Colorado workforce 2000

Material Information

Colorado workforce 2000
Blackwood, Christopher K
Publication Date:
Physical Description:
ix, 76 leaves : illustrations ; 28 cm

Thesis/Dissertation Information

Master's ( Master of Arts)
Degree Grantor:
University of Colorado Denver
Degree Divisions:
Department of Economics, CU Denver
Degree Disciplines:


Subjects / Keywords:
Labor supply -- Colorado ( lcsh )
Employment forecasting -- Colorado ( lcsh )
Economic forecasting -- Colorado ( lcsh )
Economic forecasting ( fast )
Employment forecasting ( fast )
Labor supply ( fast )
Colorado ( fast )
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )


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

Record Information

Source Institution:
University of Colorado Denver
Holding Location:
Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
21114901 ( OCLC )
LD1190.L53 1988m .B52 ( lcc )

Full Text
Christopher K. Blackwood
B.A., University of Colorado, 1980
A thesis submitted to the
Faculty of the Graduate School of the
University of Colorado in partial fulfillment
of the requirements for the degree of
Master of Arts
Department of Economics

This thesis for the Master of Arts degree by
Christopher K. Blackwood
has been approved for the
Department of Economics
Date Ikr,. 8 /9M

Blackwood, Christopher K. (M. A., Economics)
Colorado Workforce 2000
Thesis directed by Professor Suzanne W. Helbum
This study addresses many of the labor force, occupational, and
skill requisite issues that will face residents of the state of Colorado over
the next twenty-five years. Forecasts of the national and state economies
were prepared using econometric models. A Colorado specific
demographic model was linked into the economic model to provide an in-
ternally consistent state population and migration projection. Next,
growth forecasts were generated for approximately 900 occupations.
Finally, a projection was made of the skill requisites of 37 occupational
categories and for the state as a whole.
The results of this study show that the U.S. economy will continue
to expand into the twenty-first century however, at rates below the his-
torical average for the past twenty-five years. The Colorado economy also
is expected to register lower economic growth rates.
The expanding Colorado workforce will be older, more feminine,
and less Anglo. In 1988, the average age of the Colorado workforce was
34.9 years. It is forecast to rise to 38.6 years by 2000. Female labor force
participation rates are predicted to rise from 61.1 percent in 1988, to 66.1
by 2000. Colorado males are forecast to experience a declining participa-
tion rate. The minority proportion of the workforce is forecast to increase
from 1988s 15.3 percent to 19.0 percent by 2010.

Net migration will account for a lower percentage of the states
labor force needs. The forecast calls for net migration to average 4,000
per year from 1988 to 1993, and to average 14,000 per year from 1993 to
2000, in sharp contrast to the annual net migration of 40,000 recorded
during the 1970s.
The skill requisites of the workforce are not expected to dramati-
cally change for the next twelve to twenty-two years. However, more
workers at both the very low and the very high ends of the skill spectrum
will be needed to fill the new jobs, a by-product of the acceleration of a
service-oriented economy.
The form and content of this abstract are approved. I recommend its
Faculty member in charge of thesis

Dedicated to
Mom, Dad, Laurie and Nicole

I. INTRODUCTION................................... 1
Preface....................................... 1
Study Outline................................. 2
Study Origin.................................. 4
n. METHODOLOGY.................................... 7
Introduction.................................. 7
Model Development and Selection............... 8
Forecast Development......................... 12
Data Generation Process ..................... 16
m. U.S. ECONOMIC FORECAST........................ 17
Overview .................................... 17
Forecast Detail ............................. 18
IV. COLORADO ECONOMIC FORECAST ................... 28
Introduction................................. 28
Historical Review............................ 28
Overview ................................... 30
Forecast Detail ............................. 31
V. WORKFORCE DEMOGRAPHICS........................ 40
Growth Forecast.............................. 40
An Aging Workforce .......................... 41
The Feminization of Colorados Workforce ... 43

Increasing Minority Representation............ 43
Implications.................................. 44
VI. OCCUPATIONAL PROJECTIONS ...................... 48
Introduction.................................. 48
Analysis...................................... 50
VE. PROJECTED SKILL LEVELS ........................ 54
Introduction.................................. 54
Findings...................................... 54
Comparison to National Trends................. 57
VIE. ALTERNATIVE FORECASTS.......................... 59
Introduction.................................. 59
Low Growth Forecast........................... 59
High Growth Forecast......................... 60
IX. RISKS TO THE FORECAST.......................... 66
Introduction.................................. 66
Industry Forecasts............................ 67
Occupational Projections ..................... 68
Conclusions................................... 69
X. CONCLUSIONS.................................... 72
BIBLIOGRAPHY.................................... 75

1. U.S. Economic Forecast.............................. 26
2. Colorado Economic Forecast
Medium Growth Scenario.............................. 38
3. Colorado Labor Force Participation Rates............ 46
4. Colorado Labor Force Participation Rates
by Ethnic Group..................................... 47
5. Total Employment by Occupational Group........ 52
6. Change in Employment by Occupational Group . 53
7. Colorado Economic Forecast
Low Growth Scenario................................. 62
8. Colorado Economic Forecast
High Growth Scenario................................ 64
9. Industry Employment Projection Error
20 State Sample .................................... 71

1. Data Generation Process

Preface As Colorado moves into the 1990s and approaches the
twenty-first century, many facets of economic life will change. It is like-
ly that a number of the products and services in existence today will be-
come obsolete by the year 2000. Conversely, many technological
innovations will create a vast array of new consumer and industrial goods
and services. Colorados population and workforce are likely to change
as well. Migration will contribute less to the states population growth
than in the past resulting in slower population growth rates and in an in-
creasing average age. Similarly, Colorados workforce will grow less
rapidly than in the past and the average worker will be older. It is almost
a certainty that the types of jobs that exist today will be different in twen-
ty years. Unless educators possess an innate sense of these changes, a mis-
match likely will develop between the skills of the workforce and the needs
of the workforce. This study addresses the key labor force, occupational,
and skill requisite issues facing Colorado over the next quarter century.
Using models that incorporate the economy, workforce, employ-
ment sectors, occupational demand, and current skill levels, the future skill
needs of the workforce were determined. By examining how jobs are
created and what type of jobs will exist, this study attempts to uncover
some of the mystery of our working future. Without such an investiga-

tion, Colorado will move blindly into the next century, potentially miss-
ing rich opportunities, and wasting the states most important resource -
its people.
Study Outline The trends, growth patterns, and conclusions con-
cerning the skill requirements of Colorados workforce are largely depend-
ent upon the structure and expansion of the national and state economies,
and upon the changing composition of the states workforce. Hence, both
national and state economic forecasts were prepared to serve as a basis for
employment and demographic forecasts, and ultimately, for occupational
The national and state economic forecasts were prepared using
structural econometric models (see Charter n for more detail). The na-
tional forecast was obtained from the Data Resources Incorporated (DRI)
January 1988 Trend forecast, and the state forecast was developed in con-
junction with the Center for Business and Economic Forecasting Incor-
porated (CBEF). A modeled relationship was used to enable policy makers
to obtain an understanding of the economic environment that will surround
the future job market. Furthermore, such models allow assumptions about
the future to be specified and changed. By changing some of these key as-
sumptions, the resulting economic forecast also shifts. In this way, snap-
shots of our economic future can be examined under different sets of
potential circumstances. This approach should allow policy makers to un-
derstand both the absolute costs and the opportunity costs of undertaking
certain educational and training activities.

The future of the Colorado economy and its workforce was ex-
amined under three economic scenarios; high growth, medium growth, and
low growth. Each of these economic scenarios relied upon the same U.S.
economic forecast. Assumptions were altered with respect to Colorados
relative economic performance in order to generate each of the alternative
This report will focus primarily on the medium growth forecast be-
cause it is deemed to have the highest probability of occurrence. Similar-
ly, the national forecast used is described as "... the mean of all possible
paths that the economy could follow." 1 Each of the three state forecasts
yielded forecasts of employment for eight major industrial sectors.
Forecasts of net migration, total population and the age/gender/ethnic com-
position of the workforce were derived by linking the economic forecast
with the States cohort-component population projection model.
Colorado specific occupational forecasts were prepared by utiliz-
ing a computerized occupational model developed by the Utah Department
of Employment Security in conjunction with two federal agencies (Inter-
state Conference of Employment Security and the National Occupational
Information Coordinating Committee). This model allowed a Colorado
specific occupational structure to be estimated and forecast for ap-
proximately 900 occupations.
1 Data Resources Incorporated, U.S. Long Term Review, Winter
1987-88, p. 13.

The expectation that certain occupations in existence today will dis-
appear in the future, and that new professions will surface that previously
had not existed, present the major risks in forecasting occupational change.
This risk was partially offset by the application of "change factors" to the
occupational forecast. Change factors are adjustments made to certain oc-
cupation patterns given an expected change in technology. For example,
many computer and computer-related occupations were assumed to ex-
pand at faster rates, reflecting the expectation that most industries will re-
quire workers with more computer skills.
Finally, an assessment of the skill level of the workforce was
prepared. This analysis attempted to measure the skill level of workers in
the economy using a methodology developed at the Hudson Institute (an
Indiana research firm) that estimated the skill level of workers in 37 oc-
cupational groups. Calculating the skill level of the entire workforce then
became a simple matter of adding up the skill levels of the 37 occupation-
al groups.
Study Origin This thesis was prepared in response to a specific
research need of the State of Colorado. The Colorado Department of Local
Affairs (DOLA) and the Colorado Department of Labor and Employment
(DLE) jointly sought to produce a report that would uncover and clarify
major labor force and educational issues facing the state government. In
February 1988, the State contracted with me to prepare such a study. That
report was entitled Colorado Workforce 2000. Not coincidentally, this
thesis bears the same name as that report.

Much of the information contained within this thesis was directly
generated from the report I prepared for the State. Since this thesis was
generated almost entirely from my work stemming from the contract with
the State, it may be helpful to provide an explanation of my role in generat-
ing the Colorado Workforce 2000 study. First, the general study outline
was prepared in conjunction with the Deputy Director of the Colorado
Department of Local Affairs (DOLA), Bart Alexander. Upon outlining
the elements to be addressed in the study I proceeded to obtain the resour-
ces needed to complete the study. The primary nongovernmental resour-
ces used in the study was a computer-based econometric model of the
Colorado economy. I directed the process of reviewing and selecting the
state economic forecasting model to be used in the study. In addition, I
directed the effort to prepare each of the alternative scenario forecasts to
the extent of selecting the general assumptions. After the state economic
forecasts were produced, the next step of the process involved generating
demographic forecasts using the States cohort-component population
projection model. The process of solving many of the technical problems
related to linking the economic model to the States demographic model
was carried out almost entirely by DOLA staff demographers and CBEF.
My role was to oversee the process. Upon generation of an internally con-
sistent economic and demographic forecast, the next step involved prepar-
ing a statewide employment forecast at the two-digit Standard Industrial
Code (SIC) level. This task was completed wholly by me. After the
employment forecasts were generated at the two-digit level, the occupa-
tional model could be run. All of the data entry, computer work, and

forecasting of occupational growth patterns was completed by myself with
technical assistance provided by staff of the Utah Department of Employ-
ment Security. With growth forecasts generated for 900 occupations, the
focus shifted to generating skill level forecasts. The skill level forecasts
were prepared by adapting a methodology employed by the Hudson In-
stitute in their WORKFORCE 2000 report. This portion of the study was
completed jointly with Bart Alexander. The generation of skill level
forecast completed to data generation part of the study.
The remainder of the study included analyzing data, writing up
results, and producing the report. This was done solely by myself. The
lone exception was Chapter V, Workforce Demographics, which was
originally drafted by Reid Reynolds, Colorado state demographer. Reid
Reynolds and Bart Alexander reviewed drafts of the study in various stages
of completion. Certain changes to the analysis were made based on their
comments. Finally, Tom Cladis edited the final draft for grammatical er-

Introduction The general goal of this study was to examine labor
force trends in the state of Colorado and assess the implications for oc-
cupational and skill levels demands. Since this study was prepared in
response to a specific research need of the State of Colorado, an addition-
al objective was to ensure that the study could be replicated in the future.
As a result, each stage of the research project was undertaken using
methods and resources that the State would likely have access to in future
Generally, this study sought to investigate labor force trends and
related issues within a modeled framework. In this context, the models
used were computerized representations of reality models of the
economy, occupational structure and skill demands. By conducting this
study in a modeled framework, policy makers will be better equipped to
answer "what if' questions concerning the economy. Moreover, by ex-
plicitly stating the assumptions that generate each forecast, a higher level
of understanding can be garnered regarding the factors that most affect the
labor force, occupational demand, and skill level of the workforce.
The sections that follow provide detailed descriptions of the reasons
why certain models were used, how they were selected, and the way in
which each specific forecast was produced.

