A passenger train between Denver and Fort Collins

Material Information

A passenger train between Denver and Fort Collins
Putka, Robert
Publication Date:
Physical Description:
79 leaves : maps ; 28 cm


Subjects / Keywords:
Railroad trains -- Planning -- Colorado -- Denver ( lcsh )
Railroad trains -- Planning -- Colorado -- Fort Collins ( lcsh )
Railroad trains -- Planning ( fast )
Colorado -- Denver ( fast )
Colorado -- Fort Collins ( fast )
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )


Includes bibliographical references (leaves 77-79).
General Note:
Submitted in partial fulfillment of the requirements for the degree, Master of Arts, Department of Economics.
Statement of Responsibility:
by Robert Putka.

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:
19817900 ( OCLC )
LD1190.L53 1988m .P87 ( lcc )

Full Text
Robert Putka
B.B.A., Cleveland State University, 1974
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
Robert Putka
has been approved for the
Department of
7 / 2.?/yg-

Putka, Robert (M.A., Economics)
A Passenger Train Between Denver and Fort Collins Thesis
directed by Professor John Morris
The purpose of this paper is to demonstrate that
a passenger train between Denver and Fort Collins is an
economic viability if it is operated differently than a
conventional train. Three major factors are necessary in
order to provide a successful passenger train. First,
exorbitant labor costs must be avoided. Secondly,
existing railroad tracks must be used if capital costs
are to be minimized. Finally, new technology, in the
form of modern equipment, needs to be purchased by the
carrier of passengers between Denver and Fort Collins so
as to attract riders and minimize operating expenses.
Explaining various ways to keep all of the costs
at a minimum and maximize ridership are the purposes of
this paper. Initially, this.paper identifies the
existing train and highway corridors between Denver and
Fort Collins. The primary purpose of these identifi-
cations is to show that the demand for potential
ridership exists.
After ridership is established, economic
comparisons are made between a formerly operated
passenger train, the Rio Grande Zephyr, and a passenger
train proposed by the Colorado State Rail Plan Advisory
Committee. These comparisons provide sufficient

information to derive a beginning balance sheet for the
Denver to Fort Collins passenger train.
Finally, after the outline of the necessary, but
minimal, expenditures is provided for starting the
operation of the modern passenger train and the demand
for potential ridership in the Denver to Fort Collins
corridor is established, a projected revenue is proposed
to justify the expenditures and attract riders.
The form and content of this abstract are approved.I
recommend its publication.
S igned
Professor John Morris

I. INTRODUCTION...................................... 1
II. IDENTIFICATION OF CORRIDORS....................... 5
Existing Rail Corridors......................... 5
Highway Corridors............................... 9
The Ideal Corridor for the Denver
to Fort Collins Train........................ 9
Notes-Chapter II............................... 14
Rio Grande Zephyr.............................. 16
Projected Costs for a Denver to
Glenwood Springs Train...................... 23
Notes-Chapter III.............................. 38
Summary of Research Models..................... 39
Development of a Demand Model for the
Denver to Fort Collins Train................ 44
Notes Chapter IV............................. 55
TRAIN MODEL.................................... 57
Personnel...................................... 57
Capital Equipment.............................. 61
Operations................................. 6 5
Notes Chapter V

VI. SUMMARY AND CONCLUSIONS...................... 70
BIBLIOGRAPHY........................................ 77

1. Labor Costs for the Rio Grande Zephyr
from 1979 to 1982 ........................ 18
2. Capital Expenditures for the Rio Grande
Zephyr from 1979 to 1982 ................. 20
3. Operating Expenses for the Rio Grande
Zephyr from 1979 to 1982 ................. 22
4. Revenue, Expenses, and Net Earnings for
the Rio Grande Zephyr from 1979 to 1982. . 24
5. Labor Costs for the Denver to Glenwood
Springs Train in 1980 ........................ 28
6. Capital and Depreciation Costs for the
Denver to Glenwood Springs Train in 1980. 30
7. Operations Costs for the Denver to
Glenwood Springs Train in 1980 ........... 32
8. Total Costs for the Denver to Glenwood
Springs Train in 1980 ........................ 34
9. Comparisons for the Rio Grande Zephyr
and the Denver to Glenwood Springs Train
in 1980 ...................................... 35
10. Estimated Total Cost for the Denver to
Fort Collins Train..........................

1. Union Pacific Corridors............................ 7
2. Burlington Northern Corridors...................... 8
3 . Highway Corridors................................. 10

For the most part, transportation economists are
unenthusiastic about the rapid rail systems built in the
last twenty-five years (e.g., Washington, D.C. Metro:
San Francisco BART; and the Atlanta system). Many
economists feel that the systems do not represent the
most efficient use of taxpayers' money. If, however, a
system could be built at a relatively low cost per mile,
the chances are that a rail system would be economically
The purpose of this paper is to demonstrate that
a passenger train between Denver and Fort Collins has a
strong prospect for economic viability if it is operated
differently than a conventional train. A key factor in
providing an efficient passenger train is personnel. It
will be shown that exorbitant labor costs can be avoided.
Another factor in favor of a Denver to Fort Collins train
is the fact that railroad tracks already exist in this
corridor. If these tracks are used, then capital costs
can be minimized.
Chapter II discusses the various transportation
corridors between Denver and Fort Collins. The foremost

purpose of this chapter is to outline the current train
tracks between these two cities and to establish a choice
of one of these train corridors to make a potential
commuter line a profitable venture. Another purpose of
this chapter is to delineate the highway corridors that
parallel these tracks between Denver and Fort Collins in
order to establish a demand factor for potential
ridership in this corridor.
Economic comparisons between a formerly operated
passenger train, the Rio Grande Zephyr, and a passenger
train proposed by the Colorado State Rail Plan Advisory
Committee will be the topic of Chapter III. Both of
these trains provide sufficient economic history to
establish cost figures to operate a passenger train in
the Colorado region. After the balance sheets of the Rio
Grande Zephyr are discussed and the projected costs of
the latter train are outlined, suggestions for a more
efficient passenger train will be delineated in order for
the Denver to Fort Collins train to be successful.
Chapter IV discusses the research about the
development of previous demand models which provide
information about potential ridership. In order for a
Denver to Fort Collins train to be successful, ridership
needs to be maximized. As will be seen, previous models
have related the demand for ridership to factors such as
space consumption, automobile ownership, travel time, and
specific sociological factors. In addition to the demand

cited by certain transportation economists, there are
some major cities in the United States that have
successful transit systems currently in operation. The
new demand model, which will be applicable to the Denver
to Fort Collins corridor, will have a combination of the
factors from previous demand models, implementations from
the most contemporary transit systems and some additional
factors that have not been previously cited.
The proposal for a passenger train from Denver to
Fort Collins is the basis of Chapter V. With all of the
past history of previous passenger trains, some things
need to be changed in order to make the new passenger
train feasible. For example, the operation of and
equipment on this train would necessarily differ from
that of most other passenger trains. More specifically,
the proposed commuter train would have fewer crew members
than are currently used on most trains. In addition, the
proposed Denver to Fort Collins train would need to
provide courteous, on-time service to its passengers.
Furthermore, modern technology has made available a new
rail vehicle, the railbus, which not only carries
passengers but provides its own propulsion. Just as a
self-propelled vehicle minimizes labor costs, similar
reductions can be made concerning other operating costs.
The last chapter summarizes the minimal
expenditures necessary for operating a passenger train in
the Denver to Fort Collins corridor. The success of this

train hinges on several key factors. If new labor
practices can be enforced, while utilizing railroad track
already intact, then a passenger train can become a
viable alternative for a commuter in the Denver to Fort
Collins corridor. The final part of Chapter VI will
address the projected revenue based on a given demand for
the proposed Denver to Fort Collins train.

There are several existing rail corridors that
could be utilized for a Denver to Fort Collins passenger
train. This chapter evaluates existing corridors in
order to determine which has the greatest potential for
profitable operation. Such an evaluation involves
examination of initial capital costs, operating costs,
potential for ridership, and the logistics of using a
specific corridor that are compatible with the current
owner of the tracks.
Existing Rail Corridors
Two major companies, Union Pacific (UP) and
Burlington Northern (BN) have trackage from Union Station
in Denver heading north toward Fort Collins. Union
Pacific's mainline begins at Union Station and travels
north to northeast out of Denver. A train departing from
Union Station on this route passes through Commerce City,
Brighton, Fort Lupton, Greeley, Ault, and Cheyenne,
Wyoming. This line is UP's major freight corridor from
Colorado to Wyoming. In addition to transporting freight,
this line at one time accommodated a passenger train, the

former San Francisco Zephyr on a round trip route. Six
miles to the west of the mainline in Commerce City, UP
has a branchline, which runs north to Milliken and
parallels the mainline for approximately thirty-seven
miles. The final twenty-three miles of this branch head
in a northwest direction toward Fort Collins. The total
mileage from Denver to Fort Collins on Union Pacific's
branchline is sixty-six miles (see Figure 1).
Burlington Northern is the other primary railroad
company that has tracks north of Denver. Trains
travelling on the BN tracks leave Denver heading north to
northwest, a distance of approximately sixteen miles to
Broomfield. Two miles northwest of Broomfield the BN
tracks split. The mainline continues on a northern trek
through Erie and onward to Longmont. In addition, the BN
branchline follows a northwestern direction to Louisville
and Boulder. The branchline from the Broomfield juncture
to Boulder is twelve miles. A northeastern route from
Boulder to Longmont, which runs for an estimated eleven
miles through the Niwot area, returns the branch back to
the mainline. The final stretch from Longmont to Fort
Collins, via Berthoud and Loveland, is an additional
twenty-seven miles. Therefore, the complete trip from
Denver to Fort Collins, using the branchline of the
Burlington Northern, is sixty-six miles (see Figure 2).

