- 4

2.

Apportioned expenses, or those which in the first in-
stance were incurred jointly for both services but for
which a basis was prescribed. The expense of locomotive
repairs, a joint expense, was to be apportioned on a
prescribed basis in accordance with the performance in
each service.

3. Unapportioned expenses, or those which could not be di-
rectly assigned to one service or the other and for which
no basis of division was prescribed. For example, all
maintenance of way expenses, except those items incident
to tracks, stations, or other facilities regularly used
solely in one service, were to be reported as unappor-
tioned.

In its report of June 13, 1914, on the practicability of providing for the ascertainment of cost figures in the general manner indicated in Circular No. 3, the Commission specifically refused to abandon its policy of giving consideration to cost of railway services in establishing rates. The Commission agreed with the arguments advanced in support of a separation of expenses, namely: (1) that it would be a material aid in the determination of the cost of railway services, not only as between freight and passenger services as a whole, but also in determining the cost of particular classes of traffic for the reason that the separation of freight and passenger expenses is fundamental in all cost of service inquiries; (2) an authoritative formula separating operating expenses and leaving none unapportioned would simplify the work of carrier representatives in rate cases; and (3) the proposed reports would be of assistance in the study of comparative costs among railways. The Commission noted that manufacturers have found estimates of operating expenses regarding individual articles manufactured in a plant to be useful managerial tools in determining efficiency and in product pricing and, while differences probably exist, the same principles should be true for the railway industry. The Commission stated that it was erroneous to suppose that the Commission was interested in such statistics merely for the purpose of rate making. The statistics would be valuable in making year-to-year comparisons for the same railroad and for different roads in the same year. The Commission also said:

1/ In the Matter of the Separation of Operating Expenses, 30 I.C.C. 676. 2/ Among cases in which cost evidence was introduced were: Five Per Cent case, 31 I.C.C. 392; Anthracite Coal case, 35 I.C.c. 220; 1915 Western Rate Advance case, 35 I.C.C. 497; Western Passenger Fares case, 37 I.C.C. 1; Lake Erie Ports Iron Ore case, 41 1.C.C. 181; and New England Milk case, 40 I.C.C. 699.

1.

-5

It is the duty of the Commission to keep itself informed regarding the conduct of the railway business, that a knowledge of the variations in unit costs is valuable even if no rates are based thereon.

2. Approximately two-thirds of the operating expenses of a
railroad can be separated in a reasonably satisfactory
manner. The separation of the remaining one-third is
useful if a basis is selected which equitably measures
the use which either service makes of common facilities.
3. "Special purpose" statistics rarely command that confi-
dence which inheres in figures that are kept continu-
ously on accepted bases without reference to a particular
controversy.

4.

5.

Comparable use units might be found for distributing the property used in common.

The systematic development of railway cost accounting
and its free discussion in the light of all information
available is the best safeguard against the misuse of
cost figures.

The Commission concluded that Circular No. 3 should be revised so that a workable plan would be developed which would not be unduly burdensome to the carriers and which would yield statistical results of a fundamental character of value alike to the carriers and the Commission, and that no operating expenses were to be left unapportioned.

The third step in the Commission's revival of this subject was the offering of a tentative plan (March 18, 1915) of separating expenses between freight and passenger services 1 based on its Circular No. 3 but modified in accordance with 30 I.C.c. 676. The Association of American Railway Accounting Officers made numerous suggestions for improving the formula without receding from the position that no uniform basis could be adopted which would apply equitably under all conditions. With the exception of those relating to maintenance of way and structures, most of the Association's suggestions were adopted. According to this plan:

1. All expenses which were directly or naturally as-
signable to one service or the other would be di-
rectly assigned to the respective services to the

1 I.C.C. Statistical Series Circular No. 4, March 18, 1915.

- 6

fullest extent practicable without undue increase
in accounting expense.

2. Expression of opinion was invited as to whether
common maintenance of way and structures expenses
should be divided as a whole without prior separa-
tion between traffic wear and weather stress and
without separation as between terminal and line
expenses (except for directly assignable items).

3. Expression of opinion was invited as to the suita-
bility of the following bases for the separation of
certain or all maintenence of way and structures ac-
counts:

The locomotive con-mile, exclusive of the
weight of tender.

The locomotive tractive-power mile, which
was similar to "a" except that tractive
power was substituted for weight.

The direct charge under which all of mainte-
nance of way and structures expenses would
be regarded as an overhead expense and that
for every dollar directly spent in running
trains over a track, a sufficient number of
cents would be added to meet maintenance of
way and structures expenses.

The gross ton-mile which had the partial
difficulty that gross-ton-miles were not
regularly compiled by all railroads.