Model Development and Selection This report relied on struc-
tural econometric models of the economy for the United States and
Colorado. Many econometric models of the U.S. economy were available.
The list of potential vendors providing U.S. economic models and forecasts
was narrowed down to two, Data Resources Incorporated and Wharton
Econometric Forecasting Associates. The models used by these firms are
generally similar in that both are neo-Keynesian econometric structured
models. The two firms listed above were selected as "finalists" because
they represented the only two national firms whose forecast used by local
organizations as exogenous inputs to Colorado economic models. In other
words, the list of state economic models proved to be the limiting factor
in paring down the list of national models to be used.
The two Colorado economic models that I examined for use in this
study were developed by: 1) the Center for Business and Economic
Forecasting Incorporated (CBEF), and 2) the Center for Economic
Analysis (CEA) of the University of Colorado at Boulder. Each of the
state models provided by these organizations contained properties that
presented advantages over the other.
CBEF is a commercial research and forecasting firm in existence
since 1982. It uses the Data Resources Incorporated U.S. economic
forecast as an input to their state model. The CBEF model contains ap-
proximately 150 equations including identities. This model is structured
as a quarterly econometric model capable of producing forecasts to the
year 2010. This model is designed to produce economic forecasts rather
than simulations. In this sense, simulation refers to the ability to easily

change assumptions on key policy levers and generate internally consis-
tent economic results. Generally, a forecasting model has a tendency to
"blow up" if policy variables are adjusted incorrectly during a simulation
process. A forecasting model requires greater user knowledge of its struc-
ture to produce policy simulations. I have used this model in other studies
over the past five years and having a working knowledge of its structure.
CEA is a research institute located in the Economic Department at
the University of Colorado at Boulder. It provides research and forecast-
ing services to industry and government. The national forecast used as an
input the CEA state model is provided by Wharton Econometrics Forecast-
ing Associates. The CEA model is not a pure econometric model. It
employs a more eclectic modeling approach using vector autoregressive
techniques, transfer functions, and econometric methods. It is more
specifically designed to allow the user to simulate policy changes on the
economy. The CEA model contained more sectoral detail than the CBEF
model. For example, the CEA model produces forecasts of employment
at the two-digit level for all categories in the state of Colorado (ap-
proximately 74). In comparison, the CBEF model produced employment
forecasts at the two-digit level for 29 categories. The CEA model also
contained a direct linkage into an occupational model for the state of
Colorado. Unfortunately, the occupational model had not been regularly
maintained and therefore was of little use. Documentation of this model
was limited.
I selected the CBEF model over the CEA modelfor use in this study.
The primary reason for selecting the CBEF model was threefold: 1) my

higher level of familiarity with the CBEF model and 2) the longer term ex-
istence of the CBEF as a commercial vendor (which increased my con-
fidence that the model will be available in the future for study replication),
and 3) the more complete documentation of the CBEF model. Cost was
not an issue in the decision. Both CBEF and CEA would have provided
the specified forecasting services for approximately $3,000. The existence
of an occupational model linkage to the CEA model was an attractive fea-
ture. Unfortunately, this model used a generic occupational matrix for the
State of Colorado that had not been updated in over ten years. This voided
the use of the occupational feature in the CEA model because the State of
Colorado had access to a more recently developed occupational matrix.
Another key goal in this study was to directly integrate a state
economic forecast into the State of Colorados cohort-component popula-
tion projection model. In the past, the state demographer has prepared his
population and migration forecast without having a direct linkage to an
economic forecast. For example, migration was usually forecast by simp-
ly using the past ten year moving average of migration as a proxy for the
forecast of future migration levels. Considering that many studies have
shown that migration is a function of economic variables, an economic
based linkage was determined to be desirable.
The choice of occupational model was limited. Some commercial
venders provided occupational models for the State of Colorado. But,
given the limited resources of the State, the choice was narrowed to a model
that could be used at no cost to the State. The model utilized was the In-

dustry/Occupational Microcomputer Matrix System developed by the
Utah Department of Employment Security.
Generally, the occupational model applies a historically determined
staffing pattern to employment forecasts at the two-digit SIC code level in
order to determine the occupational projections. These staffing patterns
were developed using specific State of Colorado staffing pattern data
culled from the 1980 Census. Once the historical Colorado staffing pat-
tern had been determined, "change factors" developed by the U.S. Bureau
of Labor Statistics served to modify the pattern for future years. Change
factors account for projected changes in technology and for the disap-
pearance of certain occupations in the future. The model generated
forecasts for over 900 occupations in Colorado. A detailed listing of these
occupations is shown in the Appendix I of Colorado Workforce 2000.
Until they are applied to the skill levels and educational require-
ments of tomorrows workforce, occupational forecasts are of little use.
Consequently, a methodology was developed to estimate the required skill
level of the Colorado labor force in order to analyze the impacts of the
shifts in occupational structure and growth, and to determine what those
shifts portend for educational policy. Ideally, this analysis would assign
some type of quantitative or qualitative score to each of the 900 occupa-
tional groups, and sum the scores to derive a requisite total skill attainment
level. While this methodology was in theory very desirable, in practice it
proved to be a difficult task.
This study adopted a more streamlined approach of skill level es-
timation by adapting a methodology developed at the Hudson Institute for

their WORKFORCE 2000 study. First, the 900 occupations were
regrouped into 37 categories designed to reflect similar groupings in the
Hudson Institute study. The decision to collapse the data allowed for quick
model design and for the desired comparison of state and national skill
A skill level score was assigned to each of the 37 occupational
categories. The score assignments were taken directly from research data
provided by the Hudson Institute. For each skill category, subscores were
assigned for math and language on a 1 to 6 scale, with 1 reflecting a lower
skill level and 6 a higher level. Thus, the lowest possible total score as-
signment was two (1 for math plus 1 for language), and the highest twelve
(6 for math and 6 for language). In this study, the lowest group, hel-
per/laborers, was assigned a score of 2.3 and the highest group obtained a
score of 11.2.
Finally, each score was multiplied by each occupational groups
percent of total employment. Hence, the total skill score for the labor force
represents a weighted sum of the occupational group skill attainment
levels. As the percentage of employment increases or decreases in any oc-
cupational group, the relative weights for each group necessarily will
change. Thus, the total skill score of the economy shifts as the occupa-
tional structure of the labor force is altered.
Forecast Development Since the CBEF model was selected for
this study, the DRI forecast of the U.S. economy was automatically used.
The national economic variables used for each scenario in this study were
taken directly from DRI Long Term Review, Winter 1987-88. The DRI

U.S. economic model is a quarterly econometric model, with the capacity
of producing either a trend or cyclical forecast. The trend forecast version
of this model was used in this study to project all macroeconomic vari-
ables. The trend forecast was selected because this study emphasizes long
term trends rather than short term shifts in labor market conditions. There-
fore, no cyclical movement in economic conditions is shown in the forecast
except for the 1988 to 1990 period. The 1988 to 1990 portion of the
forecast was prepared to be consistent with the short-term forecast of U.S.
economy which does include a cyclical pattern of economic growth.
The Colorado economic forecast was prepared by the CBEF under
the direction of the Colorado Department of Local Affairs (DOLA), and
the Colorado Department of Labor and Employment (DLE). My role in
the forecast preparation was to determine the assumptions that would be
used to drive each forecast scenario. A meeting was convened with
economists and demographers and from DLE, DOLA, and CBEF to deter-
mine what assumptions would be used to drive each forecast scenario. A
number of forecast runs were made for each scenario until internal forecast
inconsistencies were solved.
The CBEF model produced forecasts for many variables but some
had to be estimated using other means. The model produced forecasts of
nonagricultural employment but not for self-employed individuals. The
self-employed group was therefore assumed to grow at the same rate as
the expansion in the labor force, a variable that was forecast by CBEF. In
addition, no estimate was made of agricultural employment because a reli-
able Colorado specific staffing pattern was unavailable for that category.

Upon generating state economic forecasts, the next step was to link
the forecast results into the States demographic model. The demographic
model produces forecasts of population by age and by sex. By applying a
labor force participation rate to each population cohort, an estimate of total
labor force could be implicitly derived. The CBEF economic model also
produced a forecast of the labor force. The first computer runs of the
demographic model generated implied labor force projections that were
inconsistent with the CBEF labor force forecast. A number of forecast ad-
justments, and computer runs, were made to produce internally consistent
labor force forecasts from the CBEF model and the States demographic
Detailed occupational projections were obtained from an occupa-
tional model that factored in all two-digit employment categories of the
Standard Industrial Code (SIC). The employment forecasts used in the oc-
cupational model were partially supplied by CBEF. Specifically, the
CBEF model of the Colorado economy provided forecasts at the two-digit
level for 29 of 74 employment sectors. The remaining forecasts of employ-
ment at the two-digit level were arrived at by estimation. Because employ-
ment forecasts were available for all major employment divisions
(commonly referred to as the one-digit level), a target figure was defined
for the two-digit forecasts within each one-digit sector. For example, in
the mining division, CBEF forecasts were available at the two-digit level
for three of the four employment classifications (SIC 10,13, and 14; no
employment in SIC 11 exists in Colorado). In this case, the calculation of
the final two-digit SIC employment level (SIC 12) becomes merely a mat-

ter of subtracting the CBEF forecast for SIC 10,13, and 14 from the min-
ing division total. Viewed mathematically, the mining sector employment
at the two-digit level was calculated by using the following equation:
Mining Division Total (SIC 10 + SIC 13 + SIC 14) =SIC 12
Other employment sectors used a more complicated, but
straightforward methodology. In the manufacturing sector employment
forecasts were available for both the nondurable and the durable goods
subsectors. Within the durable goods subsector, employment forecasts
were available at the two-digit level for 7 of 15 two-digit categories. To
derive a forecast for the remaining two-digit categories, the employment
totals of the known two-digit forecasts were subtracted from the durable
goods employment total. Thus, if total durable goods employment equaled
100, and the known two-digit sector forecasts totaled 60, the unknown sec-
tor must total 40. For the remaining 8 two-digit employment sectors, the
forecast amount was distributed on a pro rata basis according historical
distribution. Once employment forecasts were prepared for each two-digit
employment category, the occupational model could be run.
Upon completion of the employment forecasts, preparation of the
occupational forecasts was simply a matter of running the appropriate
programs in the microcomputer Matrix System. Forecasts of occupation-
al growth were generated for each economic scenario, low, medium, and
high. In addition, forecasts were prepared for three target years (1993,
2000, and 2010) rather than for each year. The target years were decided
upon by myself and officials at the DOLA.


Overview Generally, the national forecast assumes that the
economy will experience a smooth path of expansion over the next twen-
ty-five years. While it is likely that the growth of the U.S. economy will
be affected by economic cycles, it is also virtually impossible to forecast
the timing of such events in a long range forecast.
Key U.S. forecast assumptions are as follows:
Real growth over the forecast horizon is slower than that of the past
twenty-five years. Real GNP grows at an annual rate of 2.3% from
1988 through 2010, compared with 3.1% from 1962 to 1987.
Inflation is forecast to accelerate relative to its recent pace, but it is not
predicted to reach the levels attained in the late 1970s. The CPI is
projected to increase at an annual rate of 5.3% Over the next twenty-
five years.
The federal budget deficit continues to remain large in nominal terms,
falling to $115 billion in 1991, before rising to $259 billion by 2010.
More importantly, the budget deficit decreases as a percent of GNP
from 3.5% in 1988 to approximately 1% in 2010.
Energy prices rise faster than overall inflation, but major supply dis-
ruptions are not predicted on the same scale as in either 1973 or 1979.

The labor force grows 1 % slower than the post-war period. From 1962
through 1987, labor force growth averaged 2.1% per year. In the
forecast period, labor force gains are expected to average only 1% per
year. Because of the dwindling labor supply, the unemployment rate
should average 5.9% over the forecast horizon, a level approximately
equal to that of the 1960s. Labor productivity gains improve as an
older workforce increases efficiency.
The population projection is consistent with the U.S. Census Bureaus
middle growth series. Fertility rates are expected to stabilize at around
1.9, slightly above the current 1.8 rate. Mortality rates will continue
to decline. Net immigration is assumed to average 450,000 annually.
Overall population growth averages 0.7 % from 1988 to 2010, amarked
deceleration from the 1.1% average since 1962.
Forecast Detail From 1988 to 2010, the U.S. economy should ex-
perience slower economic growth than in the past twenty-five years.
Reflecting primarily demographic factors, the labor supply side of the
forecast dictates this slow down. And with slower labor force growth,
potential output of the economy will be reduced. As a result, both con-
sumption and housing market growth are projected to decelerate in the fu-
Because the long-term forecast is affected primarily by
demographic factors which are already set in place, it is unlikely that the
long-term labor force growth will be affected significantly by economic
shocks. Structural shifts in the economy would have to occur to change
the general growth rate of many of the production aggregates. Such shifts

would include large changes in
labor force participation rates and
in net immigration from abroad.
Total population in the
U.S. is predicted to increase from
244 million in 1987, to 269million
by 2000, and to 284 million by
2010. As the baby boom genera-
tion passes through its child bear-
ing years, births will peak at 3.9 million in 1990, posting the highest level
since the 1960s.
Similarly, the prime home-buying group, those individuals aged
25-34, peaked in 1987 at 17.9 percent of the population, and will fall to
12.7 percent by 2006. Accompanying the smaller home-buying group will
be a lower level of housing activity. As a result, housing starts will average
1.5 to 1.6 million units per year through 2010. In comparison, the 1970s
witnessed an average of nearly 2 million housing starts a year.
The labor force participation rate the ratio between the labor force
and the civilian adult population represents another key element to the
long-term growth prospects in the U.S. economy. The labor force par-
ticipation rate remained essentially unchanged from 1948 through 1963,
with declines in male participation rates offset by increases in female par-
ticipation rates. However, from 1963 to 1987, the overall participation rate
rose from 57.6% to 64.6%. Even more astounding, the female participa-
tion grew from 38% in 1963, to 56% in 1987. By 1988, it is estimated that
U.S. Population Growth
Annual Percent Change
1 0% : \S ' sVisS:
I.UAa^Nswx M. sss s- Os**ss\ wsn<-ss \ %'

mS ?
70-75 75-80 80-65 65-88 88-93 93-00 00-10

the overall participation rate will equal 65%. From 1988 to 2010, par-
ticipation is predicted to increase at a slower rate than it has over the past
quarter-century, rising to 67% by 2000, and to 68% in 2010.
For the past twenty-five years, labor force growth has averaged
2.1% per year, paced by the 1970s annual growth rate of 2.7%. While the
strong growth experienced in the 1970s can be attributed both to the entry
of baby-boomers into the workforce and to the higher female participation
rates, these sources of faster-than-average growth in the labor force will
not sustain similar expansions in the future. First, increases in participa-
tion rates are forecast to taper off. Second, growth in the adult population
will slow the number of people turning 18 in 1994,3.2 million, will be
below the 1979 level by 1.1 million. Thus, during the next twenty-five
years, the labor force should grow at only half the rate that it did during
the previous quarter-century.
A potential benefit of the slowing labor force growth is the stabiliza-
tion of the unemployment rate close to its current "natural" rate of 6%, as
the predicted slowing growth rate is accompanied by an acceleration in the
demand for labor at historical rates.
On the other hand, slower labor force growth and reduced un-
employment could cause inflationary pressures to mount. Hence, DRI
states that "...the projection is not so much a reflection of the economys

ability to absorb labor force growth as it is an assessment of what con-
stitutes a noninflationary degree of slack in the labor market."1
In terms of employment growth projections, only about half of the
growth experienced over the past twenty five years will be achieved U.S. Employment Growth Annual Percent Rate
over the next twenty-five years. For example, from 1988 through 3.5%|/1
2.5% A BSk
2010, the cumulative growth in 2.0% 4
employment is expected to total 1.5% '* vk. 1 0% ^
26%, or 27.4 million jobs. This 0.5%t |
compares with an estimated total growth of 60%, or 40.5 million
70-75 75-80 8045 8548 8843 9340 00-10 Year
jobs added from 1966 through 1988, a period of similar duration.
Over the forecast interval, the economy will add most of its jobs in
the service sector. The manufacturing share of total nonagricultural
employment will continue to dwindle as it has for most of the post-war
period, falling from 35% in 1959, to 19% in 1988, and to 14% by 2010.
Most importantly, this decline in employment will be due primarily to
higher productivity growth rather than to a contracted demand for
manufactured goods.
1 Data Resources Incorporated, U.S. Long Term Review, Winter
1987-88, p. 15.

