Gottschalk, Petter, "Railroad Decline and Revitalization", 1981

RAILROAD DECLINE AND REVITALIZATION

Petter Gottschalk
Resource Policy Group
Sagveien 21 Oslo 4, Norway

Introduction

The purpose of this paper is to introduce an integrated
framework for lang-range strategic planning in a railroad. The
framework is a computer simulation model designed to be useful
to most freight-hauling railroads. The model can help to
increase the understanding of problems facing the railroad and
to aid in developing strategies for addressing these problems.
It is designed to forecast railroad performance and to aid in
developing more effective policies for railroad management. It
can also be used by Federal agencies to evaluate impacts of

public policy on railroad performace.

The Model
The model represents dynamic interactions between
railroad track and equipment management, freight market share
and internal financial variables. It is divided into four
sectors: track module, equipment module, market module, and

finance module.
The track module traces amounts of standard and sub-
standard track in the system. Standard track is defined as

track without deferred maintenance, substandard track is track

with deferred maintenance. Track wear.- such as deterioration
of rails, ties and ballast - causes track to move from standard
to substandard condition. Track maintenance - such as rail

replacement, tie replacement, ballast conditioning, and track
surfacing - causes track to move from substandard to standard
condition. Track souerruceien increases the amount of standard
track. Track abandonment reduces the amount of both standard

and substandard track.

The equipment nodule traces the number of new and old
rail cars and the number of rail cars needing maintenance.
The number of new cars increases by car purchases and decreases
by aging of new cars. The number of old cars increases by
aging of new cars and decreases by aging of old cars (car
retirement) and by car sales. The number of cars needing
maintenance increases by car wear and decreases by car main-

tenance.

The main activities in the market module are the
calculations of revenue ton-miles and market share. Revenue

ton-miles is the product of total intercity freight in the
s 3 =

region and the railtoad's market share. The amount of regional
intercity freight is a function of the region's contribution

to GNP. The railroad's market share is determined by the
railroad's and the competitor's shipping rate and service
levels. The model distinguishes between the following service

components: reliability, frequency, and operating speed.

The finance module represents mangement accounting in
a railroad. the money balance (liquid assets) increases
through revenues and issuance of new debt and decreases
through operating expenses and costs and repayments of debt.
Increase in debt is determined by desired increase in debt
and by the debt-equity ratio, where equity is the sum of all
assets. Money adequacy, which is a variable in the finance
module, indicates the availability of funding and influences

the allocation of money to track and equipment maintenance.

Some of the feedback loops in the model are shown on
the next page. The main positive feedback loop indicates
that a reduction in freight volume leads to a reduction in
money balance which causes cutbacks in maintenance of track

and equipment.

As a consequence, the fraction of property in standard

condititon declines, and the resulting reduction in service
aoe als

NUMBER OF

RAIL CARS
o
RAILCAR
(UTILIZATION
©
FREIGHT _
* VOLUME \
MARKET MONEY
SHARE BALANCE <) ~—CDEBT
. ?
»
+
MAINTENANCE OF
SERVICE TRACK AND EQUIPMENT

te
i
FRACTION OF PROPERTY

IN STANDARD CONDITION
@ |
TRACK WEAR

Some of the feedback loops in the rail-
road model.
reduces the railroad's market share. A reduction in. market

share causes further decline in freight volume.

Consolidated Rail Corporation

Since the end of World War II, many US railroads have
been deteriorating physically and financially. Consolidated
Rail Corporation (Conrail), which began operations in 1976,
has come into being as a result of this trend. Conrail is a

consolidation of the rail properties of six bankrupt railroads.

Despite receipt of 3.3 billion dollars in federal
loans, Conrail remains in financial trouble. Additional
federal loans, rate deregulation, additional coal freight,
track abandonment, and productivity improvements are some of
the policies being considered to help Conrail and other
railroads out of their difficulties. To evaluate the impacts
of such policies on Conrail's performance, the system dynamics
model for long-range strategic planning was used by Conrail's

Planning and Control Department.