Fig. 1 : Union Pacific Corridors
Denver to Fort Collins

Fig. 2
Burlington Northern Corridors
Denver to Fort Collins

Highway Corridors
In addition to the train tracks in the northern
part of the state, numerous state highways, county roads,
and even one federal interstate connect Denver to Fort
Collins. Interstate 25 (1-25) is the major north-south
highway connecting these two cities. The total driving
distance between the Mile High City and Fort Collins is
fifty-nine miles, the last four miles of which consist of
a short western stretch on Colorado State Highway 14
(SH 14). For the purpose of future reference, this
corridor is called Highway Corridor 1 (HWC #1) (see
Figure 3).
The second route connecting Denver and Fort
Collins is referred to as Highway Corridor 2 (HWC #2);
this route runs northwest from Denver on US 36 to US 287,
another name for Denver's Federal Boulevard. Even though
this corridor is approximately the same distance as the
1-25 route, it takes a driver a longer period of time to
make the trip between Denver and Fort Collins because of
the stop lights and signs interspersed throughout the
towns of Lafayette, Longmont, Berthoud, and Loveland.
The Ideal Corridor for the Denver
to Fort Collins Train
The primary reason for utilizing existing
railroad track on a Denver to Fort Collins passenger
train is to reduce initial capital costs in order to make

Fig. 3
Highway Corridors
Denver to Fort Collins

the system cost efficient. In either case, the BN
branchline or the UP branchline provide suitable railroad
tracks for operating a safe passenger train between
Denver and Fort Collins. Just as capital costs may be
minimized for this new train, operating expenses must
also be kept to a minimum.
A major operating cost is the lease arrangement
made between the carrier of the passengers who ride the
Denver to Fort Collins train and the owners of the track.
The proper lease arrangement is one of the significant
factors in making a new passenger train an economic
If operating costs are assumed to be equal,
except for the lease arrangement, on either the BN
branchline or the UP branchline, and cost considerations
are the only factors in making a decision about possible
railroad corridors, then the railroad company offering
the better lease arrangement would be the choice of the
carrier between Denver and Fort Collins. However,
revenue plays a large role in the success of any proposed
passenger train. So the optimal choice for any train
must take both cost and the potential for ridership into
The ideal corridor for a Denver to Fort Collins
passenger train is the branchline of the Burlington
Northern which passes through Boulder and Longmont (see
Figure 2). One demand factor that influences the

decision to use this corridor is population density. The
use of this corridor would permit the commuter population
surrounding the greater Boulder community to utilize the
Denver to Fort Collins train. Another factor which would
influence the decision to choose this corridor is the
large number of automobiles and busses that travel on
SH 36 between Denver and Boulder. A passenger train
travelling from Denver through Boulder and onward to Fort
Collins could make a stop in Boulder to increase the
total number of riders on this passenger train.
A secondary consideration for choosing the BN
branchline is the logistics of train movements. The
mainline between Longmont and Broomfield accommodates
freight trains travelling in this corridor. In other
words, the split in the tracks allows all trains to use
the right-of-way without any costly time delays, regard-
less of how many trains travel on these corridors.
The total freight train movements on the BN
branchline in 1981 were seven per day.-*- The mainline can
handle up to twenty trains per day, so an additional
passenger train travelling three round trips per day
would not overburden the branchline corridor. The only
other aspect that needs to be worked out is the time
logistics of moving both passenger and freight trains on
both lines in the Denver to Fort Collins corridor.

An example of possible, departure times for a
commuter train from Fort Collins could be 7:15 A.M.,
11:15 A.M., and 3:15 P.M. The arrival times of the
proposed commuter train in Denver would be approximately
ninety minutes later. The commuter train's departure
from Denver could be at 9:15 A.M., 1:30 P.M., and
5:15 P.M.
In summary, both the demand factors and the
logistics of operating a Denver to Fort Collins passenger
train will be served well by the Burlington Northern
branchline. The lease arrangement made between the
railroad company and the carrier of the passengers would
have to be significantly different in order to alter the
decision to use the BN branchline. Thus, the ideal route
is based on ridership potential and the convenience of
moving other trains through this frequently travelled
corridor. Based on these criteria the BN branchline is
the logical choice.

^Colorado State Rail Plan 81-82 Update prepared
by the Colorado Department of Highways Division of
Transportation Planning Rail/Transit Section February,
1983, 111 2 0 .

Because the costs of utilizing commuter rail are
soaring and government subsidies for passenger train
transportation are diminishing, there are fewer passenger
trains today than there were ten years ago in the western
region of the United States. However, the decrease in
the number of trains does not imply that a passenger
train cannot be profitable. If all aspects of the
passenger train are scrutinized carefully with the
assistance of past history, a passenger train can be
successful and profitable. The recent history concerning
a private non-subsidized passenger train in the Colorado
region, the Rio Grande Zephyr, provides insight into the
estimated costs of operating a passenger train. In
addition, another passenger train, the one proposed by
the Colorado State Rail Plan Advisory Committee in 1981,
supplies further analysis of cost figures in a shorter
corridor within the state of Colorado. Together, the
information from these trains provides sufficient data to
estimate a beginning balance sheet for the Denver to Fort
Collins train.

Rio Grande Zephyr
The Rio Grande Zephyr, a privately owned
passenger train, which operated in the Rocky Mountain
region until 1983, had similar expenditures to those a
proposed passenger train from Denver to Fort Collins
would have. From 1979 to 1982 the Denver Rio Grande
Western (DRGW) was required to file revenue and expense
account information, known as an R-l report, with the
Public Utilities Commission (PUC). The information in
the R-l report included a revenue and expense record
classified according to freight and passenger traffic.
Even though R-l reports are available to the public, the
generalizations outlined in the accountant's
interpretation are sometimes unclear. Therefore, I
contacted Robert L. Jacobsen, Director of Transportation
Research for the DRGW. He forwarded "Revenue, Expenses
and Net Earnings" reports for the Rio Grande Zephyr for
the years of 1979 through 1982. The following
information concerning revenues and expenses has been
extrapolated from these balance sheets for those years.
Classifications for expenditures can be broken down into
three major categories: labor for both train crew and
non-train personnel, capital outlays, and operating
The work force necessary to run the Rio Grande
was the traditional five man operating crew (i.e., an
engineer, one fireman, one conductor, and two brakemen)

in conjunction with two coach porters. This train crew
travelled the complete trek from Denver to Salt Lake
City, a total of 570 miles. During the course of a
single year, 313 trips were made. One day of each week,
Wednesday, was used for maintenance and repair of capital
equipment. The figures stated in these R-l reports
indicated an average increase of 10% per annum over the
four year period from 1979-1982 for the seven man train
crew (see Table 1). Salaries, payroll taxes, and bene-
fits; which included vacations, health and welfare insur-
ance; comprised the expenditures for the train crews.
These figures also contained additional reimbursement for
lodging or extra time spent by the crew on one trip due
to breakdowns or inconveniences caused by inclement
The non-train labor personnel were classified
according to DRGW as personnel used to complement the
efficient use of the passenger train. Some of these
employees, namely station personnel, yard crews, and
switching operators, had overlapping duties. They served
the railroad company in a position that benefited both
the passenger trains and freight trains. So a pro-ration
was determined by DRGW to measure non-train crew salar-
ies for the passenger lines. Other personnel included
those individuals necessary to maintain the proper flow
of passenger traffic from station to station. Table 1
summarizes the expenditures for labor, both train and

Table 1
Labor Costs for the Rio Grande Zephyr
from 1979 to 1982
Year Train Crew Non-Train Crew Total Labor
1979 $752,870 $140,668 $ 893,538
(4.22) ( .79) (5.01)
1980 $843,620 $152,857 $ 996,477
(4.73) ( .86) (5.59)
1981 $912,506 $173,145 $1,085,651
(5.11) ( .97) (6.08)
1982 $984,555 $196,998 $1,181,553
(5.52) (1.10) (6.62)

non-train crew, per annum for the Rio Grande Zephyr. In
addition, each parenthetical amount below a figure in
Table 1 represents the cost per mile travelled during
that particular year. The total mileage per annum was
178,410 miles. In 1982 the average total labor cost
approached one dollar per crew member per mile.
From 1979 to 1982, labor expenses remained
constant relative to total expenditures. Annual labor
costs to operate the Rio Grande Zephyr involved, on the
average, 22% of the total outlay. In addition, capital
expenditures also remained constant relative to total
expenditures. Approximately 30% of the total expenses
per year was spent on repairs and depreciation of both
locomotives and passenger cars. Thus, the majority of
the capital outlay was funneled into constant renovation
of equipment. Table 2 summarizes the outlays for capital
equipment per annum, in addition to providing the cost
per mile for depreciation and repair. Once again, the
parenthetical amounts indicate the cost per mile based on
a total mileage of 178,410 miles per annum (see Table 2).
As Table 2 indicates, the depreciation of the
vehicles did not change drastically over the four year
time period from 1979 to 1982. However, each time the
Rio Grande Zephyr departed on a trip the cost of repairs
and maintenance on the locomotives and passenger cars
exceeded six dollars per mile. By 1982 the cost for