On the trade front, the forecast assumes that the merchandise trade
deficit will continue to improve, especially during the next few years.
Much of the recent decline in the deficit can be traced to normal, if not
delayed, J-curve effects ("J-curve" refers to a graphical depiction of the
relationship between currency price changes and the impact on trade
deficit). In fact, the exchange value of the dollar (Federal Reserve Board
index) is forecast to drop 12% in 1988, and another 6% in 1989. By the
1990s, an equilibrium situation will be at hand and the dollars decline will
slow. While this analysis indicates that the trade situation will improve
sharply in the next few years, the improvement will be tempered in the
mid-1990s when the dollar once again appreciates. In nominal terms, the
merchandise trade balance is anticipated to improve from a deficit of $170
billion in 1987, to a shortfall of only $ 115 billion in 1990. Over the remain-
ing forecast years, the trade deficit should continue to expand, totaling
$165 billion in 2000, and $318 billion in 2010. However, the trade deficit
will continue to improve as a percent of GNP from -3.8% in 1988, to -1.3%
in 2010.
Although fiscal austerity will continue to guide the federal budget-
ing process for the next few years, it is assumed that Congress will be un-
able to attain the Gramm-Rudman-Hollings budget deficit targets in fiscal
year 1989 and beyond. The DRI forecast calls for the federal government
budget deficit to fall from $169 billion in 1988, to $137 billion in 2000,
before rising to $259 billion in 2010. When these budget deficits are
analyzed in the context of a growing U.S. economy, improvement is

revealed. As a percent of GNP, the deficit falls from the historical high of
5.2% in 1984 to 4.5% in 1988, to 1.4% in 2000, and to 1.1% in 2010.
In the next quarter-century, industrial output is expected to grow at
an annual rate of 2.7%, one percentage point less than the previous twen-
ty-five years. The slowdown in industrial growth is expected to permeate
each industrial sector. Yet, the relative growth rates in each sector are not
expected to change. For example, rubber and plastics products, electrical
machinery, and instruments were three of the fastest growing sectors
during the past twenty-five years, and they are predicted to duplicate their
performance in the future. Similarly, the laggards in the economy such as
leather goods, tobacco products, and primary metal products will continue
to perform weakly in the course of the next twenty-five years. Despite in-
dustrial output slowing on a relative basis when comparing the following
twenty-five years with the previous twenty-five, gains will be on a par with
the growth that the U.S. industrial sector has experienced in the past four-
teen years.
The long-term outlook
calls for relatively low inflation
over the forecast period, largely at-
tributable to more stable energy
prices and to moderate wage in-
creases. The moderation of infla-
tion in recent years, coupled with
continued efforts by the Federal
Reserve to control inflation, has
Consumer Price Index
Annual Percent Change__
70-75 75-80 80-85 85-88 88-93 93-00 00-10

dropped the long term forecast for the Consumer Price Index from the 10%
range in the 1970s and early 1980s to amore palatable 5% over the forecast
interval. It must be stressed that a great deal of uncertainty surrounds the
long-term inflation forecast. In fact, DRI indicates that the forecast
bandwidth between its optimistic and pessimistic forecasts is three times
as wide as the range for real GNP growth rates.
The outlook for the financial markets is closely linked to the infla-
tion forecast. Specifically, real interest rates are forecast to remain above
their historical average before the pre-August 1979 average values (the
Federal Reserve changed operating policies at that time, introducing more
volatility into interest rates). The change in Fed policy raised the cost of
funds to institutions. Therefore, to account for the higher cost of funds, a
larger real interest rate premium is implied by the interest rate forecast, hi
terms of key interest rates, the 30 year Treasury bond is predicted to fluc-
tuate in the 8.5% to 9.5% range from 1988 through 2010. At the shorter
end of the maturity spectrum, the federal funds rate is forecast to hold at
around 7.2% throughout the forecast period. For the next three to five
years, oil prices are expected to tread at current price levels. Beyond 1993,
with worldwide demand increasing, OPEC should regain some of its car-
tel pricing power. According to the DRI forecast, "significant energy in-
flationis likely to return by the mid-1990s." The trend projection assumes
2 Data Resources Incorporated, U.S. Long Term Review, Winter
1987-88, p. 23.

that oil prices rise from $25 a barrel in 1990, to $33 a barrel in 1995, and
to more than $100 a barrel by 2006. Real prices increase an average of
5.1% per year over the forecast horizon. Interestingly, although energy
consumption is projected to continue to rise, by 2010 the amount of ener-
gy consumed per real dollar of GNP will be only half of its 1970 peak.
Historical investment conditions have been significantly impacted
by federal tax law changes. The Economic Recovery Tax Act of 1981
lowered the cost of investment, and, as a result, investment expenditures
boomed from 1983 through 1985. However, because the Tax Reform Act
of 1986 eliminated many of these tax incentives, investment spending in
1989 is forecast to fall to its lowest share of GNP since 1964. In the lat-
ter years of the forecast, investment spending is predicted to rebound as
the nations capital stock ages and needs replacement.

1970 1975 1980 1985 1988 1993 2000 2010
Gross National Product
Real GNP ($Billion-1982) 2,416 2,695 3,187 3,607 3,887 4,391 5,133 6,419
Annual Growth Rate (%) 2.2 3.4 2.5 2.5 2.5 2.3 2.3
Nominal GNP ($Billions) 1,015 1,598 2,732 4,010 4,708 6,629 11,225 23,140
Annual Growth Rate (%) 9.5 11.3 8.0 5.5 7.1 7.8 7.5
Prices and Wages
Implicit Price Deflator (1982=100) 42.0 59.3 85.7 111.2 121.1 150.9 218.7 360.5
Annual Growth Rate (%) 7.1 7.6 5.3 2.9 4.5 5.4 5.1
Con. Price Index (82-84=100) 38.8 53.8 82.4 107.5 118.3 150.5 219.8 369.6
Annual Growth Rate (%) 6.8 8.9 5.5 3.2 4.9 5.6 5.3
Production and Other Key Measures
Industrial Production (1977=100) 78.5 84.9 108.6 123.8 132.2 153.3 184.6 239.4
Annual Growth Rate (%) 1.6 5.0 2.7 2.2 3.0 2.7 2.6
Retail Units Car Sales (Mil Units) 8.4 8.5 9.0 11.0 9.8 10.5 11.5 12.4
Federal Budget Surplus ($Bil) -12.4 -69.4 -61.3 -196.0 -168.8 -136.4 -137.1 -258.7
Deficit/Nominal GNP (%) -1.2 -4.3 -2.2 -4.9 -3.6 -2.1 -1.2 -1.1
Labor Force
Labor Force (Millions) 82.8 93.8 107.0 115.5 121.8 129.3 139.2 152.4
Annual Growth Rate (%) 2.5 2.7 1.5 1.8 1.2 1.1 0.9
Total Emp.-HH Survey (Millions) 78.7 85.8 99.3 107.2 114.5 122.3 130.9 142.8
Annual Growth Rate (%) 1.7 3.0 1.5 2.2 1.3 1.0 0.9
Unemployment Rate (%) 6.0 8.5 7.2 7.2 5.8 5.3 5.9 6.2
Non-Ag. Employment (Millions) 70.9 77.0 90.4 97.6 104.2 112.2 120.2 131.6
Annual Growth Rate (%) 1.7 3.3 1.5 2.2 1.5 1.0 0.9
Labor Force Participation Rate (%) 59.1 60.4 62.8 63.6 60.4 62.1 67.3 67.8
Total Population (Millions) 205.1 216.0 227.7 239.9 246.2 256.7 268.9 284.3
Annual Growth Rate (%) 1.0 1.1 1.0 0.9 0.8 0.7 0.6
Population Age 18-64 (Millions) 115.2 126.1 138.4 147.8 151.9 157.3 166.4 180.1
Annual Growth Rate (%) 1.8 1.9 1.3 0.9 0.7 0.8 0.8
Foreign Trade
Current Account Balance ($ Bil) 2.3 18.3 1.9 -116.4 -138.9 -105.5 -102.5 -149.6
Merchandise Trade Balance ($ Bil) 3.7 10.6 -22.4 -120.2 -146.8 -126.1 -164.6 -318.1
Trade Deficit/Nominal GNP (%) 0.4 0.7 -0.8 -3.0 -3.1 -1.9 -1.5 -1.4

TABLE 1 (continued)
1970 1975 1980 1985 1988 1993 2000 2010
Financial Markets
M2 Money Supply ($ Billions) 623 1,014 1,627 2,551 3,086 4,390 7,429 15,235
Annual Growth Rate (%) 10.2 9.9 9.4 6.6 7.3 7.8 7.4
30 year Treasury Bond (%) NA NA 11.3 10.8 8.2 9.0 8.6 8.3
Treasury Bill Rate (%) 6.4 5.8 11.4 7.5 5.2 6.3 6.1 6.0
Federal Funds Rate (%) 7.2 5.8 13.4 8.1 5.7 7.2 7.2 7.2
Personal Income ($ Billions) 832 1,313 2,258 3,327 3,965 5,604 9,436 19,458
Annual Growth Rate (%) 9.6 11.5 8.1 6.0 7.2 7.7 7.5
Savings Rate 8.1 9.1 7.1 4.5 4.5 5.4 6.1 6.2
Profits After Tax ($ Billions) 41.7 83.0 152.3 128.1 123.2 180.6 293.8 578.0
Annual Growth Rate (%) 14.8 12.9 -3.4 -1.3 7.9 7.2 7.0
Note: Annual growth rates calculated on a point-to-point basis. For example, the annual growth rate for Real GNP
from 1970 to 1975 reflects the annual compounded growth rate over that five year time frame. This growth rate is
displayed under the 1975 column heading. This method of calculating and displaying annual growth rates is used
throughout this study.
Data source: Data Resources Incorporated

Introduction The outlook for the Colorado economy can be
divided into two stages: 1) the recession recovery period (1988 to 1993),
and 2) the historical advantage period (1993 2010). For the next five
years, Colorado will be saddled with the task of recovering from its 1985-
86 recession. Since recovery in a recessionary environment does not occur
overnight, Colorado is expected to grow at a rate below its historical
average compared to rest of the country (in terms of employment). But,
the state is forecast to match the average national growth rate during the
next five years. And, once Colorado gets past this relatively weak growth
environment, it is predicted to regain again its relative prosperity.
Historical Review How did Colorado get into its weak economic
position? The answers are uncovered by reviewing the states recent
economic history. From 1973 to 1981, the Colorado economy prospered
as energy prices rose and new firms located offices in the state. In the
course of those eight years, state employment growth trends paralleled na-
tional patterns but its growth rate was 2% above the national rate. The
Economic Recovery Tax Act of 1981, stimulated a commercial and in-
dustrial building boom in the state. Nonresidential construction contracts
rose to nearly $2 billion in 1981. In conjunction with the construction and
energy booms in the state, the residential housing market accelerated

sharply. As a result of the fast growth in these sectors, nearly all segments
of the state economy prospered. From 1970 through 1984, nonagricultural
employment (a key indicator of growth) grew at an annual rate of 4.7%,
2.5 percentage points above the national average. Retail sales, housing
starts, and personal income similarly reflected the relative strength of the
state economy. As individuals and businesses outside the state witnessed
"the boom", many moved to the state in hopes of participating in the
prosperity. The attraction of Colorado was directly reflected by the num-
ber of net migrants locating in the state. From 1970 to 1984, net migra-
tion into Colorado averaged 40,000 persons a year. And the states
population grew at an average rate of 2.7%, 1.6 percentage points above
the national average.
The relative prosperity of the state economy disappeared in 1985
as the factors responsible for the boom reversed themselves. First, oil
prices fell precipitously in 1985 and 1986. Although, the oil price decline
started in 1981, the drop was slow and overshadowed by fiscal stimulus
from the federal government. Not only did the energy industry have its
problems, but also did the construction and retail sectors begin to suffer.
Vacancy rates for commercial office space in the Denver area neared 30
percent and a glut of single and multi-family homes became evident in
many areas of the state. Total housing permits fell from 50,900 in 1983,
to a meager 18,700 in 1987. When the last round of the federal tax cuts
and incentives ran out in 1986, it became clear that the state economy was
mired in a recession.