In 1960, Conrail's predecessors carried an estimated
280 million tons an average of 375 miles, which gives a
freight volume of 105 billion ton-miles. The total intercity

freight in the region covered by the predecessors was estimated
to be 620 billion ton-miles, leaving Conrail with a market
share of 17 percent. The average price per ton-mile - the
shipping rate - was 4 cents, and the net income was close to
zero. We assume that Conrail's predecessors had the same
amount of track as Conrail has today: 40,000 miles, of which
10,000 miles were estimated to be in substandard condition in
1960. We also assume that Conrail's predecessors had 150,000

vail cars, of which 105,000 cars needed maintenance in 1960.

These numbers are the. initial conditions with which
the model simulation starts in 1960. The financial data are
in constant 1980-dollars. If the railroad's operating
environment had remained at 1960-levels, then it is assumed
that the railroad performance would have remained unchanged,
which means that the model is initialized in equilibrium in
1960. (This assumption is probably too simplistic. Although
most: railroads were making profit in 1960, they were overbuilt

for the traffic volume at that time.)

However, several changes took place in Conrail's
operating environment in the 60's and the 70's, creating the

model behavior shown in the base simulation run on page ll.
First, the GNP-contribution of the Eastern District
(in which Conrail operates) showed a steady growth in the 60's,
which led to growth in Conrail's freight volume despite
decreasing market share. In 1970, however, the growth stopped
and a slow decline has occured in the Eastern District of the
US ever since. The decline can be derived from the fact that
not just railroad freight, but also truck freight has been
declining in the Eastern District since 1970. This decline
in economic activity combined with the decline in market
share, led to a steep decline in Conrail's freight volume in
the 70's. Also, Conrail has the impression that rail-oriented
activities, such as heavy manufacturing and coal mining, have
declined even faster than the averaye economic activity in

the Eastern District.

Second, the rate of inflation increased in the 60's
and the 70's, causing railroad costs to rise faster than
shipping rates. It is assumed in the model that the net
effect of inflation on railroad performance is a lag in
shipping rate behind operating costs. When costs go up, then
the railroad applies for a rate increase. The rate increase
is approved by the Interstate Commerce Commision (ICC) only
after a considerable delay. This delay implies that the

shipping rate will always lag behind cost increases. This lag
will be greater the faster costs grow. In the base simulation
run, this effect is seen .in the declining real value shipping

rate over the period from 1960 to 1980.

Third, the competition from other modes of transporta-
tion - trucks in particular - has become more significant.
Trucks have been able to reduce their rates relative to those
of railroads. Also, substantial improvements in service
reliability, service frequency and operating speed have been
achieved in the trucking industry. These changes have caused
the trucking industry to increase its market share at the

expense of the railroad industry.

Finally, federal investments - totaling 3,3 billion
dollars - were spent by Conrail to improve property quality.
With the federal money, the car maintenance backlog was almost
eliminated and the increase in substandard track was halted.
These improvements led to reductions in operating losses
(i.e-negative net income), but market share (here Conrail's
share of the total freight market, not Conrail's share of the
vail market) and freight volume continued to decline. (It must
be noted that the calculated net income is far from exact

because it is very small compared with revenues.)
When used to simulate the period 1960 to 1979, the model
projects the following operating data for 1979 (the actual

number - when available - is put in paranthesis):

- freight volume of 90 billion ton-miles (93 billion)
- market share of 14 percent

- shipping rate of 3.8 cents per ton-miles (3.8 cents)
- 17,000 miles of substandard track

~ 118,000 rail cras (118,000 rail cars)

- 20,000 rail cars needing maintenance

- no net income (loss of 178 million dollars).

The numbers are used as initial conditions for pro-
jections of Conrail performance. For Conrail's operating
environment it is assumed in the projections that the eco-
nomic. activity in the Eastern District of the US will con-
tinue to decline slowly, that the inflation rate will remain
at the 1979-level, that there will be no federal investments
after 1980, and that the railroad's competitive environment
will not become worse, i.e. that truck rates and service levels

remain constant relative to rail.
-10-

The base simulation run on the next page shows
projected Conrail performance from 1979 to 2000. Conrail's
market share continues to decline because of continued track
deterioration. Conrail's freight volume continues to decline
because of declining market share and because of the assumed
decrease in regional economic activity. The operating loss
remains insignificant. In the year 2000, Conrail's freight
volume is 70 billion ton-miles and substandard track amounts

to 24,000 miles, more than half of the total track.