Table 2
Capital Expenditures for the
Rio Grande Zephyr from
1979 to 1982
1979 1980 1981 1982
Locomotive $ 27,810 $ 24,826 $ 25,850 $ 26,123
Passenger Car $ 77.197 $ 84.014 $ 91.733 $ 91.733
Sub-Total $ 105,007 (.59) $ 108,840 (.61) $ 117,583 (.66) $ 117,856 (.66)
Locomotive $ 502,101 $ 548,358 $ 588,403 $ 609,745
Passenger Car $ 641.060 $ 666.843 $ 786.947 $ 777.357
Sub-Total $1,143,161 (6.41) $1,215,201 (6.81) $1,375,350 (7.71) $1,387,102 (7.77)
Total Capital Expenditure $1,248,168 (7.00) $1,324,041 (7.42) $1,492,933 (8.37) $1,504,958 (8.44)

repairs approached eight dollars per mile. An increase
of this nature should have indicated a warning signal to
DRGW accountants of an exorbitant charge for the
rejuvenation of equipment.
In contrast to depreciation costs, operating
expenses increased from 45% of the total expenses in 1979
to almost 50% of the expenses in 1982. Operating
expenses quite often have been the nemesis of many
passenger trains. For example, the Rio Grande Zephyr
took pride in offering an elaborate dining and buffet car
for the comfort and convenience of its passengers during
its long trip. However, this amenity continually lost
money. Logically, therefore, this particular service
would not be a necessary item on the shorter route
between Denver and Fort Collins. In addition, the cost
of locomotive fuel increased approximately 50% from 1979
to 1980. In 1979 fuel costs were $2.11 per mile and 1980
prices jumped to $3.18 per mile (see Table 3). Train
supplies and expenses also accounted for an average of
10% of total expenditures. The final expenditure
category was a charge by Amtrak for lease arrangements.
This charge over the four year time period averaged 4% of
total expenses per annum. Some of the remaining
incidental expenses, namely insurance, repairs and train
maintenance, advertising, baggage claims, and credit card
service charges also brought an additional 4.5% per annum

Table 3
Operating Expenses for the Rio Grande
Zephyr from 1979 to 1982
1979 1980 1981 1982
Buffet Services $ 770,983 $ 713,522 $ 824,990 $ 982, 674
Locomotive Fuel $ 377,308 $ 567,408 $ 577,075 $ 534, 568
Train Supplies $ 358,473 $ 465,160 $ 537,281 $ 652, 099
AMTRAK Charges $ 128,502 $ 194,942 $ 215,299 $ 225, 505
Miscellaneous $ 160,460 $ 206,758 $ 234,467 $ 265, 725
Total Operating Expenses $1,795,726 $2,147,820 $2,389,112 $2 ,660, 571

to the operating expenditure total. Table 3 summarizes
the operating expenses per annum (see Table 3).
Now that all of the major expenditures have been
delineated for the Rio Grande Zephyr, the purpose of
these costs needs to be specified. Table 4 summarizes
all of the expenditures and revenues from 1979 to 1982
for this passenger train.^ Each parenthetical amount
stated immediately below each figure in Table 4 indicates
the per mile amount for the Zephyr (see Table 4). These
per mile figures along with the cost figures derived by
the Colorado State Rail Plan Advisory Committee will
develop a cost and revenue estimate for the Denver to
Fort Collins train.
Projected Costs for a Denver to
Glenwood Springs Train
The Rail Passenger Service Act of 1970 provided a
program so that Amtrak could contract with certain states
to provide service in corridors other than those on the
national system. In accordance with a mandate from
Congress, Amtrak and the Department of Transportation
(DOT) made a study in 1979 to restructure Amtrak
operations. The Amtrak Improvement Act of 1981 amended
Section 403(b) permitted any state, regional, or local
agency to apply to Amtrak requesting the initiation of
rail passenger service in addition to that service
provided by the basic system.^ During the first year of
service, Amtrak was responsible for 55% of the operating

Table 4
Revenue, Expenses, and Net Earnings
for the Rio Grande Zephyr
from 1979 to 1982
1979 1980 1981 1982
Labor $ 893,538 $ 996,477 $1,085,651 $1,181,553
(5.01) (5.59) (6.08) (6.62)
Operating $1,795,726 $2,147,820 $2,389,112 $2,660,571
(10.06) (12.04) (13.39) (14.91)
Capital $1,248,168 $1 , 324,041 $1,492,933 $1,504,958
(7.00) (7.42) (8.37) (8.44)
Totals $3,937,432 $4 ,468,338 $4,967,696 $5,347,082
(22.07) (25.05) (27.84) (29.97)
Passenger $1,191,654 $1,148,785 $1,506,860 $2,096,384
(6.68) (6.44) (8.45) (11.75)
Dining/ $ 364,224 $ 362,347 $ 402,488 $ 536,548
Buffet (2.04) (2.03) (2.26) (3.01)
Other $ 4,256 $ 4,381 $ 4,925 $ 6,380
Totals $1,560,134 $1 ,515,513 $1,914,273 $2,639,312
(8.74) (8.49) (10.73) (14,79)
Deficit $2,377,298 $2 ,952,825 $3,053,423 $2,707,770
(13.32) (16.55) (17.11) (15.18)

costs, including labor, of any additional passenger ser-
vice put into place and the state was responsible for 45%
of the costs. During the second and ensuing years,
Amtrak was responsible for 35% of the total operating
costs and the state was responsible for 65% of the costs.
Capital costs for providing this additional passenger
service were distributed 50% to Amtrak and 50% to the
Because of the implementation of the national
Amtrak Improvement Act and amended Section 403(b) in
1981, Colorado transportation officials decided to study
the possibility of beginning a passenger train in the
Denver to Glenwood Springs corridor. As a result, the
Colorado State Rail Plan Advisory Committee was
established to oversee this study. City government
officials, businessmen, and other interested citizens
from Aspen, Vail, Glenwood Springs, and Denver were
appointed to the committee which met in cooperation with
the DRGW Railroad and the Public Utilities Commission
(PUC). After the State Rail Plan Advisory Committee
completed its study, the Colorado Department of Highways
Division of Transportation Planning Rail/Transit Section
in cooperation with the Federal Railroad Administration
(FRA) published the committee's findings.
The committee based its cost estimates for the
proposed train between Denver and Glenwood Springs on 313

round trips of 370 miles per trip for a total of 115,810
miles per year. The committee also researched the costs
for the use of train crews, on-board personnel, purchase
of new or used equipment, operating expenses, and
estimated DRGW charges for use of the facilities entailed
in operating 313 round trips per annum.
The projected cost for labor, which was estimated
to be 28% of the total expenditures, was determined by
the Rail Advisory Committee's research of current union
pay standards which stated that an operating crew member
travelling 100 miles per day would earn one crew day.^
Therefore, each member of the operating crew would accrue
four crew days for one round trip between Denver and
Glenwood Springs. Each member of the crew, depending on
job classification, was also to receive varying salary
amounts per diem. For example, the highest paid train
personnel (i.e. engineers and conductors) would receive
$79 per crew day in 1980.^ Taking the $79 cost per crew
day times four crew days per trip times 313 trips per
year would have resulted in an expenditure of $98,908 for
an engineer's annual wages, if he ran all of the trips.
Benefits were not included in this figure; therefore, an
additional 47% of the projected yearly wages would have
to be encumbered for each of the train crew members'
salaries. Because engineers and conductors are on the
same salary scale, the gross expenditure for each of
these crew positions was identical--$ 145,395. A

fireman's salary was slightly less at $134,352 per annum,
while each of the two brakemen earned $132,511 for a
total of $265,023. As a result, all of the operating
crew wages and benefits came to a total of $690,165 per
annum (see Table 5).^
Additional on-board personnel, who assisted in
assuring passenger comfort and safety included two coach
porters and a buffet lounge cook with one attendant.
Service crewmen were paid on an hourly basis without
overtime compensation. For example, the highest paid
serviceman would be the cook. He would receive $9.73 per
hour plus 47% of his wages in benefits according to union
guidelines.^ Therefore, if a cook put in a fifteen hour
day, which was the estimated time it would take for a
round trip between Denver and Glenwood Springs, and was
paid $9.73 per hour for 313 trips per annum plus
benefits, he would earn compensation of $671153. A
buffet lounge attendant would have received $65,635.
Each coach porter's position would have been $62,598. As
a result, total compensation for the four man service
crew could have been as high as $257,'983.
The nine man train crew, for the projected Denver
to Glenwood Springs train, consisting of operating
members and service oriented personnel, would have cost
$948,148 per annum. An additional expenditure for labor
would necessarily include those employees classified as
non-train crew. The Passenger Traffic Department and

Table 5
Labor Costs for the Denver to
Glenwood Springs Train in 1980
Wage Benefits Compensation
Train Operators:
Engineer $ 98,908 $ 46,487 $ 145,395
Conductor $ 98,908 $ 46,487 $ 145,395
Fireman $ 91,396 $ 42,956 $ 134,352
Brakemen (2) $180,288 $ 84,735 $ 265,023
Sub-totals $469,500 $220,665 $ 690,165
Service Crews:
Coach Porters (2) $ 85,167 $ 40,028 $ 125,195
Buffet Lounge Cook $ 45,682 $ 21,471 $ 67,153
Lounge Attendant $ 44,650 $ 20,985 $ 65,635
Sub-totals $175,499 $ 82,484 $ 257,983
Other Non-Train Crew
Passenger Traffic Department $ 134,966
Station Employees $ 46,136
Sub-total $ 181,102
Total Labor Costs $1,129,250

station employees of a projected train from Denver to
Glenwood Springs would have cost the passenger line
$181,102 for their assistance in the proper flow of
traffic. However, some of the station employees would
also assist in the flow of freight trains. Total labor
costs for the projected train amounted to $1,129,250
annually for 313 round trips (see Table 5).
The second set of costs investigated by the
Colorado State Rail Plan Advisory Committee were for the
capital expenditures necessary to operate a passenger
train efficiently and safely in the Denver to Glenwood
Springs corridor. According to the 1980 figures, two
alternatives were investigated regarding this passenger
train. Three head end power (HEP) locomotives would pull
either new cars or rebuilt service cars. In either case
the old locomotives would be rebuilt at a cost of
$640,000 for a total locomotive cost of $1,920,000. The
annual depreciation cost, based on straight line
depreciation for twenty years, would be $96,000 for all
of the locomotives. Also, ten passenger coaches, one
food service car, and a baggage car would be needed to
run the Glenwood Springs train efficiently. If all the
service cars purchased were brand new, the annualized
depreciation cost for these twelve cars would be
$526,250. This resulted in total depreciation costs for
the new option of $622,250 per year (see Table 6). The
second option of rebuilding used cars would cost $220,000

