tive: although the rate level in the future may quite prop erly vary between divergent sections of the country, it is unlikely that discrimination between interstate and intrastate traffic, or among points, will have any considerable future significance with respect to passenger traffic. In short, rate-making in the passenger field may be expected t be even more simple in the future than in the past, with but two factors having weight: mileage and the rate per

References:

mile.

CLARKE, J. M. Standards of Rate Reasonableness in Local Freight Dcrimination, Columbia University Studies, Vol. 37, No. 1. CONVERSE, PAUL. "The Problem of the Transcontinental Rate Structure," Journal of Political Economy, Vol. XXVI, pp. 291-301.

INTERSTATE COMMERCE COMMISSION. Reports.

JOHNSON, E. R. and HUEBNER, G. G. Railroad Traffic and Rates, Vol. I, chaps. xxii-xxiv. D. Appleton & Co., 1911.

MCPHERSON, LOGAN. Railroad Freight Rates. H. Holt & Co., 1909. RIPLEY, W. Z. Railroads: Rates and Regulation, chaps. iv, v, xi. Longmans, Green & Co., 1912.

RIPLEY, W. Z. Railway Problems, chap. xvii. Ginn & Co., 1913. VANDERBLUE, H. B. and BURGESS, K. F. Railroads, chaps. ix-xiii. The Mac

millan Co., 1923.

WILLIAMS, H. G. AND FAGG, CHARLES J. The Freight Traffic Red Book The Traffic Publishing Company (Annual).

XXVI

COMPARATIVE RAILWAY CHARGES

The traffic unit in freight and passenger fields. Computation of unit costs. Railway costs and charges. Factors influencing costs: construction differences, efficiency, price level, traffic factors, quantity of shipment. Modifying factors: length of haul; type of traffic; character of service; purchasing power of money. Evaluation of passenger rate comparisons. Conclusion concerning the value of comparisons. Express charges. American freight rates. American and European charges. American passenger rates. Comparison with foreign charges. Conclusion.

DURING the past 100 years the steam railway has developed from a mere experiment into a transportation agency of fundamental significance, and in no country has that development been more striking than in the United States. Expanding rapidly, our railway net today comprises almost 40% of the mileage of the world. Yet, since transportation agencies are not regarded as ends in themselves but rather as instrumentalities of commerce, the comparisons which are of greatest value to us are those of service rather than of mileage. And as the rate is the only mathematical index available of the efficiency of that service, the rate becomes of immediate concern. How does the cost of transportation in the South compare with that in the North, in the East with that in the West? How do rates in the coal districts of West Virginia compare with those of Illinois, upon Kansas wheat with those upon Dakota grain? Indeed, how do rates upon the entire traffic of the United States compare with those charged upon the traffic of other countries? In arriving at conclusions, the comparison of individual rates or of the totals of published rates would avail little. On such a basis conclusions would be hard to draw and, when drawn, would be worth little because the tonnage moving upon some rates is negligible, upon

others tremendous. The only reasonable basis of comparison yet proposed is that of the average cost of movement of all the traffic handled by each railway or group of lines.

TRAFFIC UNITS IN FREIGHT AND PASSENGER SERVICE

In computing such an average, the first difficulty met has proved insuperable—it has been impossible to devise a common base for the two great types of traffic-for freight, upon which charges vary in general with quantity and distance, and passenger, upon which the individual charge varies with the distance factor alone. True enough, bases of equation have been devised, but the absence of agreement among these indicates the difficulty of securing a reliable consolidated figure. As a result of the inability to discover a common base, therefore, service as a whole cannot be compared with service as a whole-such comparisons as are made must be between similar types of service rendered by the carriers or groups of carriers under consideration.

In the selection of the unit to be used in the determination of average freight charges, any one of several might be employed. Of these the train-mile is one, but variations in train length are great from day to day and from season to season upon a single line, to say nothing of the variation between plains and mountainous territory and between countries. Therefore the train-mile as well as the car-mile, to which objection also lies because of wide variations from time to time and from place to place, are so open to criticism that they have been accepted by none. Indeed, the only unit which is sufficiently uniform to justify its utilization in the freight field is the ton-mile. By definition, the ton-mile is the movement of one ton one mile or its equivalent, as for example, 100 pounds 20 miles. Yet it may be well to note that even ton-mile compilations may not be comparable, because, in arriving at the figures, gross tonnage, net tonnage, or net revenue tonnage may have been

used. This difficulty is not experienced in the United States, however, because all American ton-mile statistics are prepared according to rules laid down by the Interstate Commerce Commission and cover, presumably, net tonnage alone, a clear differentiation between revenue and non-revenue traffic also being made. Several methods may be followed in compiling ton-mile data, but generally it is obtained by multiplying the weight of each shipment by the number of miles it has been moved, both items of information being gained from the way-bill. Occasionally, however, reliance in compiling ton-mile figures is placed upon the freight conductor's "wheel report" by the accounting office. It is in that office that monthly aggregates are made, not only of ton-miles but also of total tons of traffic handled, classified according to commodities.

In the selection of a unit to be used for comparative purposes in the passenger field, the problem is somewhat more difficult. Although the train-mile is as faulty here as it is in connection with freight movements, the car-mile is deserving of more serious consideration. To a degree which is not true of freight, the weight of the passengers occupying a car is negligible when compared with the dead weight of the heavily built vehicle in which they are moved. Therefore, when the factor of cost is regarded, it might well seem proper to give the car-mile serious consideration. However, because of the universal significance attached to traffic units and the fact that the charge so clearly pertains to the individual movement, the passenger-mile has been generally adopted as the basis of comparisons in this field. This unit, like the ton-mile, may be compiled upon different bases, but figures for American roads are prepared on the basis of what might be termed public movement, excluding movements of persons on company business. These figures, covering revenue passenger movement, are compiled by the accounting department, as are the ton-mile statistics, and usually are based upon the conductors' reports which are submitted at the end of each run.

COMPUTATION OF UNIT COSTS

Once the ton-mile figures of a given division, road, territory, or country are available, the average cost figure to be used for comparative purposes is obtained by a simple mathematical operation—the division of the total gross revenue from the freight service by the total number of tonmiles of business handled by the particular unit or area for which figures are desired. Similarly, the average cost of passenger movement is determined by dividing passengermiles into gross passenger revenue. The figures thus obtained show, for a given period, the average cost to the particular public involved of moving one ton and one passenger, respectively, a distance of one mile. Conclusions based upon a comparison of these figures are given great weight by many, and those able to show the smaller quotient herald the fact with pride. Such figures have been accepted, too, by the public as an infallible index of the superior efficiency or the inefficiency of particular carriers or of the carriers serving a particular territory. Immediately the question arises, "Are such comparisons to be accepted at their face value, or does careful analysis tend to discount obvious conclusions?"

RAILWAY COSTS AND CHARGES

Before entering upon such an analysis, it is desirable that the nature and elements of cost, as before mentioned, be clearly understood. Under conditions of perfect competition in any field, cost to the marginal producer of a good or service-here being included necessary profits and cost to the consumer are identical. However, the railway industry is not one in which competition functions in such manner as to insure this identity of costs, even in countries where the various properties are so owned as to preclude a community of interest. Because of this fact, we find that government control in one form or another is characteristic of modern railway operation, and that such control tends, with few ex