Policy Analysis

A revitalization of Conrail seems unlikely to occur
unless new policies are implemented. No doubt, the federal
investements have helped Conrail, but they have only prevented
Conrail from a steeper decline, they have not turned Conrail's

financial and physical condition around.

Additional Federal Investements

Conrail applied for additional federal investements
in 1980. The simulation on the page 12 (bottom), assumes
that Conrail receives another 2 billion dollars. The money
is mainly spent on track maintenance between 1981 and 1985.
The simulation shows a significant reduction in deferred

track maintenance: the amount of substandard track is reduced
= 312

NET FED. SHE
ANC. INV RATE.
400M BOOM £.3¢

"FREIGHT VOLUME

ZFEDERAL INVESTMENTS
108 atm. JN _NET INCOME || SM Ate

yy E

SHIPPING RATE _

200 LOOM 3.9¢

8 MARKET SHARE”

3% -— -
1860 1370 1980 1380

——80M 0 35¢
2000

Base simulation run showing freight volume (billion
ton-miles per year), federal investments (million
dollars per year), net income (million dollars per
year), shipping rate (cents per ton-mile), and
market share (percent).

CARS, TRACK
2407 7 %: 2 - = GBT

IOTAL TRACK

‘BOT 361

TOTAL CARS

1207 a

CARS NEEDING—~*s

MAINTENANCE
os Cn 1980 s 99” ~ 2000

Base simulation run showing total track and track
needing maintenance (thousand miles), and total cars
and cars needing maintenance (thousand cars).
208

_
— AS
. oN “tos

~ ~
fst sae eenaee Rachie 9 wieares 8 A aK sneer
; ~~) ~

MARKET SHAE

goon aot

100M GOGM 4.25

gee 1970 1see 1990 2000

Base scenario

808

50B

208

L FREIGHT VOLUME
po FEDERAL INVESTMENTS.

254 ‘ . . :
Ce NET INCOME
Neal 7 ae -——o
wa —

gohiere RATE

Additional federal investments

106

ET FED.

4 GOON 4

ye
- 13 -

from 17,000 miles in 1980 to 9,000 miles in 1985. This improve-
ment in railroad condition increases the service and, as a
consequence, Conrail's market share starts to rise. The
increase in market share compensates for the decline in regional
economic activity and the freight volume is stabilized at 85
billion ton-miles. Net income, however, shows no significant
improvement since the shipping rate is only sufficient to

cover unit operating costs.

Additional Coal Freight

The energy crisis may force utilities and other indu-
stries in the US to make significant shifts from oil to coal.
As a consequence, the demand for coal transportation will
likely rise. Traditionally, railroads have hauled twothirds
of the coal (measured in tons). Competing modes, such as
slurry pipelines, may pick up the growth because of cost advan-
tages although the high fixed costs probably will prevent the
construction of such pipelines in the near future. Also,
large-scale pipeline construction could remain impossible,
because railroads and environmental interests continue to be
successful in blocking eminent domaine legislation. Thus, an
increase in Conrail's coal freight is likely. In 1979, Conrail

hauled 78 million tons, representing 29 percent of Conrail's
total tonnage. In the simulation run on the next page

it is assumed that a doubling of this quantity will occur by
1990 (156 mill. tons) and a trippling by 2000 (234 mill.
tons). It is assumed that the. freight of other commodities

will respond as in the base scenario.

The simulation on the next page shows steep. growth in
market share and freight volume. Net income, however, shows
ae initial (though insignificant) decline because operating
expenses exceed operating revenues. Around 1990, additional
unit coal trains increase productivity and reduce unit operating
expenses, causing net income to go up- At the same time funding

is generated to reduce deterred track maintenance.

It is difficult to predict how fast unit operating
expenses can be reduced by additional coal freight, but the
projection presented here may seem too pessimistic. If a
faster reduction is assumed, then the initial decline in net

income will not occur.