Table 6
Capital and Depreciation Costs for the
Denver to Glenwood Springs Train
in 1980
Type of Car New Annual Depreciation Rebuilt Annual Depreciation
Passenger(10) $8,750,000 $437,500 $3,750,000 $187,500
Food Service(1) $1,025,000 $ 51,250 $ 350,000 $ 17,500
Baggage(1) $ 750,000 $ 37,500 $ 300,000 $ 15,000
Locomotives(3) $1,920,000 $ 96,000 $1,920,000 $ 96,000
Total Depreciation Costs $622,250 $316,000

in annual depreciation. Thus, projected costs for total
depreciation via the refurbishment of all the service
cars would be $316,000 (see Table 6).
In addition to labor and capital expenditures, a
third set of costs were estimated by the committee.
These expenses involved the anticipated operating costs
for a private passenger train. The largest expenditure
within the operations sector was listed for train
supplies and materials at a cost of $714,831. Dining and
buffet services would cost another $495,792, while
locomotive fuel for the round trip of 370 miles from
Denver to Glenwood Springs would be $275,000 annually.
An additional category involved in operations cost
included those miscellaneous items necessary to operate
an efficient train; this amounted to $201,260. Thus, the
general operating costs for the year for the proposed
Denver to Glenwood Springs train would be $1,686,883 (see
Table 7).
The final cost involved in running the Denver to
Glenwood Springs train was $500,000 which the carrier
would have had to pay to the DRGW for leasing its tracks
and facilities. This amount was $4.32 per mile based on
an annual mileage of 115,810 and appears in Table 7 as a
part of operations costs. Although this charge by the
DRGW to the user of its train tracks seems reasonable,
there is no empirical data nor a rationale to justify
this charge. So, when the time comes to begin

Table 7
Operations Costs for the
Denver to Glenwood Springs Train
in 1980
Train Supplies and Materials $ 714,831
Dining and Buffet Service $ 495,792
Locomotive Fuel $ 275,000 (2.37)
Miscellaneous $ 201,260
General Operations Costs $1,686,883
Plus: Lease Arrangement $ 500,000 (4.32)
Actual Operations Cost $2,186,883

negotiating rental costs or lease arrangements for the
use of the Burlington Northern track in the Denver to
Fort Collins corridor, the negotiators should suggest a
reduction in this rate to minimize operating costs.
Table 8 indicates the total costs for the Denver
to Glenwood Springs train. A comparison shows the
difference between total costs is $306,250, which is the
net difference between new coaches and rebuilt cars.
In summary, a comparison between the costs of the
Rio Grande Zephyr and the proposed Denver to Glenwood
Springs train needs to be made at this point for the
benefit of the Denver to Fort Collins train. In 1980,
the average cost for the total labor expenditure on the
Denver to Glenwood Springs train was four dollars higher
per mile than the Rio Grande Zephyr (see Table 9). The
primary reason for this discrepancy was a result of
having two additional crew members on the Denver to
Glenwood Springs train because the State Rail Advisory's
study determined a need for more personnel even though
the train ran two hundred miles less than the Rio Grande
Zephyr. These labor practices seem to be headed in the
wrong direction. If mass transportation is to become a
viable alternative, then labor costs per mile need to be
reduced not increased.
Another comparison should be noted in regard to
capital expenditures. At first sight, according to the
State Advisory Committee study, the usage of rebuilt

Table 8
Total Costs for the
Denver to Glenwood Springs Train
in 1980
New Coaches Rebuilt Cars
Labor $1,129,250 $1,129,250
Capital $ 622,250 $ 316,000
Operations $1,686,883 $1 686,883
Lease Arrangements $ 500,000 $ 500,000
Total Costs $3,938,383 $3,632,133

Table 9
Comparisons for the Rio Grande Zephyr
and the Denver to Glenwood Springs
Train in 1980
Rio Grande Denver to
Zephyr Glenwood Springs
Crew members
Cost per mile
Depreciation per mile
Repairs per- mile
Fuel per mile
7 9
$5.59 $9.75
$ .61 $5.37
$6.81 Not estimated
$3.18 $2.37

passenger and service cars appears to be the most
economical way of minimizing costs. As Table 6 indicates
the annualized depreciation amount for used equipment is
almost half the amount for new equipment. However, the
Rio Grande Zephyr figures have shown that minimum
depreciation costs can be misleading. The exorbitant
repair costs of $6.81 per mile for locomotives and
passenger cars experienced by DRGW for its Denver to Salt
Lake City train was one of the major reasons for the
dissolution of this line (see Table 9). So, the carrier
of the passengers on the Denver to Fort Collins train
should learn from both of these previous situations and
seriously consider purchasing new equipment, but not
necessarily in the same quantity nor the same vehicles as
suggested by the State Advisory Committee.
Operating cost is the final category for
comparison. Dining and buffet service was not a
profitable operation on the Rio Grande Zephyr and an
expensive option on the Denver to Glenwood Springs train.
Therefore, the dining and buffet service should not be
provided on the Denver to Fort Collins train in order to
minimize operating costs. In addition, other operating
costs can be reduced by comparing the fuel costs of the
Rio Grande Zephyr to the Denver to Glenwood Springs
train. In 1980, the fuel cost for the former train was
$3.18 per mile in contrast to $2.37 per mile for the
latter train (see Table 9). Quite possibly, this was

related to the weight and speed of the train. These
factors have a significant impact on the fuel efficiency
of any train. So, the fuel expenditure, which
constitutes a large portion of the cost of operating a
Denver to Fort Collins train needs to be scrutinized
carefully and continuously.
In conclusion, the total labor costs of the Rio
Grande Zephyr were lower than the total labor costs
estimated by the Colorado State Advisory Committee for
the proposed Denver to Glenwood Springs train. Even
though the Zephyr travelled a longer distance than the
latter train, it used fewer employees to operate a train
safely and efficiently. A reduction in the labor force
does not necessarily forgo efficiency. The Denver to
Fort Collins train needs an efficient crew but not
necessarily a seven man crew like the Rio Grande Zephyr.
The inefficient nature of repairing or
maintaining old equipment and the inability to monitor
operating expenses carefully proved to be the downfall of
the Rio Grande Zephyr. Therefore, the operators of the
Denver to Fort Collins train should invest money in new
passenger cars that are compatible with the old track,
negotiate a reasonable lease arrangement with Burlington
Northern, initiate a further study to maximize fuel
efficiency, and eliminate the dining and buffet service
to minimize non-labor expenses.

"Revenues Expenses and Net Earnings for the Rio
Grande Zephyr Passenger Trains Nos. 17 and 18." (The
Denver and Rio Grande Western Railroad Company, Reports
for the years 1979 to 1982).
^United States Code Congressional and
Administrative News. 91st Congress Second Session, V-I,
Section 403(b), 1970, 1555.
^Colorado State Rail Plan 81-82 Update prepared
by the Colorado Department of Highway Division of
Transportation Planning Rail/Transit Section, February,
1983, IV 3.
4Ibid. IV 7.
^Colorado State Rail Plan 81-82 Update, IV-8.
7Ibid. IV- 7.

Summary of Research Models
One priority that should be a concern of
transportation officials interested in a passenger train
from Denver to Fort Collins is to maximize ridership. In
order to maximize ridership, officials need to understand
and apply the research findings by transportation
economists such as John F. Kain, Carol C. McDonough, and
Glenn Westley. These economists have derived demand
models that can be useful in developing a model for the
Denver to Fort Collins corridor. Also, any carrier of
passengers must study those cities (e.g. San Francisco
and Toronto) which have developed recent mass transit
systems with a modicum of success.
In his research, John F. Kain develops a model
which relates rider demand for public transportation to
residential space consumption, automobile ownership,
choice of transportation mode, and the length of journey
to work. Using linear regression techniques, he
estimates this model using cross-sectional data from the
Detroit metropolitan area.

Kain's research indicates that the worker first
chooses a residential location on the basis of space
preference, income, and family size. In general, more
space is used if both income and family size increase.
Also, more spacious residences, which are less costly per
unit, are located further away from work centers.
Furthermore, Kain finds that the worker's decision
regarding residential location and availability of mass
transit service affects his decision concerning whether
or not to purchase an automobile.
The final relationship in Kain's model shows the
mean elapsed time of the journey-to-work being dependent
on the previous decisions about the consumption of
residential space, journey-to-work transportation mode,
and the cost of residential space near the work place.
In addition, Kain's model confirms that transit use is
positively related to transit service. If the service is
great and travel time is minimized, then the demand for
ridership will be maximized.^
McDonough agrees with Kain regarding the
correlation between travel time and transit use; however,
she goes one step further and cites an interesting
relationship between income and relative travel time to
rider demand which Kain does not mention. She suggests
that rider demand is sensitive to changes in time cost,
whether these variations result from changes in travel
time or in the opportunity cost of this time. In

addition, McDonough's study shows that commuters tend to
place greater emphasis on time minimization for peak-hour
trips, which are generally work trips, than on the
occasional, leisure-oriented, off-peak trips. McDonough
also states that as income and opportunity cost of time
increase, standard rail transport, with its high home-
station travel time, becomes a less than optimal choice.
Therefore, the middle income commuter may choose to
travel by automobile, not because he judges rail
transport to be unsatisfactory, but because, given his
valuation of time, auto transport has a comparatively
lower total cost.^ If McDonough's interpretation is
correct, the reduction of home-station travel time
through the provision of efficient bus transportation,
which is correlated to on-time service to and from
suburban rail stations, would be an essential component
in increasing rail demand.
McDonough fails to mention the significant
sociological factors that should be considered in
studying ridership demand. These factors can be found in
the differences between the utility functions of the
riders. Westley's study of the Lindenwald High Speed
Line (HSL) between Philadelphia, Pennsylvania, and
Lindenwald, New Jersey in 1979 contradicts the findings
of McDonough and Kain. Westley states,