The states labor force also took a hit from the recession. In terms
of nonagricultural employment, the state lost jobs in 1985, in 1986, and in
1987. In this time frame, job losses were recorded in the following in-
dustries: mining (14,800), construction (21,100), manufacturing (11,100),
transportation and public utilities (3,100), and wholesale trade (4,300).
Only the services industry and the government added a significant num-
ber of jobs between 1984 and 1987. They recorded gains of 22,500 and
17,100, respectively.
The flow of persons into and out of the state was aptly represented
by the relative loss of the states economic competitiveness. Net migra-
tion of at least 30,000 persons each year was common in the 1970s. Even
the recessionary year of 1975 saw 15,000newpersons (net) make Colorado
their home. By 1984 however, Census Bureau estimates indicated that
migration totalled only 9,000. And 1986 and 1987 both registered migra-
tion declines of -4,000. Although the states population continued to grow
(because births exceeded deaths by nearly 30,000 each year), the net ad-
dition from migration had turned negative for the first time since 1965. In
fact, the 1984-88 period is probably the longest interval of low migration
to the state since World War n.
Overview The Colorado economic forecast was prepared using an
econometric model developed by the Center for Business and Economic
Forecasting. This model directly utilizes data from the DRI U.S. model to
prepare state forecasts. In addition, the forecast was modified so that the
demographic component of the forecast was consistent with, and linked
to, the population projections prepared by the Colorado Department of

Local Affairs. (For more detail on the specific methodology, see Chapter
The medium-growth forecast was prepared by adding to the general
assumptions for the U.S. economy (DRIs January 1988 Trend forecast)
the following Colorado specific assumptions:
Labor force participation rates follow a similar pattern to those of the
nation with female participation continuing to rise (but at slower rates
than during the 1970s).
Net migration is based on the relative growth of the state economy
compared to other states in the U.S.
Colorado is assumed to retain a competitive advantage relative to other
states that will enable the state to attract and to develop manufactur-
ing businesses faster than the national average.
Forecast Detail While history tells us that Colorados economy
generally has grown faster than the U.S., in recent years the state has fal-
len behind in most of its economic aggregates. What are the implications
for the future of the state economy? The forecast indicates that recent
problems plaguing the economy will be rectified, but at a slow rate. Over
the next five years, Colorado is forecast to move back into the plus column
for employment growth; however, the state is not expected to resume its
former role as one of the countrys economic growth leaders. One reason
- energy prices are not projected to experience the sharp price spikes of
1973 or 1979. Nor is a manufacturing boom projected. But, owing to the
states many natural advantages such as climate, transportation, geog-
raphy, and leadership in higher education fields, the state is forecast to at-

tract new business and generate job growth in excess of the national
average. On a positive note, a construction boom could occur with the
development of the new Denver airport. The possibility of this situation
occurring is analyzed in Chapter VII.
In terms of forecast
specifics, the most general in-
dicator of growth, nonagricul-
tural employment, is forecast to
rise at an annual rate of 1.4%
from 1988 to 1993, slightly
below the national average. For
the remainder of the forecast
period, Colorado employment
growth will outstrip that of the rest of the nation by 0.8%, as the state in-
creases employment by 1.9% per year.
The projected rebound in the states economy will be reflected in
all employment sectors. The services sector is forecast to be the growth
leader, expanding at an annual rate of 3.1% from 1988 through 2010.
Wholesale trade employment will record the second fastest growth rate
over the next twenty-two years as the Front Range increases its role as a
distribution center for the Rocky Mountain region. The mining sector will
follow wholesale trade as the third fastest growing sector with an annual
job growth rate of 2.6% from 1988 to 2010, attributable to an expected rise
in oil prices in the mid 1990s. The remaining sectors are forecast to ex-
pand at rates of between 1.1% and 2%. Looking at employment patterns
Employment Comparison

from another perspective, the goods producing industries (mining, con-
struction, manufacturing) are
predicted to increase at an average
annual rate of 1.4%, compared to
a 1.7% services industries growth
(transportation, public utilities,
communications, finance, in-
surance, real estate, services,
wholesale and retail trade,
government). Even though goods
producing firms will continue to grow, service-related firms will produce
85% of all nonagricultural jobs from 1988 to 2000. By 2010, this figure
will expand to 88%.
On the surface the growing dominance of service industries in
Colorado may be cause for concern. But, the Hudson Institute put the ex-
pansion of services into another perspective by stating that:
"... the shift to services is an inevitable and healthy result of steady
productivity gains occurring in manufacturing. Because productivity
gains are easiest to achieve in a controlled, high volume environment
of the factory, the factory is becoming less and less important as a
source of jobs; economic value is moving away from the plant into
other activities. This process is not a threat to societys wealth-
producing potential, it is a reflection of it; it represents a shift in the
economic challenge to new sectors: retailing, health care, education,
government, food service, and other industries. The nations
Goods/Services Employment
Annual Growth Rate 1988 2000

Goods Services
Industry Group

economic future will be written by these industries not by the revival
of manufacturing."1 2
Historically, Colorados population has grown more rapidly than
the nation as a whole because of its ability to attract migrants from else-
where. People move from one state
(or country) to another for a variety
of reasons. But most long distance
moves can be sustained only if there
exists an adequate economic oppor-
tunity at the destination to sustain a
similar or superior standard of
living to that of the persons place
of origin. Certainly there are non-
economic amenities that attract people to Colorado, but most migrants
have relocated and remained because of the relative health of the states
economy. It also should be acknowledged that non-economic amenities
can have indirect economic and demographic impacts. For example, when
decision makers seek a location for a new plant or office, non-economic
Net Migration/Employment
Annual Change Thousands of Persons
70-75 75-80 80-85 85-88 88-93 93-00 00-10
1 Hudson Institute, Inc., Workforce 2000, p. 29.
2 Authors note: Climate, life style and recreational opportunities are
examples of "non-economic amenities" which affect investment
decisions. Of course, "non-economic" is a relative term, but it is
used here to distinguish these factors from more obvious economic
elements such as wage levels, tax policy, and transportation costs.

amenities may result in Colorado being chosen. The decision to locate the
new facility will, in turn, create jobs which are likely to attract new
The medium-growth forecast calls for a return to job growth, al-
though at levels below those of the recent past. As the supply of jobs ac-
celerates, net migration to Colorado
is predicted to rise as well. For ex-
ample, from 1988 to 1993, the state
anticipates an additional 98,000
nonagricultural wage and salary
jobs, but only 22,000 net migrants.
The discrepancy between job
growth and net migration can be at-
tributed to the relatively rapid
growth of Colorados labor force, which, in turn, is a by-product of both
the changing age composition of Colorados population (an increasing
proportion of the population in the prime working ages), and the increased
female labor force participation rates.
Indeed, the aging labor force that will encompass the nation will
impact Colorado as well. From 1988 to 2010, Colorados labor force will
expand at an average annual rate of 1.8%, a marked slowdown from the
1970 to 1988 period when its yearly growth was 3.5%. Thus, the con-
tinued slowdown in labor force growth, coupled with steady employment
gains should push the Colorado unemployment rate downward. By 2010,
Population Comparison
Medium Growth Scenario
Annual Percent Change
I-iI ii i:
70-76 75-60 8065 856868-83 93-00 00-10

the state unemployment rate is predicted to be 5.7%, a level on par with
the 1970s, and well below the 1988 rate of 7.0%.
As mentioned earlier, Colorados relatively prosperous economy is
assumed to affect directly the states ability to attract business and new
workers to the state. Migration to and from a state is a function of family
ties, business relationships and perceptions of job opportunities. With lit-
tle real and perceived job growth occurring until the late 1990s, net migra-
tion is predicted to stay below 10,000 per year until 1998. Specifically,
the forecast calls for net migration to average 4,000 per year from 1988 to
1993,14,000 from 1993 to 2000, and 52,000 from 2000 to 2010. On the
surface, the accelerating migration from 2000 to 2010 might seem very
large. By historical standards the sharp rise in migration is not large. For
example, from 1970 to 1980, annual net migration totaled 1.6% of total
employment growth, while the forecast calls for a 1.3 % average from 2000
to 2010.
Mostly, then, over the next quarter-century, the slowing labor force
growth is a consequence of demographic changes already sweeping the
country and the state In Colorado, population is forecast to increase at
an annual rate of 1.4%, a marked decline from the 2.3% growth rate
recorded from 1970 to 1988.
Colorados 1985-86 recession struck most visibly in the retail sales
market. From 1985 to 1988, Colorado nominal retail sales expanded just
0.7% per year in sharp contrast to the double digit rates experienced
during the 1970s. Over the forecast horizon a resumption of consumer and
business spending fueled by a recovery in the state economy, by faster

disposable income growth, and by rising inflation is expected to generate
increases in retail sales of between 5% and 9% per year.
Finally, the future housing market also will closely mirror the
demographic trends in the state. While mortgage rates and housing affor-
dability surely will play important roles in housing activity, the expansion
of the new home buying group (ages 25 to 34) is the central determinant
of long-run demand. Thus, the slower growth of the 25-34 age group will
restrain housing demand. From 1988 through 2010, housing starts in
Colorado will average 20,100 per year, a marked decline from the 24,800
average attained during the 1970 to 1988 period.

Medium Growth Scenario
1970 1975 1980 1985 1988 1993 2000 2010
Incomes ($ Billions)
Personal Income 8.6 15.5 29.5 47.5 53.8 73.2 138.5 356.3
Annual Growth Rate (%) 1Z0 0.3 13.7 10.0 4.2 6.4 9.5 10.0
Wages and Salaries 5.7 10.0 19.1 29.5 32.7 44.6 84.7 215.8
Annual Growth Rate (%) 11.0 0.8 13.9 9.1 3.5 6.4 9.6 9.8
Personal Income ($1982) 20.2 26.2 34.0 42.5 43.5 46.7 60.6 92.93
Annual Growth Rate (%) 5.0 0.4 5.4 4.6 0.8 1.4 3.8 4.5
Employment, Labor Force, Earnings, and Prices
Labor Force (Thou) 922.6 1,162.1 1,502.1 1,719.0 1,722.8 1,890.0 2,087.0 2,497.0
Annual Growth Rats (%) 4.7 5.2 2.7 0.1 1.9 1.4 1.8
Unemployment Rate (%) 3.5 5.0 5.9 5.9 7.0 7.1 5.9 5.7
Hourly Earnings Mfg. ($/Hr) 3.50 5.03 7.63 9.51 10.29 13.06 21.04 40.42
Annual Growth Rate (%) 7.5 8.7 4.5 2.6 4.9 7.0 6.7
Denver CPI-U (1982-84=1.00) 0.345 0.484 0.784 1.149 1.236 1.521 2.186 3.801
Annual Growth Rate (%) 7.0 10.1 7.9 2.5 4.3 5.3 5.7
Non-Ag. Wage and Salary Employment
Mining 14.1 19.8 36.2 32.9 21.2 25.4 30.0 37.6
Annual Growth Rate (%) 7.0 12.8 -1.9 -13.7 3.7 2.4 1.5
Construction 42.4 53.3 77.1 86.4 64.6 73.5 80.5 88.3
Annual Growth Rate (%) 4.7 7.7 2.3 -9.2 2.6 1.3 0.6
Durable Manu. 72.7 81.2 116.6 123.7 119.0 129.9 140.6 161.1
Annual Growth Rate (%) 2.2 7.5 1.2 -1.3 1.8 1.1 1.2
Non-Durable Manu. 48.1 56.0 63.8 68.5 66.5 66.9 68.5 72.5
Annual Growth Rate (%) 3.1 2.7 1.4 -1.0 0.1 0.3 1.0
Trans. & Pub. Util. 49.6 59.4 79.3 88.5 86.4 91.9 98.4 110.3
Annual Growth Rate (%) 3.6 5.9 2.2 -0.8 1.2 1.0 1.1
Wholesale Trade 47.1 55.8 75.2 80.9 77.2 85.9 103.0 139.5
Annual Growth Rate (%) 3.4 6.1 1.5 -1.6 2.1 2.6 3.1
Retail Trade 127.2 177.6 229.6 271.1 275.7 289.8 337.0 427.1
Annual Growth Rate (%) 6.9 5.3 3.4 0.6 1.0 2.2 2.3
Finance, Insur., Real Estate 38.7 55.2 76.4 95.8 98.3 97.6 103.6 121.2
Annual Growth Rate (%) 7.4 6.7 4.6 0.8 -0.1 0.9 2.5
Services 132.9 188.6 253.5 321.8 339.6 376.6 463.2 623.6
Annual Growth Rate (%) 7.3 6.1 4.9 1.8 2.1 3.0 2. .9
Government 177.2 216.6 243.5 248.8 264.6 273.5 300.8 335.4
Annual Growth Rate (%) 4.1 2.4 0.4 2.1 0.7 1.4 1.0
Total-Colorado 750.2 963.5 1,251.2 1,418.6 1,413.1 1,510.9 1,725.5 2,116.8
Annual Growth Rate (%) 5.1 5.4 2.5 -0.1 1.3 1.9 2.0
Total-U.S. (Millions) 70.9 76.9 90.4 97.5 104.2 112.2 120.2 131.6
Annual Growth Rate (%) 1.7 3.3 1.5 2.2 1.5 1.0 0.8

TABLE 2 (continued)
1970 1975 19B0 1985 1988 1993 2000 2010
Population Variables
(Millions of Persons)
Colorado Population 2.21 2.59 2.89 3.23 3.33 3.49 3.74 4.48
Annual Growth Rate (%) 3.2 2.2 2.2 1.0 0.9 1.0 1.8
Births 0.042 0.04 0.05 0.055 0.053 0.049 0.047 0.061
Deaths 0.017 0.017 0.019 0.020 0.021 0.022 0.025 0.033
Net Migration (Annual Average) 0.045 0.046 0.031 -0.003 0.004 0.014 0.052
Sales and its Determinants
Disposable Income ($Mil) NA 13,404 24,870 40,553 45,838 61,633 115,800 296,540
Annual Growth Rate (%) 13.2 10.3 4.2 6.1 9.4 9.9
Retail Sales ($Mii) NA 8,474 16,585 24,410 24,930 32,536 59,338 141,923
Annual Growth Rate (%) 14.4 8.0 0.7 5.5 8.9 9.1
Retail Sales ($ Mil-1967) NA 5,335 6,905 8,164 8,012 8,219 10,149 14,703
Annual Growth Rate (%) 5.3 3.4 -0.6 0.5 3.1 3.9
Sales/Disposable Income 0.632 0.667 0.602 0.543 0.528 0.512 0.479
Construction Variables
Households (Thou) 685 863 1,071 1,222 1,284 1,401 1,557 1,827
Annual Growth Rate (%) 4.7 4.4 2.7 1.7 1.B 1.5 1.6
Housing Stock (Thou) 747 991 1,172 1,386 1,477 1,582 1,746 1,997
Annual Growth Rate (%) 5.8 3.4 3.4 2.1 1.4 1.4 1.3
S.F. Housing Permits (Thou) 16.0 14.2 19.4 20.2 13.9 15.9 16.3 18.1
Total Housing Permits (Thou) 31.1 16.7 30.0 33.6 15.0 19.6 22.6 23.2
Mining and Agriculture
Crop Sales ($Mil) 268 601 1,023 1,145 938 1,565 2,353 3,897
Livestock Sales ($Mil) 941 1,318 2,211 2,019 2,664 2,823 4,224 6,819
Rotary Rig Count-Rocky Mtns 153 253 449 265 179 268 330 413
Note: Annual growth rates calculated on a point-to-point basis. For example, the annual growth rate for Colorado
population from 1970 to 1975 reflects the annual compounded growth rate over that five year time frame. This
growth rate is displayed under the 1975 column heading. This method of calculating and displaying annual
growth rates is used throughout this study.
Data source: Center for Business and Economic Forecasting
Colorado Department of Local Affairs. State Demographer's Office

Growth Forecast The medium-growth forecast calls for slower
population and employment increases, which is consistent with the
forecast of slower labor force growth. From 1970 to 1988, Colorados
labor force grew at an annual rate of 3.5 percent; from 1988 to 2010 the
annual growth rate is expected to drop to 1.7 percent. The substantial drop
in workforce growth will be accompanied by important changes in the age,
gender and ethnic composition of Colorado s workforce. Before discuss-
ing changes in composition, a brief review of workforce growth is ap-
Until recently, Colorados workforce grew rapidly. Strong demand
for labor from a growing economy was met by an influx of migrants, by
the graduation of Colorados baby boomers into working ages and by the
increasing labor force participation rates for Colorado women. Since
1985, however, the slowdown in Colorados economy has served to
decrease the state workforce by 22,000 workers, resulting mostly from a
drop in participation rates and from a net outmigration of the working age
The medium-growth economic forecast calls for a return to
workforce growth in the 1988-93 period as the states economy recovers
from the recent recession, thereby increasing die demand for labor.