Track Abandonment
Conrail operates approximately 40,000 miles of track and
20,000: miles of road (there are two track-miles per road-mile

on average). Many of the areas served by Conrail have lost
505

20B

12108

808

50B

208

254

208

Sf SHU IS RATE
¥

Lo ee ee ee mw

1960. 1970 ised 1985 2000

Base scenario

C FREIGHT VOLUME
FEDERAL
L INVESTIENTE

100m

206%

a.

gen nunines TAK gl whem a 3 8 SS = feneN
Narwer srarc* ~ 4 stem RATE
08H
1960 ive ayso 2999 2000

Coal transportation

600m

arom

eo

‘x

500M 20 4.75

sie
their industrial base. So far, however, the Interstate
Commerce Commision has prevented Conrail from abandoning the
track in these - or any other - areas. Deregulation of the
railroad industry and/or federal intervention may change this
situation. Since Conrail has stopped maintaining track where
it would like to.abandon the line, the simulation model
assumes that track is abandoned in respons to the amount of

substandard track.

On the next page, the model's simulation of this
abandonment policy is shown. The total number of track miles
is reduced to 27,000 miles by the year 2000. The main effects
of track abandonment are that only a slight reduction in
freight volume will occur, while operating expenses show

significant improvements, leading to growth in net income.

In the simulation, market share goes up while freight
volume goes down. The reason for this result is that market
share is calculated based on the region covered by the
railroad. When track is abandoned, the region covered by the
railroad decreases in the model proportionally, while only

marginal freight is lost leading to an increase in market

share.
Fat.
Yoh

sob

1iop

BoB

508

208

10¥

1970

‘ise0 1990 2000

Base scenario

FEDERAL
INVESTMENTS.
b wor.

_ ; a
aA- ner incone

_-

200M

‘i ——

MARKET SHARE.

1960

wy

rose ise suv

Track abandonment

600m

soe

200!

Other Policies

The US Congress was debating a deregulation bill which
- among other things - can give the railroads the freedom to
determine their own shipping rates. Given this deregulation,

the model shows that Conrail increases its average shipping

rate in response to negative return on investment.

Another policy is related to Conrail's productivity.
Compared with other major railroads, Conrail's productivity is
low. For example, the number of ton-miles per employee per
year is far below average. Conrail had a workforce of 87,000
in 1979. Through negotiations with the United Transportation
Union, Conrail has attempted to achieve furthre reduction in
train crew size. Labor expenses account for 50 percent of
total expenses. In a model simulation, it was assumed. that
labor productivity is improved by 10 percent by 1990 and by 20
percent by 2000 relative to the productivity level in 1980.
The reduction in operating expenses makes funding for property

improvements available and stops the decline in market share.

Comparison of Policies

The policies discussed in this paper are not actions
which depend on a single decision-maker. Most of them involve

federal authorities as well as Conrail. Additional federal
-19 -

loans are granted by the US Congress, but the spending of

such loans on competing railroad needs is decided by Conrail
management. Additional coal freight is mainly dependent upon
such factors as OPEC, the federal authorities' support to
industries which convert from oil to coal and pollution
standards, but Conrail's equipment and service will also
influence its coal freight. Track abandonment can either be
accepted by the Administration or decided upon by Conrail
itself in the case of deregulation, but political factors

will still influence its decision. The bill which will
deregulate the railroad industry must pass the US Congress,
but the use of rate freedom as yranted in the bill is derermined
by Conrail. Productivity improvements in Conrail's operations

will be achieved more quickly with cooperation of the unions.