On May 21, 1915, a hearing was held at which the matter was submitted. The Commission stated that in discussions before it there was apparent acceptance of the necessity for such a division, and that there was no serious difference of opinion as to the methods proposed except as to maintenance of way and structures expenses for facilities used in common. The representatives of state commissions advocated use of "gross-ton-miles" and those of railroads "engine-ton-miles," apparently influenced by possible effect on future statistical evidence in passenger-fare cases. Neither method was approved. The Circular No. 4 method was made effective and the carriers were required to report the disputed portion of the expenses as "undivided." Carriers also were required to compile locomotive-ton-mile

data.

On December 7, 1915, 1/ the Commission commented on various methods of separating expenses. Allocation of expenses based upon locomotive-tonmiles was rejected. The Commission said that to a certain extent this

1/ Western Passenger Fares, 37 I.C.C. 1.

- 7

method forms a measure of the wear and tear on track and structures, but a large proportion of the maintenance of way and structures expenses are influenced only to a small extent and some expenses are not influenced at all by the weight and speed of trains. Action of elements and deterioration of materials continue regardless of whether trains pass over track or not. An example was the replacement of ties. It was unknown how much replacement was due to action of elements and how much due to wear. The Commission also indicated belief that there was no doubt but that high velocity of passenger trains, compared with freight, necessitates a better maintenance standard. Allocation based upon revenue train-miles was not approved because it does not take into account relatively greater speed of passenger trains or relatively greater weight and length of freight trains. The locomotive repair and transportation direct cost basis was not approved because there was little attempt on the part of the railway industry to support it. Locomotive ton-miles, including yard engine ton-miles, was not approved for reasons indicated in the disapproval of the locomotive tonmile method. The gross ton-mile basis was not approved because it does not directly take speed into consideration as indicative of increased wear.

After rejecting all of the bases offered both by the railroads and by the protestants, the Commission divided maintenance of way and structures expenses on the basis of the actual division of seven accounts:

Road enginemen

Fuel for road locomotives

Water for road locomotives

Lubricants for road locomotives

Other supplies for road locomotives
Road trainmen

Train supplies and expenses

The Commission indicated that its decision to use this basis of allocation was not necessarily to be regarded as conclusive of the method that should ulitmately be used to divide maintenance of way and structures expenses between freight and passenger. Objections to the direct charge method were known and appreciated but the method was used because objections seemed less forceful than those waged against any of the other methods proposed. Resulting rules for the separation of expenses between freight and passenger services ("Rules Governing the Separation of Operating Expenses between Freight and Passenger Service on Large Steam Railways") were made effective as of July 1, 1915, and were used for the year ended June 30, 1916.

Part of the evidence before the Commission, (July 11, 1916, in the New England Milk case) concerned a method of dividing maintenance of way and structures expenses on the basis of fuel issues.

In the words of

1/ New England Milk case, 40 I.C.C. 699. Although the Commission here rejected this basis of apportioning maintenance of way and structures expenses, it subsequently became the basis used for 16 years. (See p. 9, following.) The method is fully described in The Separation of Railroad Operating Expenses between Freight and Passenger Services, William J. Cunningham, QUARTERLY JOURNAL OF ECONOMICS, February 1917.

one authority who was active in this case:

"The theory upon which the fuel basis is offered as a substitute for train, locomotive, and car miles or combinations thereof, is that it is a more scientific measure of use. Fuel consumption, in a large degree, is proportional to the horsepower developed. Horsepower is the resultant of locomotive tractive force and speed. Speaking in general terms, in locomotives of similar design, the horsepower developed is closely related to the volume of steam used in the cylinders, and the steam production in boilers of similar design is in turn closely related to the amount of fuel consumed in the firebox. There is, however, a critical point in the piston speed at which the horsepower efficiency is greatest.

"Since the horsepower developed is directly propor-
tional to the speed of the train, it follows that the fuel
consumption by a given train increases somewhat in propor-
tion to the speed of the train. Fuel consumption, however,
does not vary directly with speed, yet it bears a close
relation to it. It is impracticable to show the relation
exactly by formula because of the many variable factors
which affect train resistance and locomotive boiler and
cylinder efficiency; but the indications of many tests
support the statement that as between trains of different
classes on the same division, the fuel consumption, as a
measure of train horsepower, automatically equates for
both weight and speed. It takes account of the weight or
power of the locomotive, the weight or number of cars, and
the speed of the train. It affords, then, the only least
common multiple yet suggested that has some claim to the
term scientific.

"From an accounting viewpoint the fuel basis is
desirable as it requires no special compilations when
(as should be done for other purposes) fuel consumption
statistics are kept separate for the three classes of
service--freight, passenger and switch." 1/

In order to reduce statistical reporting for the period of the war, the Commission, on October 23, 1917, withdrew its requirements for apportioning common expenses between freight and passenger services and for maintaining a record of locomotive ton-miles, as prescribed by the rules

1/ The Separation of Railroad Operating Expenses between Freight and Passenger Services, William J. Cunningham, QUARTERLY JOURNAL OF ECONOMICS Vol. XXXI, February 1917, p. 236.