Richer people may prefer the HSL more than
poorer people because it may be more important to
the former group to be able to sit quietly and
read a newspaper or to have time to contemplate
and collect thoughts before work, or because in
their circle of friends the HSL is the "educated
thing to do" in view of social problems, such as
pollution and congestion, known to be caused by
the automobile.^
Thus, Kain, McDonough, and Westley cite
convenience and on time service as the primary motivators
in people using mass transportation. However, Westley is
the only one who addresses a sociological factor as a
necessary part in making mass transit successful. Thus,
regional planners must consider the uniqueness of the
area and their target population when planning an
appropriate transit system.
Regional planners who did consider the special
circumstances of their area in creating a successful mass
transit system were the Bay Area Rapid Transit (BART)
officials. The BART system was developed to structure
the future of the San Francisco region. Even though some
economists feel that BART does not represent the most
efficient use of taxpayers' money, citizens still obtain
certain benefits. BART was intended to do far more than
bring commuters into San Francisco. It was conceived
from the start as a regional system that would foster the
growth of the entire Bay area. Civic leaders who
promoted BART chose a rail system over additional highway
improvements because they feared that the prophets
predicting intolerable congestion in the future might be

right. The prospect that more people and more
automobiles would overload the capacities of road systems
seemed plausible enough to command a system that
simultaneously had high capacity yet was conservative in
its space demands. Also, it was intended to generate
development of subcenters throughout the region, raise
land values, and reduce land area devoted to
transportation facilities. Although the intention of
BART officials to create sub-centers throughout the
region was not completely fulfilled, a more optimal
situation than previously existed did occur. Initially,
BART's proposed construction provoked citizens of the
Rockridge neighborhood of Oakland through which the train
would run. The community members protested against
potential high-density or commercial development. As a
result, zoning regulations were implemented which
prohibited apartment houses and shops along this part of
the line, thus securing the established single family
housing pattern. Following the zoning change to single-
family houses, land values rose steadily with the highest
being nearest to the station.^ Because of the increase
in property values, residents along the BART line began
to take even greater pride in maintaining their
properties which further raised real estate values.
Another city that experienced growth along its
public transportation corridors was Toronto. The
building boom at stations along Toronto's Yonge Street

subway was early evidence cited by planners seeking a
positive example of how fixed rail transit stimulated the
economy in a metropolitan area. Toronto's experience was
used by the promoters of BART in the 1960's and similar
arguments were heard prior to the construction of the
Washington D.C. Metrorail system, the Atlanta Subway, and
the Miami-Dade County rail system. There is little
question among planners today that real estate values
escalate rapidly along rail transit lines.
Development of a Demand Model for
The Denver to Fort Collins Train
The key to developing a successful mass transit
is an effective demand model that would determine the
number of riders who would use the Denver to Fort Collins
commuter train. The demand model for this corridor would
have to take into consideration the following factors:
the population density in cities surrounding the
corridor, vehicle traffic counts indicating the actual
movement of people within the corridor, time values,
modal choice which is determined by price and time
values, and finally, those factors directly linked to
improving our society.
Population density, the first factor in
constructing a mass transit system, is similar to Kain's
space preference in the sense that the choice to reside
in a certain neighborhood depends on the convenience of
the area to the place of work. So, a statistical survey

of the Standard Metropolitan Statistical Area (SMSA) of
the Denver region is necessary for the Denver to Fort
Collins demand model. The Denver to Fort Collins region
includes Adams, Boulder, Denver, Jefferson, and Larimer
Counties. The general trend of population growth in this
area would be applicable to the demand model.
One of the governing bodies of the Denver region
is the Denver Regional Council of Governments (DRCOG).
This council constantly refines and evaluates the future
plans and policies for the SMSA, or the Denver region.
According to a 1975 study, over 95% of the 1.5 million
people in the region live in a 360 square mile area which
includes Denver. Population density per square mile is
approximately 4,200 people. Although the core city of
Denver has grown very little since 1960, growth in the
rest of the region has been explosive. From 1960 to
1970, the SMSA grew by 32%, almost 2.5 times the national
average. However, since the mid-seventies, the growth
rate has moderated, and it should be fairly stable for
the remainder of the century, with population increases
at about 1.1% per annum.^ Compounding this growth over
twenty-five years, from 1975 to 2000, sets the population
at 1,970,000 people in the year 2000. This implies an
increase in population density per square mile of the
metropolitan region to 5,500 people.
In 1950, half the metro population resided in the
city and county of Denver. Now, only about one-third of

the metro population lives in Denver. It is quite
possible by the year 2000 that little more than one-
fourth of the population will live in the core city.
This movement to suburban areas will lead to demands for
new housing, transportation, and a whole range of social
The second criterion for determining the proper
demand model for the Denver to Fort Collins commuter
train involves the vehicle traffic counts in the various
highway corridors. The Average Daily Traffic (ADT)
counts per highway are published every two years by the
Colorado State Highway Department. These figures
indicate the vehicle traffic counts at various junctures
along the aforementioned corridors. The most populous
corridor for vehicle traffic is HWC #1 (see Figure 3).
According to a 1984 study in which the Highway Department
divided HWC #1 into eight sections for purposes of
recording daily traffic tallies, the densely populated
area just north of the central Denver business district
had the highest ADT count. As the traffic moved along
HWC #1, a gradual decline in the number of vehicles using
this corridor occurred. Each exit claimed a few more
commuters along the sixty mile stretch between Denver and
Fort Collins.
A logical choice for any commuter travelling
directly between Denver and Fort Collins is 1-25, that is
HWC #1. The ADT count south of SH 14, one of the exits

to Fort Collins, is 19,100 vehicles per day. The count
just north of SH 14 on 1-25 is 8,400 vehicles per day.
Therefore, a difference of 10,700 vehicles represents the
total number of vehicles travelling between Denver and
Fort Collins per day.
Similar statistics exist for HWC #2. The
commuter has two routes from which to choose in
travelling from Denver to Broomfield because the mileage
variation is not significantly different. First, he can
travel on 1-25 to SH 128, which is the exit for
Broomfield. The ADT counts south and north of SH 128 on
1-25 are 52,900 and 36,800 respectively. The net
difference of 16,100 vehicles per day is the traffic
travelling from Denver to Broomfield north on 1-25 and
west on SH 128. A second choice is to travel north from
Denver on 1-25 to US 36 and use the exit on the west side
of Broomfield at SH 128. Likewise, the ADT counts south
and north of the Broomfield exit on US 36 are 41,200 and
38,300 respectively. The net difference on US 36 is
2,900 vehicles per day. Therefore, the total for the two
alternatives from Denver to Broomfield is 19,000 vehicles
per day.
In addition to these statistics, other sub-
divisions of HWC #2 are needed to determine the proper
demand model. First, the traveller from Denver to
Boulder would probably utilize US 36. The lowest ADT
count on US 36 between Denver and Boulder is 38,300

vehicles per day. Secondly, the commuter's logical
choice of highways between Boulder and Longmont is
SH 119. Once again, the lowest ADT count on SH 119
between Denver and Longmont is 11,900 vehicles per day.
Finally, the last highway sub-division is from Longmont
to Fort Collins. The lowest ADT count on US 287 between
Longmont and Fort Collins is 9,500 vehicles per day.^
The Department of Highways continually monitors
the vehicle movements on a variety of different roads and
highways throughout the state of Colorado. One of the
major purposes for keeping an accurate count is to
anticipate possible areas of congestion in the future.
Any corridor that has the potential for reaching its
highway capacity during a particular peak hour would deem
a recommendation by the highway department for expansion.
Although 1-25 has never experienced grid-lock, bumper to
bumper traffic does occur on a regular basis on this
freeway during morning and evening rush hours.
The third set of data to evaluate in order to set
up a realistic demand model for the Denver to Fort
Collins commuter train is time value. According to
Webber, the average commuter using public transit counts
his waiting time for a bus or train three or four times
more heavily than his riding time.^ Webber's inter-
pretation of time values implies that the marginal
utility of a trip is inversely related to the amount of
travelling, time involved. That is, the discomfort of

commuting increases with travelling time and differences
in the elapsed time of different modes serving the same
route are reflected in differences in the utilities these
modes yield.
The association of modal choice with the relative
amount of discomfort involved in travelling by the
various modes provides a useful explanation of travelling
patterns. Unfortunately, this approach is short of
analytical power. It relies on a cardinal measure of
utility, assuming can compare the utilities
derived by individuals with different incomes and that
one can measure the utilities yielded by travelling on
different routes using various transportation modes. If
travelling time affects only tastes, one cannot predict a
change in tastes caused by the introduction of a new and
unknown mode of transportation. As appealing as this
explanation may seem at first, it has to be rejected in
favor of a more rigorous analysis.
Although time, whether it is waiting or travel
time, along with modal choice render different utilities
to each commuter, the primary variable, time, should
command a positive price because it is a scarce resource.
Time must be allocated in income producing work,
transport, and leisure. If work itself provides no
utility, working time is valued at a price equal to the
marginal wage rate. Given perfect substitution between
work and non-work activities, non-working time should be

equal to the marginal wage rate. But institutional
conventions specifying relatively fixed bounds for the
workday and work-week frequently limit this substitution,
thereby decreasing the opportunity cost of non-working
time. The opportunity cost of time should, nevertheless,
vary directly with wages and income for several reasons.
Income may provide some measure of the scarcity of an
individual's time as a higher income indicates a greater
time scarcity and, thus, a higher value to be placed on
non-working hours. Furthermore, in a society where
worth is often measured by monetary success, non-working
time may well be valued in proportion to the market value
of working time, even when they are not freely inter-
changeable .
Also, the opportunity cost of time should vary
significantly between different times of the day. In
general, peak hours for commuting to work are 7 o'clock
to 9 o'clock in the morning and the early evening hours
of 4 o'clock to 6 o'clock. If trip time is categorized
by peak versus off-peak hours, a peak hour should have
the higher opportunity cost. Because the peak trip is
generally a work trip, there is a closer relationship
between peak travel time and working hours. Furthermore,
the peak work trip is taken more often and during a
period when time is relatively scarce, since many weekday
hours are normally allocated to work. Because the value
of time varies directly with its scarcity, discretionary