Average annual growth for 1988-93 is forecast to be 33,400 workers, or a
1.9 percent annual growth rate. Relatively little net migration is required
in this scenario as the movement of more people into the working ages and
a return to earlier labor force participation rates will yield sufficient
workers to accomodate the forecasted demand for labor at an unemploy-
ment rate of 7.1 percent. In the
1993-2000 period, workforce
growth will drop to 28,100 per
year (1.4 percent), and by the
twenty-first century, growth
should accelerate to 41,000 per
year (1.8 percent). The aging of
the population and slower in-
creases in labor force participa-
tion rates at the turn of the century will require increases in net migration
to meet the forecasted demand for labor.
An Aging Workforce The Colorado workforce will continue to
age. This year, 1988, the median workforce age is 34.9 years. In other
words, half of the workforce is younger, and half is older than 34.9 years.
By 2000, the median age will increase to 38.6. 2010 will see a slight
decline to 37.6, as large numbers of new workers move into the state to
meet the demand for labor consistent with the medium-growth forecast
scenario. The aging phenomenon reflects a new trend in Colorado. From
1970 to 1980, the median age of Colorados workforce decreased from
37.2 to 33.1 years. However, much of the fall in the median age during
Colorado Labor Force
Participation Rates
1970 1980 1988 1993 2000 2010

the 1970s can be attributable to the large flow of young workers to the
While the median age is a useful summary statistic describing the
aging of the workforce, for many purposes, it is more appropriate to
analyze the relative and absolute changes in the sizes of different age
groups. Between 1988 and 2000, the under-35 age group will present the
most dramatic change in the age composition of the workforce, falling from
50.3 percent in 1988, to 40.6 percent in 2000. In contrast, the 35-54
category will grow from 39.7 percent in 1988 to 48.6 percent by 2000. The
turn of the century should witness an increase in the percentage of workers
in the 55-64 age group from 8.1 percent in 1988 to 12.1 percent in 2000,
and, a return to growth in the under-35 group, though it will still represent
5.7 percent less of the workforce than it did in 1988.
Expressed simply as changing shares of the workforce, these trends
do not appear to represent dramatic shifts. From 1988 to 1993, the num-
ber of workers under 35 in Colorado is predicted to decline by 5,400 per
year while workers in the 35-54 category are forecast to grow by 40,700
per year. In the 1993-2000 period, growth is projected to return to the
youngest age group with the annual addition of5,300 workers aged 16-24.
The 25-44 age group will lose 2,600 workers annually and the 45-64 age
group will gain 25,700 workers per year during the same time period. And
at the beginning of the the twenty-first century, workers in the 55-64 cat-
geory are predicted to grow by 11,200, second only to the 25-34 category
which will grow by 15,200.

The Feminization Of Colorados Workforce While women have
always worked, until relatively recently, a substantial majority did not seek
or obtain work for pay outside the home. (Only those working for pay or
seeking such work are officially counted as part of Colorados workforce).
By 1988, however, women made up 43.5 percent of the Colorado
workforce, representing a substantial increase over their 36.5 percent share
in 1970. A steadily increasing rate of labor force particpation accounts for
womens greater share of the Colorado workforce. In 1970,42.6 percent
of the Colorado women aged 16 and over were either working or seeking
work (compared to 78.5 percent for men). By 1988, the rate is estimated
to have increased to 61.1 percent. According to projections of the U.S.
Bureau of Labor Statistics, the rate is expected to reach 66.1 by 2000 (still
less than the 80.3 rate for men). The continued increase actually will rep-
resent a deceleration in the rate of growth of female labor force participa-
tion from what has been observed during the past 20 years, as more women
move into the older age groups where it is expected that they will follow
the declining participation rates of men that have been recorded in recent
decades. While women stand to remain a minority in Colorados
workforce, they nevertheless will account for 54.0 percent of workforce
growth between 1988 and 2000.
Increasing Minority Representation Colorados workforce has
been, and will continue to be, predominantly Anglo. However, the
minority proportion will increase from 15.3 percent in 1988, to 19.0 per-
cent in 2010, due to significantly higher growth rates. From 1988 to 1993,
the minority workforce is expected to increase by 18.1 percent, while the

Anglo workforce will increase by only 9.7 percent. Similar differentials
should persist throughout the forecast period.
The forecasts are based on projections of the total population of
four minority groups in Colorado Hispanic, Black, Asian and American
Indian and on relatively conservative assumptions about changes in fer-
tility, mortality and migration. Obviously, a larger-than-projected influx
of Hispanic or Asian immigrants will affect their share of the workforce
proportionately. The forecast also assumes that the lower labor force par-
ticipation rates observed for Colorado s Hispanic, Black and American In-
dian residents to date will continue during the forecast period.
Although Anglos will continue to statistically dominate Colorados
workforce throughout the forecast period, Anglo males will account for
only one-third of workforce growth between 1988 and 2000; Anglo
females will account for 41 percent, and minority males and females will
each account for 13 percent.
Implications For the past two decades, Colorado has had a rapid-
ly growing, young and predominantly Anglo workforce. The winds of
change have begun to blow across the Colorado frontier, however. The
most dramatic change has been the slowdown in workforce growth. A
prolonged state recession has prevented Colorados economy from adding
a significant number of new jobs since 1984. The aging of Colorados
workforce, underway for some time, has accelerated with the dropoff in
net migration (slowing the influx of young workers) and the aging of the
baby boom generation. Only after the turn of the century, when large num-
bers of retirees are expected to be replaced by a new influx of young

workers, will the median age of the workforce begin to decline. Closing
out the twentieth century, Anglo females as well as minority males and
females, will account for a growing share of Colorados workforce.
Slower workforce growth is largely a function of the expected slow-
down in the demand for labor. Should a repeat of the economic boom of
the 1960s and 1970s occur, it is most likely that workers would be recruited
from elsewhere to meet the demand. Even under a rapid-growth scenario,
however, the workforce will age, and a smaller portion of jobs will be held
by Anglo males. In other words, the forecast changes in the demographic
composition of Colorados workforce will occur. The speed with which
they will take place will depend on the pace of economic growth.
Thus, Colorado employers seeking to plan for tomorrows
workforce needs should reexamine their employment policies for women,
minorities and older workers in light of the declining pool of Anglo male
From a workers point of view, changing workforce demographics
will affect bargaining power in the market place. Relatively young
workers, both Anglo and minority, and both male and female, will have
fewer younger workers with whom to compete for entry level jobs. Older
workers, however, may face barriers to advancement as their relative share
of the workforce increases dramatically.

Age 1980
16-19 63.4
20-24 80.3
25-34 82.9
35-44 82.3
45-54 76.8
5534 57.6
65+ 13.8
TOTAL 69.6
16-19 67.1
20-24 88.4
25-34 96.2
35-44 96.8
45-54 93.1
55-64 74.2
65+ 20.6
TOTAL 82.6
16-19 59.6
20-24 72.1
25-34 69.2
35-44 67.7
45-54 60.9
55-64 42.4
65+ 9.1
TOTAL 57.0
1987 1993
Total Workforce
62.2 64.2
82.0 83.7
85.8 87.8
86.0 88.4
79.9 82.4
55.8 55.6
11.9 10.6
71.3 72.5
All Males
64.4 65.5
88.1 89.2
95.7 95.3
96.3 95.9
92.9 92.7
69.5 67.3
17.0 14.6
81.7 81.2
All Females
60.0 62.8
75.5 78.0
75.5 80.0
75.5 80.7
67.2 72.1
43.4 44.7
8.3 7.7
61.1 64.2
2000 2010
66.1 66.1
85.4 85.3
89.8 89.8
90.5 90.6
84.2 83.7
56.3 55.9
9.0 9.3
73.1 70.7
67.0 67.0
89.9 89.8
95.1 95.0
95.7 95.7
92.2 92.0
66.2 65.6
12.0 12.5
80.3 78.0
65.2 65.2
80.6 80.6
84.2 84.2
85.2 85.2
76.1 75.4
46.8 46.4
6.8 6.9
66.1 63.5
Data Source: Colorado Department of Local Affairs, State Demographer's Office

Age 1980 1987 Anglo 1993 2000 2010
16-19 64.0 62.6 64.7 66.9 67.1
20-24 80.9 82.8 84.8 86.7 86.7
25-34 83.2 86.4 88.6 90.7 90.7
3544 82.5 86.2 88.7 90.9 90.9
45-54 77.1 80.3 82.7 84.6 84.2
55-64 57.8 56.0 55.8 56.7 56.4
65+ 15.1 13.1 11.9 10.2 10.8
TOTAL 69.4 71.1 72.3 72.9 70.1
16-19 56.6 52.8 54.5 56.2 56.5
20-24 74.8 75.4 77.1 79.2 79.1
25-34 74.7 77.1 79.6 81.7 81.9
3544 76.5 78.9 81.8 83.8 83.9
45-54 69.2 69.4 72.0 74.1 73.9
55-64 45.9 44.2 43.7 44.5 44.7
65+ 10.1 6.6 4.7 3.3 3.3
TOTAL 64.8 65.8 67.5 68.4 66.5
16-19 56.4 58.6 60.1 61.4 60.9
20-24 81.1 82.0 83.5 84.2 83.4
25-34 85.7 86.3 87.2 87.9 87.9
3544 83.2 86.8 88.4 89.6 89.5
45-54 77.9 79.9 81.7 83.5 82.9
55-64 57.9 56.5 54.7 52.5 53.2
65+ 15.2 12.7 9.8 8.4 8.3
TOTAL 73.0 74.8 75.3 74.8 72.5
16-19 56.3 52.7 53.1 52.8 52.7
20-24 62.7 64.9 65.3 66.4 66.0
25-34 72.9 74.0 75.4 78.1 78.1
3544 73.6 75.6 77.8 80.9 81.0
45-54 73.5 74.9 78.5 80.6 80.9
55-64 56.4 57.0 58.0 59.0 59.1
65+ 20.1 17.1 17.5 14.5 14.5
TOTAL 65.4 65.9 66.6 67.2 65.0
American Indian
16-19 58.0 57.6 57.1 57.1 57.3
20-24 70.1 69.9 69.5 69.3 69.5
25-34 78.6 78.6 78.6 78.3 78.2
3544 76.8 76.9 76.9 76.8 76.6
45-54 72.2 71.5 71.3 71.3 71.2
55-64 49.5 49.7 50.1 49.8 49.5
65+ 16.4 16.8 17.1 17.1 17.3
TOTAL 68.4 68.8 68.2 66.9 64.9
Data Source: Colorado Department of Local Affairs, State Demographers Office

Introduction Every industry in Colorado requires a distinctly dif-
ferent mix of occupations to create its goods and services. The finance in-
dustry, for example, needs more accountants, analysts, and computer
operators than the transportation industry; the transportation industry
needs more engineers. In manufacturing, high tech firms require more
scientists and engineers than do their lower tech counterparts in the tradi-
tional industries such as a concrete or steel.
While each industry contains a different occupational mix, they are
bound together by a common thread their occupational mix gradually
changes over time.
"The health-care industry, for example, is increasingly dominated by
large national firms that manage a wide variety of hospitals, clinics,
and other facilities. For this reason, the industry is employing in-
creasing numbers of managers and administrators whose jobs did not
exist 15 years ago."1
By projecting the changes in occupational mix according to in-
dustry, and by applying the resultant matrix to employment levels in the
1 Hudson Institute, Workforce 2000, p. 96.

respective industries, it becomes possible to forecast coming changes in
occupational structure for any future year. This study chose to forecast the
industrial and occupational structures in 1993,2000, and 2010, and arrived
at over 900 projected occupations for those years. In order to simplify
the analysis of the occupational data, detailed occupational forecasts have
been aggregated into 37 major occupational groups (see Table 5). The
Chapter II outlines the specific detail on the methods used to produce the
occupational projections.
Occupational growth pat-
terns were analyzed according to
the following yardsticks: 1) ab-
solute growth from the base year
(1988), and 2) on a percent
change basis. These analytical
rulers produce two snapshots of
the future. The absolute growth
analysis identifies those areas in
which the largest volume of resources should be dedicated to provide for
future workforce needs. Complimenting the absolute growth examination,
the percent change analysis gives clues to the growth professions those
occupations that may not generate large volumes of jobs over the next 2
Colorado Employ.Distrib.
Medium Growth Scenario
Percent of Total
ii7Q E53 ibm B3m Mata
2 See Appendix I of Colorado Workforce 2000 for detailed
occupational forecasts.

twelve years, yet whose high growth would seem to ensure long-term sig-
nificance, In the future, they are the high growth professions that could
prove to be the backbone of the economy.
Analysis Forecasting employment by sector demonstrates that the
service sector will be the largest provider of new jobs in 1993, 2000 and
2010 (see the graph on the previous page). The wholesale trade and
finance, insurance, and real estate sectors will also prove to be major sour-
ces of service-oriented job growth. It should come as no surprise then that
the largest absolute number of jobs to be created from 1988 to 2000 will
fall in "administrative support" categories (38,882), "other service occupa-
tions" (36,875), "officials and administrators" (28,160), and "other sales
occupations" (24,597). In fact, of the 37 occupational groups shown in
Table 4, the four just listed will account for 36.6% of all new jobs
Other occupations not attaining large absolute growth, but ex-
periencing quick expansions, are worth noting. Of the six fastest growth
occupational groups, four are in professions requiring high degrees of
education. Natural/computer/mathematical scientists, lawyers and judges,
accountants and auditors, and engineers, architects, and surveyors are
projected to expand between 25% and 28% from 1988 to 2000. Due to
the small size of the current workforce in these occupations, however, none
of them are projected to make the top ten in terms of absolute growth.
Employment related to goods-producing industries will experience
medium to low job growth. For example, construction trade occupations
will add an estimated 10,991 jobs from 1988 to 2000, an increase of less

than 1,000 a year. Stepping down on the growth scale, precision produc-
tion occupations, machine setters and operators, and plant and system oc-
cupations each will generate less than 5,000 new jobs during the next
twelve years. The slow growth outlook for these manufacturing-related
jobs is consistent with the flat growth profile for the manufacturing in-
dustry as a whole. As mentioned earlier, productivity increases in this in-
dustry will limit job opportunities for the rest of the century.
At the low end of the job creation scale are communication equip-
ment operators, extractive workers, short order cooks, and bank tellers. It
is interesting to note that these occupations cross-cut many different in-
dustry groupings. And some of them short order cooks and bank tellers
- are classified in the fastest growth sector services. Although the
economy will become more dependant on the service economy in general,
some service jobs will become less in demand as a result of changing tech-
nology (bank tellers, for example).
Close inspection of the extractive workers group (mining, oil, and
gas) proves to be enlightening; namely,the extractive industries will
generate the fourth-lowest number of jobs while, at the same time, record-
ing the fastest growth rate of the 37 occupational groups. This
phenomenon can be attributed to the relatively low volume of workers clas-
sified in these occupations. In fact, in 1988, only an estimated 681 in-
dividuals could be classified in this occupational grouping. Hence, small
absolute changes create large percentage changes.