Below, a summary of simulation results are
presented. All policies will improve both Contrail's financial
and physical condition compared with the base simulation in
the year 2000. Additional federal investments and coal
freight will improve Conrail's psysical condition, but reduce
its financial performance compared with the situation in
1980. Only track abandonment and productivity improvements
seem to revitalize Conrail, i.e. permanent improvement in the

railroad's physical and financial condition is achieved by 2000.
~ 20 -

Net Substandard
Income Track
(Mill.$) (Miles)

SITUATION IN 1980 0 17,000
BASE SCENARIO 2000 ae 150. 23,000
RATE DEREGULATION 2000 - 100 22,000
ADDITIONAL COAL FREIGHT 2000 - 80 13,000
ADDITIONAL FEDERAL LOAN S$ 2000 me 20) 10,000
PRODUCTIVITY IMPROVEMENTS 2000 + 120 13,000
TRACK ABANDONMENT 2000 + 340 2,000

Reference

Gottschalk, Petter: The Dynamics of Railroad Decline and

Revitalization. GRS-31l1. Resource Policy Group 1980. 159 payes.
CLASS II DOCUMENTATION STANDARDS

_——_ FOR SIMULATION MODELS

1.

2.

3.

ACCESS TO MODEL:

Name of Model: _ Railroad Simulation Model

Name and current address of the senior technical Petter Gottschalk, Resource Policy
person responsible for the model's construction: Group, Sagveien 21, Oslo 4, Norway

Who funded the model development? Stipend from Norwegian Government

In what language is the program written? DYNAMO

On what computer system is the model currently
implemented? IBM

What is the maximum memory required to store and
execute the program? — 5000 words

What is the length of time required for one typical
run of the model? at seconds

Is there a detailed user's manual for the model? Yes

PURPOSE OF THE MODEL:

For what individual or institution was the model
designed? Consolidated Rail Corporation

What were the basic. variables included in the model?

Track, equipment, freight, income and expenses

Over what time period is the model supposed to provide useful information on real
world behavior? * a

From 1960 to 2000
Was the model intended to serve as the basis of:
an academic exercise designed to test the implications of a set

of assumptions or to see if a specific theory would explain his-
torical behavior No

communication with others about the nature and implications of an
important set of interactions No

projecting the general behavioral tendencies of the real system Yes

predicting the value of some system element(s) at some future
point in time Yes

MODEL SPECIFICATION AND THEORETICAL. JUSTIFICATION:

Provide two diagrams illustrating the extreme behavior modes exhibited by the major
model elements:

Substandard track .
Freight volune
‘reight volume Substandard track
ees
1960 1980 2000 1960 1980 2000

If they are not included in the body of the paper indicate where the reader
may find:

a model, boundary diagram that indicates the important
endogenous, exogencus and excluded variables ____None.

a causal influence diagram, a flow diagram, the com-

puter program and definitions of the program elements GRS-311

Is the model composed of:

simultaneous equations No
difference or differential equations Yes
procedural instructions Yes
Is the model deterministic Yes or stochastic
continuous Yes _. or discrete

4. DATA ACQUISITION

What were the primary sources for the data and theories incorporated in the model?

Data Technical and financial data from Conrail

Theory Discussions with Conrail managers

What percent of the coefficients of the model were obtained from:
measurements of physical systems ___30
inference from social survey data
econometric analyses
expert judgment _ 60
the analyst's intuition 10

What was the general quality of the data?__ Good

5. PARAMETER ESTIMATION

If they are not given in the publication, where may the reader obtain detailed infor-
mation on the data transformations, statistical techniques, data acquisition proce-
dures, and results of the tests of fit and significance used in building and analyzing
the model? None. The computations are not documented

6. MODEL PERFORMANCE AND TESTING

Over what period was the model's behavior compared with historical data?

From_1960 to 1980
What other tests were employed to gauge the confidence deserved by the model?

Discussions with Conrail about model structure and robustnesstest

“"Where may the reader obtain a detailed discussion of the prediction errors and the
dynamic properties of the model? Partly in GRS-311. Petter Gottschalk:

The Dynamics of Railroad Decline and Revitalization Resource Policy Gronn 1980.

7. APPLICATIONS

What other reports are based upon the model? None

Name any analysts outside the parent group that have implemented the model on another
computer system. None

List any reports or publications that may have resulted from an evaluation of the
model by an outside source. None

Has any decision maker responded to the recommendations derived from the model?
Conrail's Planning and Control Department

Will there be any further modifications or documentation of the model? No
Where may information on these be obtained?