Your letter states that the studies made by members of the Commission staff since 1949 have uniformly classified motor truck transportation from farm to city, or city to farm, as "intercity" traffic. No reference was made therein to rail, water, and pipeline traffic from rural areas which is comparable to city-farm motor traffic. Rail rural stations frequently have stock pens, factories, and/or grain elevators, but no resident population. Water traffic from gravel pits, pipeline gathering movements, etc., are similar in character. The inclusion of motor traffic from farm to city and city to farm is therefore quite consistent and proper. Your letter further states that the reports of intercity ton-miles have been so prepared as to make the increase in motor traffic-largely noncompetitive, and in great part a change from horse-drawn and other unrecorded traffic-[appear] as diversion from railway traffic. According to the Bureau of Public Roads, 50 percent of the ton-miles on main rural roads arise from hauls of over 200 miles, based on the only nationwide study available. The average haul for class I motor common carriers in 1963 was 267 miles, and for class I common carriers of general freight, 342 miles. Motor hauls of this length are competitive with rail transportation. As for the "horse-drawn" traffic, the Bureau of Public Roads in the 1930's found such traffic in quantity sufficient to be recognized only at isolated locations, and was of no significance then. This traffic was and is inconsequential in relation to the total. The transfer from "horse-drawn" to motor vehicles could have had only an infinitesimal influence on the increase in intercity ton-miles between 1939 and 1964 from 52.8 to 349.8 billions, or from less than a sixth to more than half of the rail ton-miles in the respective years.

Total revenues of the class I, II and III property motor carriers were $792.2 million in 1939, compared with $4,140.3 million for railroads. Both have been estimated in the $10 billion range for 1965. Freight revenue for 1949 and 1964 for ICC regulated carriers were:

The foregoing revenues show that changes occurred in revenue distribution roughly parallel to the changes in distribution of intercity ton-miles.

While not specifically so stated, the reference by you to the possibility that motor vehicle ton-miles represent to some extent prior "horse-drawn" traffic indicates that you may be assuming that the Commission is representing the motor vehicle ton-miles as diversions from the railroad. If so, this assumption is incorrect. No statement to that effect has ever been made by Commission compilers of the data. The ton-mile data are shown as media totals, and shares are shown related to the grand totals.

It is contended in your letter that the omission since 1949 of coastal and intercoastal water transportation has made it appear that the great proportional increase in pipeline transportation has been a diversion from the railways, when it represents largely a shift from coastal tankers.

Bureau of Public Roads data on gasoline and fuel oil consumption and Bureau of Mines data on energy derived from fuel show great increases in use of petroleum products. ICC rail Freight Commodity Statistics for 1949 and 1963 (the latest published) show a reduction in tons originated by class I railroads of petroleum and products in the six petroleum and products classifications from 60.1 million tons in 1939 to 48.2 million tons in 1949 and 26.7 million in 1963. These figures indicate a very evidenciary diversion of petroleum traffic from the railroads. The tabulation below from data of the Association of Oil Pipe Lines includes both internal, coastal, and intercoastal transportation of petroleum in each year. It indicates clearly that the increase in pipeline traffic since 1949 could not have resulted from diversion of coastal tanker traffic. The increase in pipeline traffic between 1949 and 1963 was 260,125,000 tons, which was greater than the water carrier traffic in 1949. During the same period, water carrier traffic increased

105,683,000 tons. Motor carriers also had large increases. The decrease in rail traffic between 1949 and 1963 was only 3.9 percent of the increase in pipeline, water carrier, and motor carrier tons carried in domestic transportation during the same period.

Source: Association of Oil Pipe Lines, "Shifts in Petroleum Transportation," table 1, March 1965.

Your statement that the great decline in railway use of coal has reduced revenue traffic of those railways supplying others [railroads] with that fuel but has correspondingly improved the position of other carriers is probably true, since the increased use of oil and gas has provided traffic for tanker, tank truck, and pipeline but not for railroads, as indicated above. This is a concrete example of loss of traffic by the railroads.

As to the inclusion of ton-miles generated on the Great Lakes by traffic between the United States and a foreign port, railroads of the United States offer competitive rates and solicit for the traffic via land routes. As a matter of fact, they secure some of it during the period of closed navigation on the Great Lakes and St. Lawrence Seaway, the remainder which moves via rail going to the Canadian roads. While the Great Lakes may be international waters, the railroads are in direct competition for the movement in those waters in the same manner as they are competitive for traffic moving via highways along the south shore of the Lakes or domestically by water between Lake ports.