Ranked By Occuaptional Skill Score
Medium Growth Low High
Skil Growth Growth
Occuoationai Group Score 1988 1993 2000 2010 2000 2000
Helpere/laboreis 2.3 48246 51983 59489 73275 55308 70518
Janitors & Cleaners 2.4 37146 40863 48937 65118 44772 57828
A Hand working occupations 2.9 26457 29230 31980 37539 28070 37605
Total 111849 122078 140406 175932 128150 165949
Percentage of total employed 7.1% 7.3% 7.3% 7.6% 7.2% 7.8%
Machine setters and operators 3.1 33830 34505 37498 42491 34254 42367
Extractive workers 3.6 5214 5895 6979 8890 6602 7934
Transportation & material moving 3.8 58437 62339 71875 88029 66572 83041
Waiters and waitress 3.9 30355 32308 37869 48581 36099 42473
Agricdture/forestry/fishing 4.1 28287 31162 34024 38334 32957 37590
Precision production occupations 4.5 22452 24224 26874 31616 24551 30963
Cashiers 4.6 30338 31892 37183 47325 35471 41675
B Cooks, except short order 4.7 19400 20585 23887 29916 22866 26385
Short order cooks 4.7 1727 1815 2117 2708 2124 2370
Other service occupations 4.8 154706 164768 191581 239930 180827 214269
Total 384746 409493 469687 577820 442323 529067
Percentage of total employed 24.5% 24.3% 24.5% 24.8% 24.7% 24.2%
i Meehan jcs/instaJtors/re parrots 5.1 64087 68548 77629 95034 71897 92311
Other administrative support occs 5.3 177199 189845 216081 265079 199236 248748
Communication equipment operators 5.3 13260 14002 15503 18455 14311 17975
Plant and system occupations 5.4 1210 1254 1415 1595 1360 1462
Secretaries/stenographers/typists 5.6 59210 63750 73135 89054 67304 82951
Construction trades 5.7 48501 53775 59492 66300 57748 71459
Financial record processing 5.7 25966 26975 29723 36100 27043 34470
Banktallers 6.0 5464 5424 5761 6873 5167 6548
c Blue collar worker supervisors 6.1 26420 28419 31827 36985 29748 37281
Other sales occupations 6.6 108024 115419 132621 164258 123793 153273
Writers/artists/en terrain ers/athletee 6.9 19002 20431 23317 27902 21388 26935
Salespersons 6.9 48726 51426 59670 75265 57036 66805
Total 597069 639068 726174 882900 676029 840218
Percentage of total employed 38.0% 38.0% 37.8% 37.9% 37.7% 38.5%
Other ptofessionaVparaptoTlechnicians 7.2 16594 17619 20751 23108 18282 22141
Teachers/! ibrariana/counselors 7.8 91594 95959 108943 125105 105107 112072
Other management occupations 7.9 35556 37835 42292 50407 38948 48454
Technician occupations 7.9 41182 44961 51139 61926 46778 58405
n Sodal/recreational/reiigious workers 8.0 10645 11418 13322 15810 12471 14228
Officials & administrators 8.6 131213 140481 159373 191441 148725 182323
Total 326784 348273 395820 467797 370309 437623
Percentage of total employed 20.8% 20.7% 20.6% 20.1% 20.7% 20.1%
| Health diagnosis snd treating 9.5 55187 59119 66087 79803 65184 72229
Social scientists 9.6 13157 14092 16381 19343 15369 17521
Lawyers and judges 9.7 10228 11096 13093 15401 12015 14225
c Accountants A auditors 10.0 14909 16178 18778 22831 17180 21221
c Engineers, architects, & surveyors 10.1 33753 37532 42416 50043 37526 49028
Total 127242 136015 156733 187421 147274 174224
* Percentage of total employed 8.1% 8.2% 8.2% 8.0% 8.2% 8.0%
Natural/computer/mathematical scientists 11.3 23317 25740 30035 38160 27265 35143
F Total 23317 25740 30035 38160 27265 35143
Percentage of total employed 1.5% 1.5% 1.6% 1.6% 1.5% 1.6%
TOTAL 1571007 1682665 1918855 2330030 1791350 2182224
Data Source: Colorado Workforce 2000

Ranked By Occupational Skill Score
Medium Growth Low
SMI Growth
Occupational Group Score 1993 2000 2010 2000
Hetpen/teborers 2.3 3737 11243 25029 7062
Janitors & Cleaner* 2.4 3717 11791 27972 7626
Hand world ng occupations 2.9 2773 5523 11082 1613
Total 10227 28557 64083 16301
Percentage of total employed 9.2% 8.2% 8.4% 7.4%
~~ Machine setter* and operatore 3.1 675 3668 8661 424
Extractive workers 3.6 681 1765 3676 1388
Transportation & material moving 3.8 3902 13236 29592 8135
Waiters and waitress 3.9 1953 7514 18226 5744
AgricUturaforestry/fiahing 4.1 2875 5737 10047 4670
Precision production occupations 4.5 1772 4422 9164 2099
Cashiers 4.6 1554 684S 16987 5133
Cooks, except short order 4.7 1185 4487 10516 3466
Short order cooks 4.7 88 390 981 397
Other service occupations 4.8 10062 36875 85224 26121
Total 24747 84941 193074 57577
Percentage of total employed 22£% 24.4% 25.4% 26.1%
Mschanics/instaUets/repairera 5.1 4461 13542 30947 7810
Other administrative support occs 5.3 12446 38882 87880 22037
Communication equipment operators 5.3 742 2243 5195 1051
Plant and systsm occupations 5.4 44 205 385 150
Secratariesfeenographers/typists 5.8 4540 13925 29844 8094
Construction trades 5.7 5274 10991 17799 9247
Financial record processing 5.7 1009 3757 10134 1077
Bank tellers 6.0 -40 297 1409 -297
Blue cotar worker supervisors 6.1 1999 5407 10566 3326
Other sales occupations 6.6 7395 24597 56234 15769
Writem/artiBtB/entertainera/aihletse 6.9 1429 4315 8900 2386
Salespersons 6.9 2700 10944 26539 8310
Total 41999 129105 285831 78960
_ Percentage of total employed 37.6% 37.1% 37.7% 35.B%
Other professjonal/paraprovtechnicians 7.2 1025 4157 6514 1688
Teachers/librarians/counselora 7.8 4365 17349 33511 13513
Other management occupations 7.9 2279 6736 14851 3392
Technician occupations 7.9 3779 9957 20744 5594
Sodalfrecrsaiional/tefigious workers 8.0 773 2677 5165 1826
Officials & administrators 8.6 9268 28160 60228 17512
Total 21489 69036 141013 43525
_ Percentage of total employed 19.2% 19.8% 18.6% 19.8%
Health diagnosis and treating 9.5 3922 10890 24606 9987
Social scientists 9.6 935 3204 6186 2212
Lawyers and judges 9.7 870 2867 5175 1789
Accountants a auditors 10.0 1267 3867 7922 2271
Engineers, architects, & surveyors 10.1 3779 8663 16290 3773
Total 10773 29491 60179 20032
_ Percentage of total employed 9.6% 8.5% 7.9% 9.1%
[ Naturai/computer/tnathemabcai scientists 11.3 2423 6718 14843 3946
Total 2423 6718 14843 3948
L Percentage of total employed 2.2% 1.9% 2.0% 1.8%
TOTAL 111658 347848 759023 220343
Data Source: Colorado Workforce 2000

Introduction In order to assess the impact of labor force growth
trends in the state of Colorado with respect to the educational policies of
government and private industry, first one must seek to analyze current
and future skill levels of the labor force. The following analysis divides
the 37 occupational groupings presented in Tables 5 and 6 into six skill at-
tainment categories. Each occupational group is assigned a score of one
to six for mathematics and language skills. Hence, a maximum of twelve
points could be earned for those occupational groups demanding the
highest level of skill in both categories. The lowest skill level group will
be denoted as "Group A" and the highest skill level group will be desig-
nated as "Group F".
Findings Although the overall skill level of the population is not
expected to change materially in the future, the mix of skills demanded
will shift. Group A, the lowest skill level group, will rise as a percent of
the new jobs in the future. In 1988, Group A accounted for 7.1% of the
jobs in the state. From 1988 to 1993, it will capture 9.2% of the hew jobs;
from 1993 to 2000, 8.2%; and from 2000 to 2010,8.4%. Because of this
rising share, by 2010 group A will account for 7.6% of the total jobs in the
state, a 0.5% rise from 1988. Most of the Group A increases can be tied
to the accelerating demand for low level, service-oriented jobs.

Percent of Total
Group B, the second-lowest skill category, nets a relatively unstable
share of the new jobs compared to its 1988 proportion of 24.5%. From
1988 to 1993, this group attracts only 22.2% of the jobs created, but
rebounds in 2000 to 24.4% and, by 2010, jumps above the 1988 average
to 25.4%. As with Group A, Group B will add most of its jobs in the ser-
vices industries.
The moderate skill level group, Group C, consistently attracts a
smaller share of new workers in the future. Yet, the shift in the skill mix
is small. In 1988, this category
u uioa riri. ri a Skill Score Distribution
held jo % Ol all JODS in Golorado. Medium Growth Scenario
From 1993 to 2010, this group is
predicted to capture between
37.1% and 37.7% of all new
The next skill level
group, Group D, is predicted to
experience the same type of
decline as Group C. Currently holding an estimated 20.8% of all jobs in
the state, Group D will secure between 19.2% and 18.6% of all the new
jobs from 1993 to 2010. Hence, the middle proficiency categories, Groups
C and D, which held 58.8% of all jobs in 1988 are forecast to attract only
56.6% of the jobs created from 1988 to 2010.
One of the most important trends identified in this analysis is that
two skill groups (with the highest skill scores in Groups E and F) are
predicted to capture markedly higher percentages of the jobs in the future.

Occupational Group
I 1988 E23 2000

These groups, which include doctors, lawyers, accountants, engineers, ar-
chitects, and scientists accounted for 9.6% of all Colorado employment in
1988. In 1993, Groups E and F will account for 9.7% of the total job market
in Colorado, moving up one-tenth of a percentage point seven years later.
In terms of new jobs, these two groups are predicted to account for 11.8%
of the new jobs through 1993, and 10.4% through 2000. This section
returns to its 1988 proportion in 2010 Overall as 9.9% of the new jobs
created over the next 22 years will be classified in these groupings.
The skill score data indicate that the total skill demands of the state
will not change materially over the forecast horizon. Specifically, in 1988,
the average skill score in Colorado was 6.091. This figure is anticipated
to move to 6.094 in 1993, to 6.092 in 2000 and to 6.071 in 2010. Even the
largest movement, 0.021, from 2000 to 2010 is extremely small compared
to skill level differences between occupations. Hence, it cannot be con-
cluded with any degree of confidence that the states labor force skill needs
will materially decline from 2000 to 20101.
In summary, tomorrows jobs generally will require the same
amount of skill proficiency as todays jobs. However, the distribution of
skills will not be the same. Tomorrows workforce must fill a greater
demand both at the lower end and the higher end of the skill levels. More
laborers will have to perform simple labor tasks for more professionals.
1 See Chapter IX for more explanation.