weekday hours should be more highly valued, with an
opportunity cost more closely related to income,^
Gronau states that the price of a trip, P,
consists of two parts: the monetary costs involved, M;
and the opportunity costs of the elapsed time, KT. That
P = M + KT
where K denotes the price of time and T measures the
elapsed time. The monetary costs, M, include the fare,
travelling expenses incurred on the way to and from the
terminal, and any additional expenses incurred enroute.
The elapsed time consists of time spent on the way to or
from the terminal, waiting time, and time spent
enroute.^ For example, a trip from Denver to Fort
Collins by bus involves a monetary cost of nine dollars
and approximately two hours and fifteen minutes of
travelling time in addition to, on the average, a half-
hour access time to each bus terminal in Denver and Fort
Collins. The price of the trip to a passenger who
assigns to his time a value of five dollars per hour adds
up to twenty-five dollars.
The demand for trips is a function of their price
and the price of other inputs required for the production
of the visit. Also, the derived demand for trips depends
on the demand for visits which may be undertaken for
either personal or business purposes. The demands for
personal and business visits reflect the marginal utility

and the marginal productivity, respectively, of the
visits. Given tastes, the demand for personal visits is
a function of the visit's price, the price of related
activities, and the household's income.
A business trip can be regarded as a short run
migration movement. The incentive to migrate is
inversely related to the cost of migration and directly
related to the immigrant's marginal product differential.
The demand for business visits depends, therefore, on
price factors similar to those affecting the demand for
personal visits and on the difference between the
traveller's marginal products at the point of origin and
at the point of destination. This difference would seem,
as in other cases of migration, to increase with the
traveller's skills and, hence, with his income.
A potential traveller is faced with a decision as
to which mode to use for a trip. The various modes yield
different utilities and involve different time and money
costs. The direct utility derived from a trip is a
function of the convenience, risk, and prestige
associated with travelling by the specific mode. If one
temporarily sets aside these factors, the choice of mode
hinges solely on the cost of travelling by the various
modes. The rational traveller will always choose the
least expensive mode available. That is, mode X is
preferable to mode Y, if and only if Px < Py.

The final category which needs to be considered
if the demand model for the Denver to Fort Collins train
is to be complete involves the effect of mass transit on
our society. First of all, the quality of life would
improve for everyone in terms of better air and less
noise, if people used the train instead of the
automobile. A major contributor to the brown cloud,
composed primarily of hydrocarbons and carbon monoxide,
which frequently looms over the Denver region, is the
automobile. According to measurements by the
Environmental Protection Agency (EPA), the average hourly
emissions from automobiles over a one mile section of
highway are twelve pounds of hydrocarbons and ninety-six
pounds of carbon monoxide,^ Also, the average hourly
automobile traffic is approximately equal to the number
of passengers that can be transported by five locomotives
over a one mile stretch of track. In contrast, the
average emissions for five locomotives over one mile are
approximately four pounds of hydrocarbons and six pounds
of carbon monoxide. The difference between the pollution
levels in these two measurements is significant. Just as
automobile traffic adds to the number of air borne
pollutants, so it adds to environmental noise. Although
the EPA's limit on noise discernible at one hundred feet
for automobiles is 70dB and 90dB for locomotives,^ the
train noise is not as pervasive as automobile noise since
the total exposure seldom lasts more than three or four

minutes. The improvements in our environment should draw
some riders to the Denver to Fort Collins train.
Another attraction to a rider in this corridor is
safety. Although automobile commuters often feel secure
in their ability to move about on highway corridors,
professional people movers have far fewer passenger
accidents on the railways than drivers have on the
highways.-*-^ Overall, rail passenger traffic is safer
than automobile traffic. As a result, demand for
ridership should vary inversely to the number of
passenger rail accidents.
In summary, the research has shown that the
demand for ridership is related to space preference,
automobile ownership, choice of transportation mode,
travel time, which is related to transit service, and the
valuation of time. In addition to the convenience
provided by efficient transit service, some benefits
accrue to society when traffic congestion and pollution
are reduced. Also, the San Francisco and Toronto systems
have shown that property values rise when mass transit is
implemented properly.
In conclusion, an effective demand model for the
Denver to Fort Collins corridor would have to take into
consideration population density, vehicle traffic counts,
time values, and choice of modes. Additional factors
that should be included are property values and the
measurement of particulates in the air.

Ijohn F. Kain. "A Contribution to the Urban
Transportation Debate: An Econometric Model of Urban
Residential and Travel Behavior," Review of Economics and
Statistics. 46 (1964): 55-64.
^Carol C. McDonough, "The Demand for Commuter
Rail Transport," Journal of Transport Economics and
Policy. (May 1973): 134-143.
^Glenn Westley, "The Demand for Urban Rail
Transportation," Journal of the American Statistical
Association. 74 (September 1979): 576-587.
^Melvin M. Webber, "The BART Experience-What Have
We Learned?", The Public Interest. 45 (Fall 1976): 120-
129 .
^Denver Regional Council of Governments, "A
Summary of the Regional Growth and Development Plan for
the Denver Region", Denver, Colorado in 1984.
^"Traffic Volume Map Colorado State Highway
System," prepared by Colorado State Department of
Highways Division of Transportation Planning in 1984.
^Webber, 125.
^Reuben Gronau, "The Effect of Travelling Time on
the Demand for Passenger Transportation," Journal of
Political Economy. 78 (1970): 378-379.
^M.E. Beesley, "The Value of Time Spent in
Travelling: Some New Evidence," Economics. (1965):

^Gronau t 3 7 9.
^"Colorado State Rail Plan," prepared for the
Legislative Council of the General Assembly by the State
Department of Highways in cooperation with the Federal
Railroad Administration, January 1979, URS Company,
V 5 0 .
12Ibid. V 40.
^Statistical Abstract of the United States 1985.
105th Edition/United States Department of Commerce Bureau
of Census Table No. 1067, "Railroad Accidents," 611.

In order for the Denver to Fort Collins train to
become feasible a modern train needs to be developed.
The Rio Grande Zephyr was a conventional train. The
Colorado State Advisory Committee, in its proposal to run
a train between Denver and Glenwood Springs, used
conventional wisdom. If the Denver to Fort Collins train
is to become successful, newer methods need to be
employed. First, labor practices must change. Second,
modern technology, which has provided us with new equip-
ment that is more efficient and less costly to run than
the vehicles used by the Zephyr, must be utilized.
Finally, operating expenses need to be minimized in a
variety of different ways. This final category is very
significant in order for the Denver to Fort Collins train
to be economically possible.
Pers onne1
One of the largest problems involved in operating
a successful commuter train is restrictive labor
practices. The current standards for a minimal crew, as
defined by the railroad unions, are outdated and

unrealistic because the availability of modernized
equipment has reduced the number of workers needed to
operate a train safely and efficiently.
An example of the inefficiencies pervasive in
today's train operations is the current requirement of a
fireman's position. When the diesel engine, which ran on
oil, was introduced in 1936, the fireman's job should
have been been eliminated as railroads no longer needed a
worker to shovel coal. However, the position was
protected by the union, and so it continued in spite of
the fact it was superfluous. Even a government panel
established by the Kennedy administration in 1963 was
unable to receive union endorsement of its decision to
eliminate 90% of the firemen from freight trains.
Because the union opposed the decision, no fireman's jobs
were eliminated.^ At that time, the government chose not
to intercede.
The existence of the firemens' positions on
freight and passenger trains continued until 1985 when
the Reagan administration appointed an emergency board to
try to win acceptance of a compromise contract settlement
between the railroad line and 130,000 member United
Transportation Union (UTU), the largest union
representing train crews. Once again, the fate of the
firemen was a key point of contention. Although the
firemen were not needed, the negotiating panel determined
that freight trains would maintain the firemens'

positions, while passenger trains could run without
As a result of the agreement reached between the
railroad companies and the union, AMTRAK trains on the
Washington-to Bos ton corridor today operate with a lone
engineer and a conductor with one assistant conductor.
This reduction in crew is a step in the right direction,
and a similar crew configuration would enable a commuter
train from Denver to Fort Collins to be a fiscal
In addition to the elimination of the firemen,
conductors' positions could also be eliminated or
minimized. Instead of using a conductor to take tickets
inside the cars, ticket cancelling machines could be
placed at each entrance. As a passenger enters, he
inserts the pre-purchased ticket in the machine to record
the date and time. The passenger keeps his ticket and no
further stamping is necessary. In other words, the
passenger is on his honor to stamp his ticket on boarding
and to travel no farther than its value permits. This
system is in use throughout many countries in Europe on
numerous bus lines. Such a fare system could be a key
characteristic in enhancing high productivity of the
commuter rail mode of transportation. The question is:
Would the procedure for tickets work in the Denver to
Fort Collins corridor? The answer is: Yes, it would,
even if abuse of the honor system approached the 1-2%

estimated by Europeans,^ Quite possibly, the enforcement
of this honor system could be implemented by one of the
employees on the train.
In order for the modern train to operate safely
and efficiently, two crews of two men would be able to
run three round trips between Denver and Fort Collins.
According to the current labor statistics, the average
hourly earnings in the transportation industry as of June
1985 were $13.71 wage rate per hour.^ So, on the basis
of two-thousand hour work year per employee, the total
labor costs, which include benefits of 47%, for the
Denver to Fort Collins passenger train would be
approximately $160,000 annually. This cost results in a
savings of almost $150,000 from the proposal of the
Denver to Glenwood Springs train, if the compensation of
an engineer and a conductor are added together.
Although the economics of labor personnel are
simple, political negotiations are not. Labor unions do
not prefer to change their position on the number of
personnel necessary to operate a train. However, if the
passenger train between Denver and Fort Collins is to be
successful, the railroad company and the carrier of
passengers need to negotiate an agreement that is
acceptable to the labor union. Therefore, the political
ramifications of these negotiations are a major factor in
the reduction of personnel, which is necessary for the
success of the Denver to Fort Collins passenger train.