You allege that the dimensions of the error in compiling intercity ton-mile data may be in part indicated by the fact that the Commission study represents the railway proportion of intercity freight traffic to have declined from 62.4 percent in 1939 to 43.5 percent in 1964 when the actual decline has been little if any more than 1 percent. Our comments above relative to specific allegations show that the decline in the railroad proportional share of total intercity traffic has been real.

CHANGES IN REPORTS

The subject of the closing section of your letter has two aspects. You suggest that when changes in reports are made the greatest care should be taken either to make the new data continuous with that earlier compiled or to revise all earlier reports to insure as great continuity as possible. Changes in data called for by reports are made in the light of current conditions. It would require enormous clerical work, and in some cases it might be impossible because of the characteristics of the data, to revise all earlier data to insure continuity. The advantages of continuity are considered when revisions are made in Commission report forms.

The Commission staff holds preliminary discussions with representatives of the railroad industry regarding changes in reports which the industry proposes. Labor organizations and others similarly always have the same right to propose changes and discuss matters of this kind with us. The Commission does not intend to make substantial changes in prescribed reporting programs without ample notice to all parties. It is bound by the Administrative Procedure Act and the Federal Reports Act in such matters.

Revisions of the reports of railroad employee hours and compensation, Monthly Report of Employees, Service, and Compensation, Wage Forms A and B, have been carried on the Commission's agenda for review since 1959. The frequency of report filing was subject to proposed rulemaking in 1962-1963 in Docket No. 34141. Our decision to retain monthly reporting corresponded with the position advocated by your organization.

The reporting of railroad wage statistics remains subject to study and analysis as a part of the Commission's continuous reports review program, and no con

templated change is now before the Commission for consideration. Before any action is taken on future proposed changes, all interested parties will be given full opportunity to submit objections or other comment.

THE COMMISSIONERS,

RAILWAY LABOR EXECUTIVES' ASSOCIATION,
Washington, D.C., June 17, 1966.

Interstate Commerce Commission,
Washington, D.C.

GENTLEMEN: On April 1, I directed a letter to all members of the Interstate Commerce Commission submitting certain suggestions of the Railway Labor Executives' Association on transportation matters. Among the comments in that letter was a reference to rumors of proposals by carriers to revise reports on employees, hours and compensation. As I stated, knowledge of these proposals had come to the employees indirectly. We had not been advised by the carriers or by the Commission that such a project was under consideration.

Under date of June 7, I have a reply to my letter of April 1, over the signature of the Chairman of the Commission. In an Appendix to that reply, prepared apparently by members of the staff of the Commission, there is the categorical statement on its page 27 that "no contemplated change is now before the Commission for consideration." I respectfully suggest that this statement is in error. Since April 1 the Railway Labor Executives' Association has been advised by the Commissioner of Labor Statistics that proposed revisions of the M-300 are under consideration by that Bureau as well as by other government agencies. The Bureau of Labor Statistics has invited representatives of railway labor unions to confer on this subject of these revisions, and has placed the content of those proposals before our representatives. We have also incontrovertible evidence that the Bureau of the Budget directed a communication to the Interstate Commerce Commission as of the latter part of 1965 specifically requesting a study and comments of the proposed revisions to be submitted to that Bureau by September 30, 1966.

As the members of the Interstate Commerce Commission must know, the principal use of employment and compensation statistics as reported in the Commission's Statement M-300 is in collective bargaining and legislation affecting the employees represented by the Standard Railway Labor Organizations. This method of reporting was originally established in 1921 in a joint effort of the Commission and the United States Railroad Labor Board, to facilitate application of the labor provisions of the Transportation Act of 1920. The occupational classification system specifically was established by the Labor Board which was a tri-partite body; thus, railway employees had a strong voice in originating the system through which employment and compensation information has been collected and published for over forty years. When revisions were made in later years, the employees were consulted. In 1962, when the Commission entertained proposals to change the reporting system, the employees were advised of the project; they protested, and utlimately the Commission rejected the proposals. In the light of this background, it is shocking that the Commission would entertain proposals such as those outlined to us by the Commissioner of Labor Statistics without advising the representatives of Railway Labor.