At the same time, the middle educational groupings will need fewer resour-
ces dedicated for the future as the demand for their skills are expected to
decline moderately.
Comparison to National Trends Over time, as the Colorado
economy matures, the structure of the state economy will change such that
its economy looks more and more like that of the national economy.
Analysis of employment by sector sheds light on this trend, hi 1970,
Colorado held 23.6% of all jobs in goods-producing industries (mining,
manufacturing, and construction), and 76.4% in services, while the U.S.
had 33.3% of jobs in goods industries and 66.7% in service-oriented jobs.
By 1988, the disparity between
the U.S. and Colorado figures
narrowed, as goods jobs ac-
counted for 24.0% of all jobs in
the nation, and for 19.2% in
Colorado. Looking ahead to the
year 2000, goods-oriented jobs
will comprise 21.4% of the jobs
in the U.S., and 18.5% of the
jobs in Colorado. The same trend can be found in the services-producing
industries. In 1988, Colorado holds 80.8% of jobs in this category, and
the U.S. cited 76.0%. Projecting forward to 2000, services jobs should ac-
count for 81.5% of all jobs in Colorado and for 78.6% in the nation.
Clearly, the occupational structure and skill demands both for
Colorado and for the U.S. will become more similar. And as the twenty-
U.S. & Colorado
Services Employment
^ Percent of Total Non-Ag. Employment

first century approaches, one would expect Colorado s predicted skill level
more and more to approximate that of the U.S. Based on data from the
Hudson Institute, and on forecasts of Colorado skill requirements, the skill
scores of the entire country will in-
crease from 5.65 in 1984, to 5.85
in 2000, while those of Colorado
advance minimally from 6.091 in
1988, and to 6.092 in 2000.
The ramifications are evi-
dent. For the nation, the shift to a
more service-oriented economy
will require a higher skill level in
the workforce. On the other hand, Colorado merely needs to maintain the
skill level of its current workforce until 2000. On the surface, Colorados
task would seem to be an easy one simply maintain the status quo to
provide for its future needs. But there is a catch. Colorados highly-edu-
cated workforce is not entirely home-grown. A significant portion of its
labor force has been imported from out of state. The massive migration
of workers to Colorado in the 1970s and early 1980s helped to bring not
only over 500,000 new workers, but also half a million of highly educated
new workers. Unless a similar migration of workers occurs in the future,
Colorado will be faced with a new challenge that of inheriting a growing
workforce not meeting the skill level demands of employers. And unless
Colorado takes the necessary steps to prevent it, the states overall skill
level will decline.
Skill Score Comparison
Colorado vs U.S.
Average Skill Score_____
1988 2000
I Colorado £523 u.8.

Introduction In addition to the medium-growth forecast, two
other economic and occupational forecasts were prepared. The alternative
economic forecasts both were based on the same national forecast, yet dif-
fered with respect to their assumed Colorado growth rates. Alternative na-
tional forecasts were not used to drive alternative state forecasts because
the demographic factors critical to Colorado are not shifted enough in the
alternative national scenarios to account for a realistic future variation in
the state economy. For example, the alternative national forecasts do not
shift net migration between states at a level consistent with historical varia-
tions recorded from 1970 to 1985. Thus, the need arose to produce alter-
native Colorado forecasts in which net migration shifts significantly.
Since the primary determinant of net migration to the state is relative
employment growth, the alternative forecasts were developed with large
variances in employment growth rates.
Low Growth Forecast The low-growth economic forecasts were
developed by changing growth assumptions for the following key employ-
ment sectors: 1) durable manufacturing, 2) telecommunications, 3) busi-
ness services, 4) other services, and 5) federal government.
Generally, the low-growth forecast assumes these industries ex-
pand at the national average over the forecast period. Further, since the

growth in these industries affects many other growth rates, the forecast for
all major variables increases at the national rate as well.
Although this forecast reflects low-growth by historical standards,
it does not reflect a worst case scenario. A more pessimistic forecast would
feature a state economy that consistently underperformed the national
average. However, I determined that a scenario in which the state economy
underperformed the national average to be an extremely unlikely low-
growth alternative.
The low-growth occupational forecasts are directly related to the
employment forecast. The occupational projections were prepared using
exactly the same methodology as for the medium-growth scenario.
High Growth Forecast The high-growth economic forecast was
prepared by modifying the same employment sectors as did the low growth
forecast. However, the growth in the modified sectors was assumed to be
higher than the national average. In particular, the high-growth forecast
assumes that the Colorado economy outstrips national employment growth
rates by an amount slightly less than that achieved during the 1972 to 1984
period. Furthermore, an assumed construction boom in conjunction with
the construction of the new Denver airport represents another key dif-
ference in this scenario. As a result, construction employment growth is
forecast to accelerate at an annual rate of 7.5%, from 1988 to 1993.
As with the low-growth forecast, the high-growth scenario does not
represent the highest conceivable relative growth rate measured against
the nation. Yet, this high-growth forecast should be considered as the most

likely high-growth path realizable for Colorado in the long-run. Shown
below is a graphical analysis of the alternative forecasts.
Population Growth
Alternative Scenarios
Non-Ag. Employment
Alternative Scenarios
Skill Score Comparison
Alternative Scenarios
Skill Score in 2000
Growth Scenario

Low Growth Scenario
1970 1975 1980 Incomes ($ Billions) 1985 1988 1993 2000 2010
Personal Income 8.6 15.5 29.5 47.5 53.8 72.0 128.3 290.7
Annual Growth Rate (%) 12.0 0.3 13.7 10.0 4.2 6.0 8.6 8.4
Wages and Salaries 5.7 10.0 19.1 29.5 32.7 43.7 78.1 175.2
Annual Growth Rate (%) 11.0 0.8 13.9 9.1 3.5 6.0 8.7 8.3
Personal Income ($1982) 20.2 26.2 34.0 42.5 43.5 46.0 56.2 75.8
Annual Growth Rate (%) 5.0 0.4 5.4 4.6 0.8 1.1 2.9 3.0
Employment, Labor Force, Earnings, and Prices
Labor Force (Thou) 922.6 1,162.1 1,502.1 1,719.0 1,722.8 1,878.5 2,040.3 2,116.3
Annual Growth Rate (%) 4.7 5.2 2.7 0.1 1.7 1.2 0.4
Unemployment Rate (%) 3.5 5.0 5.9 5.9 7.0 7.2 6.0 6.1
Hourly Earnings Mfg. ($/Hr) 3.50 5.03 7.63 9.51 10.29 13.07 21.05 40.35
Annual Growth Rate (%) 7.5 8.7 4.5 2.6 4.9 7.1 6.7
Denver CPI-U (1982-84=1.00) 0.345 0.484 0.784 1.149 1.236 1.521 2.186 3.801
Annual Growth Rate (%) 7.0 10.1 7.9 2.5 4.2 5.3 5.7
Non-Ag. Wage and Salary Employment
Mining 14.1 19.8 36.2 32.9 21.2 25.4 30.0 37.6
Annual Growth Rate (%) 7.0 12.8 -1.9 -13.7 3.7 2.4 1.5
Construction 42.4 53.3 77.1 86.4 64.6 73.1 79.0 85.1
Annual Growth Rate (%) 4.7 7.7 2.3 -9.2 2.5 1.1 Q.5
Durable Manu. 72.7 81.2 116.6 123.7 119.0 124.3 120.2 119.5
Annual Growth Rate (%) 2.2 7.5 1.2 -1.3 0.9 . -0.5 -0.2
Non-Durable Manu. 48.1 56.0 63.8 68.5 66.5 66.6 65.1 58.8
Annual Growth Rate (%) 3.1 2.7 1.4 -1.0 0.0 -0.3 -0.9
Trans. & Pub. Util. 49.6 59.4 79.3 88.5 86.4 88.8 84.6 72.9
Annual Growth Rate (%) 3.6 5.9 2.2 -0.8 0.6 -0.7 -1.9
Wholesale Trade 47.1 55.8 75.2 80.9 77.2 85.1 97.3 118.2
Annual Growth Rate (%) 3.4 6.1 1.5 -1.6 2.0 1.9 1.8
Retail Trade 127.2 177.6 229.6 271.1 275.7 287.0 322.0 373.0
Annual Growth Rate (%) 6.9 5.3 3.4 0.6 0.8 1.7 1.2
Finance, Insur., Real Estate 38.7 55.2 76.4 95.8 98.3 95.9 93.0 79.8
Annual Growth Rate (%) 7.4 6.7 4.6 0.8 -0.5 -0.4 -1.2
Services 132.9 188.6 253.5 321.8 339.6 367.7 419.1 488.4
Annual Growth Rate (%) 7.3 6.1 4.9 1.8 1.6 1.9 1.2
Government 177.2 216.6 243.5 248.8 264.6 272.3 291.9 302.6
Annual Growth Rate (%) 4.1 2.4 0.4 2.1 0.6 1.0 0.1
Total-Colorado 750.2 963.5 1,251.2 1,418.6 1,413.1 1,486.0 1,602.1 1,735.7
Annual Growth Rate (%) 5.1 5.4 2.5 -0.1 1.0 1.1 0.6
TotaMJ.S. (Millions) 70.9 76.9 90.4 97.5 104.2 112.2 120.2 131.6
Annual Growth Rate (%) 1.7 3.3 1.5 2.2 1.5 1.0 0.8

TABLE 7 (continued)
1970 1975 1980 1985 1988 1993 2000 2010
Population Variables
(Millions of Persons)
Colorado Population 2.21 2.59 2.89 3.23 3.33 3.48 3.66 3.87
Annual Growth Rate (%) 3.2 2.2 2.2 1.0 0.9 0.7 0.6
Births 0.042 0.04 0.05 0.055 0.053 0.049 0.046 0.047
Deaths . 0.017 0.017 0.019 0.020 0.021 0.022 0.025 0.031
Net Migration (Annual Average) 0.045 0.046 0.031 -0.003 0.003 0.003 0.004
Sales and Its Determinants
Disposable Income ($Mil) NA 13,404 24,870 40,553 45,838 60,634 107,318 241,931
Annual Growth Rate (%) 13.2 10.3 4.2 5.7 8.5 8.4
Retail Sales ($Mil) NA 8,474 16,585 24,410 24,930 31,889 54,936 116,298
Annual Growth Rate (%) 14.4 8.0 0.7 5.1 8.1 7.7
Retail Sales ($Mil-1967) . NA 5,335 6,905 8,164 8,012 8,063 9,406 12,062
Annual Growth Rate (%) 5.3 3.4 -0.6 0.1 2.2 2.6
Sales/Disposable Income NA 0.632 0.667 0.602 0.543 0.526 0.512 0.481
Construction Variables
Households (Thou) 685 863 1,071 1,222 1,284 1399 1530 1712
Annual Growth Rate {%) 4.7 4.4 2.7 1.7 1.7 1.3 1.1
Housing Stock (Thou) 747 991 1,172 1,386 1,477 1581 1732 1933
Annual Growth Rate (%) 5.8 3.4 3.4 2.1 1.4 1.3 1.0
S.F. Housing Permits (Thou) 16.0 14.2 19.4 20.2 13.9 15.2 13.5 11.9
Total Housing Permits (Thou) 31.1 16.7 30.0 33.6 15.0 18.8 19.8 17.1
Mining and Agriculture
Crop Sales ($Mil) 268 601 1,023 1,145 938 1,565 2,353 3,897
Livestock Sales ($Mil) 941 1,318 2,211 2,019 2,664 2,823 4,244 6,819
Rotary Rig Count-Rocky Mtns 153 253 449 265 179 268 330 413
Data Source: Center for Business and Economic Forecasting
Colorado Department of Local Affairs, State Demographer's Office

High Growth Scenario
1970 1975 1980 1985 1988 1993 2000 2010
Incomes ($ Billions)
Personal Income 8.6 15.5 29.5 47.5 53.8 86.8 169.0 471.0
Annual Growth Rate (%) 12.0 . 0.3 13.7 10.0 4.2 10.0 10.0 11.1
Wages and Salaries 5.7 10.0 19.1 29.5 32.7 52.8 102.8 282.5
Annual Growth Rate (%) 11.0 0.8 13.9 9.1 3.5 9.9 10.0 10.9
Personal Income ($1982) 20.2 26.2 34.0 42.5 43.5 55.4 74.0 122.8
Annual Growth Rate (%) 5.0 0.4 5.4 4.6 0.8 4.9 4.2 5.5
Employment, Labor Force, Earnings, and Prices
Labor Force (Thou) 922.6 1,162.1 1,502.1 1,719.0 1,722.8 1983.0 2,306.0 3,018.0
Annual Growth Rate (%) 4.7 5.2 2.7 0.1 2.9 2.2 2.7
Unemployment Rate (%) 3.5 5.0 5.9 5.9 7.0 6.4 5.5 5.0
Hourly Earnings Mfg. ($/Hr) 3.50 5.03 7.63 9.51 10.29 13.50 21.49 40.98
Annual Growth Rate (%) 7.5 8.7 4.5 2.6 5.5 6.9 6.6
Denver CPI-U (1982-84=1.00) 0.345 0.484 0.784 1.149 1.236 1.521 2.186 3.801
Annual Growth Rate (%) 7.0 10.1 7.9 2.5 4.3 5.3 5.7
Non-Ag. Wage and Salary Employment
Mining 14.1 19.8 36.2 32.9 21.2 25.8 30.4 38.0
Annual Growth Rate (%) 7.0 12.8 -1.9 -13.7 4.0 2.4 1.5
Construction 42.4 53.3 77.1 86.4 64.6 92.8 103.8 113.9
Annual Growth Rate (%) 4.7 7.7 2.3 -9.2 7.5 1.6 0.7
Durable Manu. 72.7 81.2 116.6 123.7 119.0 138.7 167.7 214.3
Annual Growth Rate (%) 2.2 7.5 1.2 -1.3 3.0 2.8 2.3
Non-Durable Manu. 48.1 56.0 63.8 68.5 66.5 69.2 71.1 81.3
Annual Growth Rate (%) 3.1 2.7 1.4 -1.0 0.8 0.4 2.3
Trans. & Pub. Util. 49.6 59.4 79.3 88.5 86.4 105.8 126.9 170.5
Annual Growth Rate (%) 3.6 5.9 2.2 -0.8 4.2 2.6 3.1
Wholesale Trade 47.1 55.8 75.2 80.9 77.2 103.6 123.2 176.1
Annual Growth Rate (%) 3.4 6.1 1.5 -1.6 6.0 2.5 4.0
Retail Trade 127.2 177.6 229.6 271.1 275.7 328.7 376.5 505.1
Annual Growth Rate (%) 6.9 5.3 3.4 0.6 3.5 2.0 3.2
Finance, Insur., Real Estate 38.7 55.2 76.4 95.8 98.3 115.4 117.8 153.0
Annual Growth Rate (%) 7.4 6.7 4.6 0.8 3.2 0.3 4.2
Services 132.9 188.6 253.5 321.8 339.6 26.0 540.3 789.0
Annual Growth Rate (%) 7.3 6.1 4.9 1.8 4.6 3.5 4.0
Government 177.2 216.6 243.5 248.8 264.6 283.3 312.2 361.6
Annual Growth Rate (%) 4.1 2.4 0.4 2.1 1.4 1.4 1.6
Total-Colorado 750.2 963.5 1,251.2 1,418.6 1,413.1 1,689.2 1,969.1 2,602.9
Annual Growth Rate (%) 5.1 5.4 . 2.5 -0.1 3.6 2.2 3.0
Total-U.S. (Millions) 70.9 76.9 90.4 97.5 104.2 112.2 120.2 131.6
Annual Growth Rate (%) 1.7 3.3 1.5 2.2 1.5 1.0 0.8