Capital Equipment
In addition to problems with excessive numbers of
personnel, some passenger train companies face the diffi-
culty of operating with outdated equipment. For example,
there are lines still using passenger cars born at the
turn of the century which do not possess modern
conveniences. If a passenger train is to be successful,
it must have sufficient heating, proper ventilation,
aesthetically pleasing surroundings, and minimal noise.
In addition to the antiquated passenger cars still in
use, many outdated locomotives also are being used.
Although these locomotives are sufficient for hauling
freight, they are inefficient for passenger traffic
because they are too slow and cumbersome. Today,
lighter, faster equipment has been developed and could be
utilized to make a Denver to Fort Collins train a viable
option for the riders in this corridor.
The technology of comfortable, energy efficient
vehicles is available but there has to be a willingness
and available funds for companies to use this modernized
equipment. For example, when Ansco Investment Company, a
subsidiary of Rio Grande Western Railroad, took over the
operation of the Ski Train to Winter Park from Denver's
Union Station in the summer of 1987, one of the first
changes implemented in order to improve ridership was the
replacement of old ski train passenger cars with newer
stock bought from the Canadian government.

Another company which successfully modernized its
operation was British Rail (BR). Under the leadership of
Sir Peter Parker, BR opened, or in some instances
reopened, its twenty-six railway stations, many of them
rural lines. A major reason for the success of BR was
its use of the railbus, a mass produced bus body,
developed by British Leyland in 1983, which was placed on
a railway underframe. Specifically, the Railbus Class
141 used by BR was composed of a two-axle high
performance suspension system and turbo charged diesel
engine.^ The composition of a railway underframe with a
high quality suspension system enabled the railbus to
accommodate any gauge of railroad. In addition to
railroad track compatibility, greater traction was
provided by the underfloor engines and transmissions
which were part of the new railbusses. Thus, a basic
marriage occurred between bus and rail technology which
brought not only more efficient vehicles but also safer
vehicles to the transportation industry.
An additional safety feature of the railbus was
the fact that the drivers who operated the train had the
convenience of radio based signalling from train to
train. If a commuter train is to operate safely on the
proposed Denver to Fort Collins line, a line which it
would share with freight trains, radio communication
between the commuter train and the freight train is
essential. With the assistance of radio based

signalling, the operator of the passenger train would
know the exact position of any freight train on the
Burlington Northern line at any given point in time.
Such a communication network would assure the passengers
of a safe ride and enable the driver of the train to move
the railbus from city to city without costly time delays.
If a radio communication system does not exist, it is
quite possible, with the unpredictable nature of freight
train departures, that the passenger train would be at
the same position on the track as a freight train at a
given point in time.
A final factor which assures safety of the
passengers and enables the train operator to move without
stopping is the right of way. The automation of level
crossings, which are road crossings level with the train
tracks, would minimize potential accidents and provide
the operator with the right of way. Also, the
installation of warning systems within the railbus would
be essential so as to prevent potential disasters.
All of the safety features previously mentioned
bring security to the passenger riding the train. Also,
these features provide efficiency in operating the
railbus at speed comparable to the automobile or bus. In
order for the Denver to Fort Collins train to be a viable
alternative to the automobile, the operator of the
railbus must not consume more time than the driver of an
automobile travelling between Denver and Fort Collins.

In addition to their safety features, the
railbusses have other attractive qualities. The busses
all have lavatory facilities and vestibules for baggage
which are large enough to accommodate baby carriages and
bicycles. A spacious baggage area is complemented by
seating which comfortably accommodates ninety-four
passengers. Conventional passenger cars have a capacity
of sixty or seventy-eight seats per car.
Thus, a successful commuter train between Denver
and Fort Collins would.need state of the art
communication equipment, which could be adapted to the
railbus easily; automatic warning systems; automated
crossing gates; and creature comforts. The cost of
purchasing such a two car unit railbus based on 1982
prices was 160,000 pounds (US $335,000).^ According to a
United Kingdom producer price index for railroad
equipment a 20% increase occurred between 1982 and 1986.^
So, the cost of a British Leyland Class 141 railbus would
have been 192,000 pounds in 1986. In 1986, the average
exchange rate was $1,2963; therefore, the cost of the
British railbus to a United States carrier would have
been approximately $250,000 for both two car units with a
total seating capacity of 188 seats.
In summary, the Denver to Fort Collins train
could initiate its trial run with two Class 141 railbus
units from British Leyland at a cost of a half a million
dollars. Also, an additional $100,000 contingency fund

is necessary for radio based signalling, automatic
warning systems, arid automatic crossing gates. In
conclusion, the total capital expenditure for the Denver
to Fort Collins train would be approximately $600,000.
If this amount is amortized over a period of ten years at
a 10% interest rate the expenditure for all capital
equipment in the first year of operation would be
approximately $95,000.
In addition to reducing personnel and minimizing
capital expenditures, operating expenses need to be
streamlined. The primary costs of operating a railbus
between Denver and Fort Collins are fuel, maintenance and
repair, and leasing expenses with Burlington Northern for
track and facilities use.
The largest expenditure for operations is fuel.
The anticipated cost for locomotive fuel for the Denver
to Fort Collins commuter train could be based on the
total number of miles travelled.' That is, three round
trips at 132 miles each for a total of 396 miles. The
Rio Grande Zephyr travelled 1,140 miles round trip
between Denver and Salt Lake City. So, a factor of 0.35
is used to compute the locomotive fuel cost from the 1980
balance sheet. In 1980, the annual locomotive fuel cost
was $567,408 for the Rio Grande Zephyr. Also, in 1980,
#2 diesel fuel, which was used by the Zephyr locomotives,

was $.818 per gallon. In December 1986, the same type of
fuel was $.823 per gallon.^ This cost for fuel is not
significantly different than the 1980 price, and it is
the same type of fuel used in the railbus. Therefore,
0.35 of $567,408 is approximately $200,000 for fuel costs
to run the Denver to Fort Collins train.
The second expense to operate the new railbus
from Denver to Fort Collins is maintenance and repair.
If the new cars purchased are maintained properly and
checked at regular intervals this maintenance expense
should be minimal and far less than the Rio Grande
Zephyr's maintenance cost. The allotment for maintenance
and repair of the equipment is $100,000.
The final expense for operating the train is an
appropriate arrangement with Burlington Northern to use
its branchline track and station facilities. The Lease
Arrangement cost estimated by DRGW to use its tracks and
facilities in the Denver to Glenwood Springs corridor was
a half a million dollars in 1980. If this cost is a
function of the depreciation of the tracks, then the
amount of mileage between Denver and Fort Collins, which
is 132 miles for a round trip, should be used. If a
round trip on the Denver to Glenwood Springs train would
have been 370 miles, then a factor of 0.36 should be used
to compute the leasing cost of the Burlington Northern
tracks that make up the corridor from Denver to Fort
Collins. Also, an adjustment needs to be made for the

increase in leasing arrangements from 1980 to the
present. The lease arrangement for the Denver to Fort
Collins train should not exceed $200,000.
In conclusion, the Denver to Fort Collins train
is a new train. A two man crew is sufficient personnel
to operate the modern train efficiently and carefully.
Both employees should be able to operate the train.
While one man operates the train, the other can serve as
a ticket conductor to enforce the honor system and as a
service attendant, or a relief driver, if necessary. The
Rio Grande Zephyr had separate positions for each duty,
there is no need for this on the Denver to Fort Collins
Another modification, in addition to altering
labor practices, is the utilization of a railbus. As
evidenced by the Ski Train to Winter Park, riders can be
attracted by new equipment. The railbus, which is
affordable, has proven itself in Great Britain, and it
can be adapted to meet the needs of commuters in the
Denver to Fort Collins corridor.
Finally, the changes in the operation of a
passenger train are necessary if the new train is to
become successful. The past history of the Rio Grande
has indicated the unprofitable nature of the dining and
buffet service; therefore, the elimination of this
service should be obvious for the Denver to Fort Collins
train. Also, fuel consumption should be less relative to

the Zephyr because of the use of the light railbusses.
The new vehicles should need fewer repairs than the
Zephyr cars and be less damaging to the tracks because of
their weight. Therefore, the final operating expenses,
maintenance and repair, plus the lease arrangement should
be minimal.

^-Richard Corrigan, "Railroads Keep Striking Out
in Efforts to Rid Their Trains of Firemen", National
Journal. 17 (40), September 5,1985: 2242.
2 lb id. 2245.
8Urban Transportation Another Alternative "A
World Wide Survey of Light Rail Technology", by the
Heritage Foundation, 13.
^Corrigan, 2243 (Bureau of Labor Statistics).
^Railbus Systems published by Mechanical
Engineering Publications limited on November 24, 1982,
29 .
8Paul Goldsack, "Railbus Rides to the Aid of
British Rail Commuters", Mass Trans i t. 8 (9):
September 24, 1981, 17.
^Monthly Bulletin of Statistics. 40 (12)
December 1986; Table 56: Index Numbers of producer prices
and wholesale prices; published by the Statistical Office
of the United Nations, 188-9.
8 lb id. 192.
^Monthly Energy Review. May, 1987, Energy
Information Administration, 97.

