340
0.057752341
0.
0.043881335
0.003773585
Sale
-
0.010526316
0.039109912
0.000949067
0,012244898
0.053125000
0,140911618
0.057813911
0,
0.132430331
0.092281879
0.
0.022372529
0.
0.009270705
0.
0.007012812
0.
0.
0.
0.
0.
0.
0.
0.224832215
0.059278351
0,
0.
0.
PHHS
row label
missd
0.
0.004315577
0.
0.016496599
0.
0.
0.001806685
0.
0.
0.
0,
0.027575442
0.
0.007107540
2035
1.069
503
1592
1318
1320
396
578
92
117
106
1341
89
0
rowsum
uut
mana
-rof
sale
.e r
craf
t
t ran
labo
farm
flab
Psr
Se TV
PHHL
Missi
5035
7415
'A161
5830(
4480
598
1107
1844
151
596
388
3844
148
3236
kkhSUL'
Ue
outs
146 771
20ib
E.ALD
0o616G mZo .62
Ai C tzi,.o2
9ana
Prot
Clerc
in 3
1733
13 .l1
1430
42147
11U5
255
20-4
2891
1339
19
894
2412
015
1134
134
44
zo
630
158
240
53
(j
19
247
42
1973
462
3 /b7
2035
IMAbiUSE
c
1430
703
720
2261
204o
1 /33
251
709
347
0
21o
4
544
34
frof
2U0 1
U
1144
75
1 l
9
12492
1002
1154
b7 2 u
24q.
3
72
0
U
423U
100.
(J
oU3
41
374
284
238
U
45
507
45
45
15
3o2
23
145
100
259
244
4b
114
62
17
145
U
24b
01
2/83
sa.le
1214
2!).'
1o3
233
2.2
373U
16
0
0
0
0
1506
04
20
20971
U
0
49
0
22
0
3
10,
0
179
49
49
0
2
0
131 e
1320
396
Cler
crdtt
Lper
1338
t;15
2'.12
218
0
0
0
0
71
71
0
55
134
486
47
0
0
117
0
48
350
424
104
97
1 b,2
0 9i1
57d3
62
0
0
0
92
44
11 b4
,23
11,1
62
220
244
445
56
1154
264
242
424
4u735
io0
9
3/4
0
1U02
41
103
233
Ivu
57
0
259
350
0
U
U
0U
14
513
71
21
27 b
U
I
0
1/
9S33
641
1452
4U7
201
Trans Laoor Farr
13.d
44
094
U
635
b5
15722/
7
1284
40/
111e
16
21
1971
1b1
144
J
891
347
83b
523
0
317
23
122
ild
Cratt Upers Irans Labor Farr FLa
PrS
187
21
44
0
14
21
48
26
0
23
0
2139
4
0
106
serv
PHHS
1974
248
201
72
513
278
222
105
86
0
0
39
9473
lob
1341
122
0
0
0
0
1
0
9
0
0
0
0
Missing
Data
1744
299
0
1
2
3
4
5
6
7
8
9
10
11
12
13
Pisdata
85
17
516
30
142
5
30
0
0
28
317
23
7375
lb
789
b9
row label
ffactill.edaz
Mana
1264
244
L 144
jol1
52
1U69
ut
cuts
o91
120
251
0
L.
392
220
290
Sales
1982-83
516
2/*/,/
30
17Oc4
c
45
0
636
!3
290
3
72
23b
507
64
0
U
52
0
U
0
(J
0
0
PrSer
220
19
0
0
2
45
45
247
42
62
0
U
15
0
3
0
0
47
0
49
9
0
0
5
30
0
22
62
0
0
0
0
0
5490
4530
2464
950
104
0
71
0
0
20
4b
12b
0
222
0
142
105
139t4
158
FaLab
55
86
134
23
0
Serv
PHHW
1973
462
544
100
253
105
179
49
49
Mdata
2a
0
16
317
392
0
32
0
97
0
0
2
0
0
0
0
106
0
23
3757
2035
1069
503
1592
1318
1320
396
578
92
117
106
1341
89
0
2594
13207
1418
14313
39
0
0
0
4
row label
outs
Mana
Prof
Sales
Clerc
Craft
Opers
Irans
Labor
Farm
FLa
PrS
serv
PHHS
Misdata
colsum