TABLE 8 (continued)
1970 1975 1980 1985 1988 1993 2000 2010
Population Variables
(Millions of Persons)
Colorado Population 2.21 2.59 2.89 3.23 3.33 3.56 4.11 5.40
Annual Growth Rate (%) 3.2 2.2 2.2 1.0 1.5 1.8 3.7
Births 0.042 0.04 0.05 0.055 0.053 0.051 0.055 0.077
Deaths 0.017 0.017 0.019 0.020 0.021 0.022 0.025 0.034
Net Migration (5 Yr Avg) 0.045 0.046 0.031 -0.003 0.033 0.039 0.092
Sales and its Determinants
Disposable Income ($Mil) NA 13,404 24,870 40,553 45,838 73,065 141,300 391,955
Annual Growth Rate (%) 13.2 10.3 4.2 9.7 9.9 11.0
Retail Sales ($Milj NA 8,474 16,585 24,410 24,930 40,419 74,270 193,388
Annual Growth Rate (%) 14.4 8.0 0.7 10.0 0.1 10.3
Retail Sales ($Mil-1967) NA 5,335 6,905 8,164 8,012 10,029 12,555 19,845
Annual Growth Rate (%) 5.3 3.4 -0.6 4.4 3.3 5.1
Sales/Disposable Income NA 0.632 0.667 0.602 0.543 0.553 0.526 0.493
Construction Variables
Households (Thou) 685 863 1.071 1,222 1,284 1,460 1,702 2,150
Annual Growth Rate (%) 4.7 4.4. 2.7 1.7 2.6 2.2 2.4
Housing Stock (Thou) 747 991 1,172 1,386 1,477 1,613 1,854 2,237
Annual Growth Rate (%) 5.8 3.4 3.4 2.1 1.8 2.0 1.9
S.F. Housing Permits (Thou) 16.0 14.2 19.4 20.2 13.9 26.4 23.9 30.2
Total Housing Permits (Thou) 31.1 16.7 30.0 33.6 15.0 31.9 32.0 37.1
Mining and Agriculture
Crop Sales ($Mil) 268 601 1,023 1,145 938 1565 2353 3897
Livestock Sales ($Mil) 941 1,318 2,211 2,019 2,664 2832 4224 6819
Rotary Rig Count-Rocky Mtns 153 253 449 265 179 268 330 413
Data Source: Center for Business and Economic Forecasting
Colorado Department of Local Affairs, State Demographer's Office

Introduction Lyle Kyle, former director of the Colorado Legisla-
tive Council Staff, used to say that the only thing certain about forecasts
was that they were bound to be wrong. Rather than an indictment of the
forecasting process, his remarks sought only to remind his audience of the
inherent risks in attempting to predict the future. To the extent that present
forecasts do not accurately reflect future realities, prediction will suffer.
The risk of error should neither discourage nor deter the forecasting effort,
however, as approaching the future blindly presents far greater risks.
Because this document focuses on forecasts which could have
policy implications for the State of Colorado, an assessment of the mag-
nitude of forecasting error would be useful. Ideally, the best way to get a
sense of potential prediction errors would be to examine previous forecasts
of similar studies prepared by the state. It is not an option in this instance,
however, as this is the first study undertaken by the state using the exact
methodologies outlined in this report. Yet, many other states have
prepared employment and occupational forecasts (some using the
methodology outlined in this report) which allow for the identification of
the magnitude of errors made by the other states in forecasting employ-
ment and occupational growth. The U.S. Bureau of Labor Statistics recent-
ly completed such an analysis.

In October 1987, the BLS released an article entitled, "An evalua-
tion of State projections of industry, occupational employment". The
paper evaluated the accuracy of projections for twenty states, including
Colorado. Survey data was compiled from forecasts generated from 1976
to 1982.
Industry Forecasts Starting first by examining the errors involved
in predicting industry employment, the results indicate that the more
detailed the industry category, the larger the error. Not surprisingly, strikes
and plant closings will have larger impacts on employment at more disag-
gregated levels (two-digit or three-digit SIC code level) because of the
smaller number of establishments. As shown in Table 9, the mean absolute
percent error averaged 16.7% and 22.6%, respectively, for the two-digit
and three-digit levels, compared to 11.8% for one-digit level forecasts.
Projection errors also varied significantly among industry
divisions. The most volatile industries in the economy, mining and durable
goods manufacturing, both recorded mean absolute percent errors of over
30%. Wholesale trade, retail trade, and services had the smallest errors.
The paper also revealed that differences in employment growth
rates explained the largest portion of the variation on the projection error.
Industries which lost 15% or more of their total employment during the
projection period contained the largest prediction mistakes (45.7%). In-
dustries losing between 0% and 15 % of their employment were misf orecast
by an average of 20.2%. However, industries growing between 0% and
15%, or expanding by more than 15%, recorded errors of only 11.2% and

19.3%, respectively, indicating that most forecasts were prepared with an
optimistic bias.
Generally, there is an unwillingness among analysts to forecast
employment declines.
"There may indeed be a sincere fear of creating a self fulfilling
prophecy, because economic growth is less where overall local
economic activity is seen to be stagnant or declining."1
While the results of the BLS article did not explicitly correlate the
level of economic activity with forecasts, they did show that most states
were inclined to forecast increases in employment. In the article, employ-
ment had been predicted to grow in 90.9% of the cases, but actually
recorded increases in only 62.1%.
"Put another way, if employment in an industry sector actually
declined, the chances that the decline had been predicted were less
than one in six."2
Occupational Projections The occupational forecasts prepared
by the states surveyed in the BLS study were developed by multiplying
projections of industry employment by staffing pattern estimates entered
into an industry-occupation matrix. This method gives rise to two types
of errors: 1) errors resulting from industry forecasting errors, and, 2) er-
1 Harcye A. Goldstein and Alvin M. Craze, "An Evaluation of State
projections of industry, occupational employment", Monthly Labor
Review, October 1987, p.31.
2 Ibid.

rors in projecting the distribution of employment by occupation within an
industry; that is, errors in projecting staffing patterns.
An analysis of the sources of occupational forecast error indicates
that inaccurately projected staffing patterns proved to be the central com-
ponent of occupational forecast error.
"The absolute value of the occupational component was 3.4 times
greater than the absolute value of the industry component, indicating
that changes in staffing patterns over the 6 year period were a greater
source of error in the occupational employment projections than were
errors in projecting industry employment."3
Conclusions The employment and occupational forecasts in-
cluded in this study will vary from actual results in the future. Changes in
the economy, technology, and workforce habits are certain to alter the fu-
ture in ways beyond the scope of educated prediction. The errors in
projecting occupational and employment growth are likely to be large,
rendering some estimates statistically unreliable. Although little hope can
be held in predicting the exact level of employment or occupational
growth, it is likely that the general direction of growth can be forecast with
some assurance.
In order to derive the maximum benefit from the forecasts presented
in this study, emphasis should be placed on the trend rather than on the
3 Harcye A. Goldstein and Alvin M. Craze, "An Evaluation of State
projections of industry, occupational employment", Monthly Labor
Review, October 1987, p.33.

specific numerical predictions. Focus on the forecast values for industry
employment, population, occupations, and any other economic or
demographic variable should be minimized in favor of revealed trends -
high growth, low growth, or negative growth. Furthermore, the un-
reliability of the science of forecasting to predict specific future figures
greatly enhances the value of the alternative forecasts which assign the
most likely upper and lower bounds to future growth patterns. Only in this
way can the economy and population in Colorado responsibly be studied.
Any policy that depends upon a specific forecast value is likely to allocate
resources incorrectly compared to a policy that distributes assets based on
definitive trends.

(Error in Percent)
Characteristic Sample Size Mean Error Absolute Standard Deviation
Industry Level
Total, All Industries 20 6.9 4.3
1-Digit SIC 157 11.8 11.7
2-Digit SIC 1,120 16.7 14.5
3-Digit SIC 3,010 22.6 20.7
Industry Sector
Mining 35 32.0 22.0
Construction 139 23.5 20.3
Durable Goods Manufacturing 611 30.6 23.3
Nondurable Goods Manu. 540 23.4 20.1
Transportation 123 23.3 21.5
Communications & Utilities 100 18.7 21.3
Wholesale Trade 306 16.9 16.6
Retail Trade 532 18.4 17.6
Finance, Insur., & Real Estate 208 20.8 19.2
Services 416 19.5 20.7
Growth Rate
15.0% or Less 550 45.7 24.7
14.9% to-0.1% 591 20.2 12.1
0.0% to 14.9% 641 11.2 9.7
15.0% or more 1,228 19.3 18.5
Data source: Monthly Labor Review, Bureau of Labor Statistics, October 1987.

This study reflects an attempt to investigate certain economic and
demographic issues related to the states labor force. A number of sig-
nificant trends were uncovered in the analysis. First, it is clear that the na-
tional economy will continue to expand in the future, but at rates below
the past twenty five years. The national labor force is forecast to increase
1% annually from 1988 to 2010, more than a full percentage point below
the average recorded from 1962 to 1987.
The long term growth prospects for Colorado are expected to be
above average compared to other states. Yet, Colorado is not predicted to
regain a relatively strong growth rate until the turn of the century. The
recent statewide recession has uncovered some key weaknesses in the state
economy. The forecast indicates that it will take at least five years before
a marked increase in any of the states major economic aggregates is
The face of the states labor force will change markedly over the
next quarter century. The expanding Colorado workforce will be older,
more feminine, and less Anglo. In 1988, the average age of the Colorado
workforce was 34.9 years. It is forecast to rise to 38.6 years by 2000.
Female labor force participation rates are predicted to rise from 61.1% in
1988, to 66.1% by 2000. Colorado males are forecast to experience a

declining participation rate. In addition, the minority portion of the labor
force is projected to rise from 15.3% in 1988 to 19.0% in 2000.
While the states labor demands will grow in the future, there is lit-
tle chance that the state will be able to in-migrate sufficient workers to
meet its labor market needs. The forecast calls for net migration to average
4,000 per year from 1988 to 1993, and to average 14,000 per year from
1993 to 2000, in sharp contrast to the annual net migration of 40,000
recorded during the 1970s. This trend is the result of two factors. First,
the nationwide labor force will be aging in the future. Generally, older
workers are less likely to move than younger workers. Hence, the supply
of potential migrants will be reduced. Second, the remnants of an
economic recession are still hanging over the state. Until economic con-
ditions improve in the future, workers will have little incentive to relocate
to Colorado.
Although the states labor force and occupational structure will
Change dramatically from 1988 to 2010, skill requisites are forecast to shift
only small amounts. The occupational forecasts reveal that demand will
increase for both high skilled and low skilled workers. Jobs requiring mid-
level skills will be in lesser demand. The net result of this changing job
mix on total skill requisites will be negligible. For example, the skill score
of Coloradans is forecast to move from 6.091 in 1988, to 6.094 in 1993,
to 6.092 in 2000, and to 6.071 in 2010. That compares to a marked rise in
national skill requisites from 5.65 in 1984 to 5.85 in 2000.
The relative movement of skill scores reflects some key differen-
ces between Colorados economic structure and that of the nation. First,

the Colorado economy has historically had a larger concentration of ser-
vice oriented jobs than the nation as a whole. The Colorado economy is
forecast to continue to become more service oriented, but at a slower rate
than the nation. These trends are the result of the national economy losing
manufacturing employment at a faster pace than Colorado while record-
ing increases in service jobs at approximately the same rate. Since workers
in the national economy are forecast to require significantly higher level
skills over the next twelve years, one can conclude that new service
oriented jobs will require a greater amount of training than the manufac-
turing jobs they replace. The convergence of Colorados skill requisites
with that of the nation reflects the growing similarity of their respective
economic structures.
The predicted inability of Colorado to migrate large numbers of
workers coupled with the need to maintain the skill level of its workforce
poses major challenges to state policy makers. How does the state main-
tain the skill level of its workforce when historically much of its highly
trained workers were imported from out-of-state? Clearly, the answer is
that either private industry and/or government must take move foreword
to provide the skilled workers needed throughout the state. If the state
maintains the current educational status quo in an effort to provide for fu-
ture needs, it is highly probable that much of the states highly trained labor
force needs will go unmet. Without a definitive effort to upgrade the
quality of the states workforce, Colorados future economic prosperity
will be constantly at risk.

Computer Models
Alexander, Barton, and Christopher K. Blackwood. Skill Level Forecast-
ing Model. May 1988.
Center for Business and Economic Forecasting Incorporated. Com-
puterized Econometric Model of the Colorado Economy. December
Colorado Department of Local Affairs, Cohort Component Population
Projection Model. May 1988.
Utah Department of Employment Security. Industry/Occupational
Microcomputer Matrix System. March 1988.
Center for Business and Economic Forecasting Incorporated. The CBEF
Review of the Colorado Economy. Denver: Center for Business and
Economic Forecasting, Winter 1988.
Data Resources Incorporated. U.S. Long Term Review. New Yoik: Mc-
Graw Hill Publishing, Winter 1987-88.
Diegh, Robb, and Jill Rachlin and Amy Saltman. "Best Jobs for die Fu-
ture." U.S. News and World Report, 25 April 1988, pp. 60-80.
Hudson Institute. WORKFORCE 2000, Work and Workers for the Twen-
ty-first Century. Indianapolis: Hudson Institute, June 1987.

Government Publications
U.S. Department of Commerce, Bureau of the Census. 1970 Census of
Population, Characteristics of Population, Colorado. Washington:
U.S. Department of Commerce, Volume I, Part 7.
Colorado Department of Labor and Employment and the Colorado Depart-
ment of Local Affairs. Colorado Workforce 2000. Denver: Colorado
Department of Labor and Employment, September 1988.
Colorado Department of Labor and Employment. Occupational Projec-
tions 1988 1993. Denver: Colorado Department of Labor and
Employment, April 1988.
Goldstein, Harcye A, and Alvin M. Cruze. "An Evaluation of State projec-
tions of industry, occupational employment." Monthly Labor Review,
Bureau of Labor Statistics, October 1987, pp. 29-38.