If a passenger train between Denver and Fort
Collins is to be a successful venture, those in charge
must use the train track currently in existence and avoid
the mistakes made by other passenger train carriers.
These mistakes fall into three basic categories:
excessive expenditures for labor, use of outdated
equipment, and excessive operating expenses.
The ideal train track for the Denver to Fort
Collins train is Burlington Northern's branchline. This
line provides the highest potential for ridership and
does not hinder the movement of freight trains along the
route. In addition, the utilization of existing train
tracks keeps expenditures for the passenger train to a
Another way to keep costs to an absolute minimum,
and thus make the Denver to Fort Collins train
successful, is to avoid excessive expenditures for labor.
In the past, the Denver and Rio Grande Western (DRGW) ran
the Rio Grande Zephyr with a five man train crew and two
service crew, who operated an unsuccessful dining and
buffet service. From 1979 to 1982 the revenue from the

dining and buffet service (see Table 4) fell far short of
the expenditure for this service (see Table 3). The
Colorado State Rail Advisory Committee should have
learned a lesson from DRGW's mistake. Instead, it
increased the service personnel for the proposed train
from Denver to Glenwood Springs. This increase in labor
inflated costs, which was one reason why this train never
came to fruition. So, the lesson to be learned from
either mistake is to eliminate the dining and buffet
service on the Denver to Fort Collins train. By
eliminating this service, train personnel are reduced,
thereby minimizing costs for the passenger train.
Expenditures for personnel can be reduced even
further if the Denver to Fort Collins passenger train
makes use of the new technology provided by British
Leyland, which developed the Class 141 Railbus. This
train would run very efficiently using the railbus which
would need only a two man crew. Both crew members would
hold a dual capacity. While one person operates the
railbus, the other would be a service man selling
refreshments from a portable cart, similar to the ones on
airplanes, and would serve as a fare conductor to enforce
the honor system of ticket fares. During a five day work
week, thirty one-way trips would be run according to the
proposed schedule. So, another two man crew would be
necessary to divide up the total number of trips in one

week. The estimated cost for these two crews of two men
on each crew is approximately $160,000 per annum (see
Table 10).
The second mistake which operators of a Denver to
Fort Collins transit line must avoid is the use of
outdated equipment. The cost of purchasing new equipment
would be less than maintaining rebuilt equipment. Two
British Leyland Class 141 railbus units would cost
approximately $500,000. Another major cost saving
feature of the new railbus for the Denver to Fort Collins
train is its compatibility with any gauge of railroad
track. This feature would make it easily adaptable to
the Burlington Northern line and, once again, minimize
costs. In addition to the fact that the use of modern
equipment minimizes costs, it also has been found to
attract riders. For example, when the DRGW's Ski Train
to Winter Park began using attractive new vehicles, it
experienced an increase in ridership. A similar
situation occurred on British Rail's urban and rural
lines when new equipment was introduced. Another piece
of modern equipment, which is available for use on the
Class 141 railbus is radio based signalling. Although
additional money would need to be spent, the radio based
signalling feature ensures the safety of the passengers
and maintains the proper flow of traffic among all trains
using the Denver to Fort Collins corridor. So, a
contingency fund of $100,000 is to be allotted for radio

Table 10
Estimated Total Cost for the Denver to
Fort Collins Train
2 Two man crews
(a) $27,420 annually @4... $110,000
(b) Plus: Benefits of 47%.. $ 50,000
Total Labor Costs
2 British Leyland Class 141
Railbus Cars @ $250,000
Contingency Fund
Total Capital Costs
Amortized Capital Cost
Operating Costs
Lease Arrangement
Maintenance and Repair
Total Operating Costs
$ 95,000
Total Costs for the Initial Year
of Operation for the Denver
to Fort Collins Train

based signalling. Therefore, the total capital costs
amortized over a ten year period at a 10% interest rate
would be approximately $95,000 (see Table 10).
In addition to reducing labor costs and making
wise capital investments, the operators of the Denver to
Fort Collins train must keep operating expenses as low as
possible. First of all, fuel consumption is not just
based on mileage, as the weight and speed of the vehicles
can modify fuel consumption. Even though the railbus
might weigh less than older conventional vehicles, the
speed necessary to maintain on-time service might negate
the gain provided by lighter railbus vehicles. So, the
estimate for fuel costs is approximately $200,000 based
on the pro-ration of mileage.
Another operating expense which can be minimized
involves the lease arrangement between the carrier of
passengers on the Denver to Fort Collins train and
Burlington Northern. This amount should not exceed
$200,000. Also, maintenance and repair of new equipment
should be minimal, thus $100,000 is allotted for this
expense. The last category is classified as
miscellaneous: it covers train supplies, insurance, and
ticket distribution. So, the total amount budgeted for
operating costs is $650,000 annually. Therefore, the
minimal amount necessary to operate a passenger train
from Denver to Fort Collins is $905,000 for the first
year of operation.

The revenue needed to cover these minimal
expenditures could be generated by diverting a small
percentage of the vehicular traffic travelling the
various highway corridors. For example, 2% of the 10,700
vehicles, the Average Daily Traffic count between Denver
and Fort Collins, is 214 one-way tickets per day. If
five dollars is the price of the ticket and 260 trips are
made each year, then the revenue generated by these
riders is approximately $280,000 annually.
In addition to the passengers that ride the
Denver to Fort Collins train for the complete trip, some
people might utilize this train for partial segments of
the Denver to Fort Collins corridor. The passengers from
Denver to Boulder or from Longmont to Fort Collins should
not be charged full fares. A fair price based on the
pro-ration of mileage of these corridors is two dollars
and fifty cents for either passenger. In order to assure
every rider on the train a seat, a smaller percentage of
the ADT counts is necessary. So, 1.5% of these ADT
counts are 575 and 143 riders respectively for these
partial segments. The total ridership at the two dollars
and fifty cents' fare is 718 per day. Therefore, the
anticipated revenue from these passengers is
approximately $470,000 per year.
The last type of rider on the Denver to Fort
Collins train is the traveller from Denver to Broomfield
or Boulder to Longmont. Likewise, a fair price for these

riders is one dollar and twenty-five cents. Once again,
1.5% of these ADT counts results in 285 and 179 riders
respectively per day. Therefore, a total of 464
passengers per day would generate an additional yearly
income of $150,000.
If the seating capacity of the new train from
Denver to Fort Collins is 1,128 per day, these revenue
figures are possible. In summary, the total revenue
generated by the passengers of this train is $900,000 per
year. This amount is sufficient to cover the minimal
expenditures necessary to operate a modern passenger
train between Denver and Fort Collins.

Beesley, M.E. "The Value of Time Spent in Travelling:
Some New Evidence." Economica 32 (May 1965):
174-85 .
Black, John. Urban Transport Planning. Baltimore:
The John Hopkins University Press, 1981.
Colorado Department of Highways, Division of Transporta-
tion Planning, Rail/Transit Section. Colorado
State Rail Plan 81-82 Update. Denver: 1983.
________. "1984 Traffic Volume Map Colorado State
Highway System." 1984.
Corrigan, Richard. "Railroads Keep Stiking Out in Efforts
to Rid Their Trains of Firemen." National Journal
17 (September 1985): 2242-45.
Demoro, Harry W. "New Rail Transit the Economic Payoffs."
Mass Transit 10 (February 1983): 22-33.
Denver Regional Council of Governments. A Summary of
the Regional Growth and Development Plan for the
Denver Region. Denver: 1984.
Goldsack, Paul. "Railbus Rides to the Aid of British Rail
Commuters." Mass Transit 8 (September 1981):
Gronau, Reuben. "The Effect of Travelling Time on the
Demand for Passenger Transportation." Journal of
Political Economy 78 (1970): 377-94.
Hensher, David A. and Truong P. Truong. "Measurement of
Travel Time Values and Opportunity Cost From a
Discrete Choice Model." The Economic Journal 95
(June 1985) : 438-51.
Kain, John F. "A Contribution to the Urban Transportation
Debate: An Econometric Model of Urban Residential
and Travel Behavior." Review of Economics and
Statistics 46 (1964): 55-64.
Kraus, Dorothy. "Val Vive La Difference!" Mass Transit 11
(September 1984): 46-54.

Lane, Bob. "Vancouver's LRT." Mass Transit 12
(July 1985): 14-33.
McDonough, Carol C. "The Demand for Commuter Rail
Transport." Journal of Transport Economics
and Policy (May 1973): 134-43.
Moses, Leon N. and Harold F. Williamson. "Value of Time.
Choice of Mode, and the Subsidy Issue in Urban
Transportation." Journal of Political
Economy 71 (1963): 247-64.
Pegrum, Dudley F. Transportation: Economics and Public
Policy. Georgetown: Irwin-Dorsey Limited,
1973 .
Statistical Office of the United Nations. Monthly
Bulletin of Statistics 40 (1986): 188-92.
Stopher, Peter R. and Arnim H. Meyburg. Urban
Transportation Modeling and Planning.
Lexington: D.C. Heath and Company, 1975.
The Denver and Rio Grande Western Railroad Company.
"Revenues, Expenses and Net Earnings for the Rio
Grande Zephyr Passenger Trains Nos. 17 and 18."
Denver: 1983.
The Institution of Mechanical Engineers, Automobile and
Railway Divisions. Railbus Systems. London:
Mechanical Engineering Publications Limited,
1982 .
United States Code Congressional and Administrative
News. 91st Congress, Second Session.
Washington, 1970.
U.S. Department of Commerce. Bureau of Census.
Statistical Abstract of the United States:
1985. 105th edition. Washington, 1986.
________. Statistical Abstract of the United States:
1986. 106th edition. Washington, 1987.
U.S. Department of Energy, Energy Information
Administration. Monthly Energy Review.
Washington, 1985.
U.S. Department of Labor. Bureau of Labor Statistics.
Monthly Labor Review. Washington, 1985.
"U.S. People Movers: An Update." Mass Transit 12
(March 1985): 44-5.

Webber, Melvin M. "The BART Experience What Have We
Learned?" The Public Interest 45 (1976): 120-29.
Westley, Glenn. "The Demand for Urban Rail
Transportation." Journal of American Statistical
Association 74 (September 1979): 576-87.