Other statements in the Appendix demonstrate very clearly the need for a public hearing upon the matters covered in our original letter of April 1. We should advise you that we believe a large part of the material in the Appendix is open to serious question, and cannot be adequately handled in the process of exchanging memoranda. It was not the intention of our original letter to submit more than seemed necessary to indicate that the whole of the Commission's reporting and statistical procedures should be reviewed with the participation of all of the interested parties. We must renew that suggestion, fortified in our conviction by the Appendix to the letter of June 7, 1966.

Very truly yours,

G. E. LEIGHTY,

Chairman.

(The following material was submitted for the record:) RAILWAY LABOR EXECUTIVES' ASSOCIATION,

Hon. HARLEY O. STAGGERS,

Washington, D.C., June 4, 1968.

Chairman, Committee on Interstate and Foreign Commerce,
House of Representatives, Washington, D.C.

DEAR CHAIRMAN STAGGERS: The testimony of various witnesses on behalf of the Association of American Railroads included some references to factual elements that call for some comment, clarification and correction. Among these elements, some have already been covered adequately in the statements submitted by Messrs. Chesser, Crotty and Homer in materials already submitted in these hearings. These include statements on two technical questions:

1. The use of gross ton-miles as a measure of exposure of employees, or other human beings, to conditions causing railroad accidents;

2. The effect of prices on the number of "train accidents" reported.

Mr. H. C. Daulton's statement adds a number of technical comments that cannot be left without some analysis.

Beginning on page 3, Mr. Daulton criticizes the use of "train accidents" in any analysis of the railroad safety record. This is indeed an odd position for the carriers to adopt. For a "train accident" in simplest terms is a train wreckan incident involving collisions of cars or locomotives, or derailments of such equipment or contacts between such railroad equipment and highway vehicles. For purposes of statistical consistency, a minimum "cost of damage" standard is used to define a "train accident." This minimum, of course, is far below the actual cost of most train accidents which had an average cost per accident in 1966 of over $14,000. Now, for the carriers to contend that "train accidents" are not a measure of railroad safety is absurd. The "train accident" or train wreck is the railroad counterpart of the airplane crash; although the latter entails more casualties per incident, it is no more or less a guide to air safety than is the train wreck a guide to rail safety.

Of course, railroading, by its nature, is a hazardous business. As a result, there are thousands of circumstances occurring every year that cause deaths and injuries but do not involve train wrecks or "train accidents." Men are caught between two moving cars and are maimed; some fall beneath the wheels of moving equipment and are killed. The railroad accident reports are coded in such a way as to delineate hundreds of such individual causes of casualties. If such accidents occur in service connected with train or car movement, they are referred to as train-service accidents; if not, they are called non-train accidents. In common safety parlance, where they concern employees, they are referred to as industrial accidents. Casualties in train accidents must be added to such accidents to derive the totality of the railroad safety record.

But train accidents are the major occurrences that create heavy potential safety problems for everyone-employees, and the general public. They are a direct and historically consistent-measure of railroad safety, long accepted by everyone as a meaningful standard for this purpose.

Mr. Daulton makes reference on page 6 to some differences between the railroads and other industries in methods of reporting casualties. Such differences were largely eliminated in the changes made in the railroad reporting rules in 1961. Since then, the 24-hour disability standard has been followed in both reporting systems. Other minor procedural distinctions raised by Mr. Daulton on page 6 are of little significance. Moreover, they are consistent factors and do not invalidate historical trends.

Mr. Daulton's technical objections based on inter-industry comparisons attain the height of absurdity on page 13, where he alludes to comparisons in Mr. Homer's Memorandum of fatalities on the railroads with fatalities on the airlines and on regulated motor carriers. Mr. Daulton-after characterizing the comparisons as "improper"-states that he "must assume Mr. Homer was unaware of these reporting differences." Mr. Homer is aware of some technical difficulties inherent in the reporting differences for non-fatal injuries. However, he believes a death is a death, and is no less final and complete on the railroads than elsewhere. Expert distinctions in this facet of safety reporting are quite meaningless, and the fatality comparisons made in Mr. Homer's original memorandum are quite valid.

The Daulton statement also criticizes the Homer comparison of certain categories of train accidents based on the use of "millions of locomotive and motor

train-miles." For his information (apparently he was not aware of this fact) locomotive and motor train-miles are the standard exposure measure used for this purpose by the Interstate Commerce Commission for many years.1

It is suggested that it would somehow have been more appropriate in the specific comparisons referred to for Mr. Homer to have used gross ton-miles, or carmiles. Among the factors that have caused the increase in railroad accidents in recent years have been the longer and faster trains and the greater weight of such trains. These developments have caused greater wear to car wheels, and brakes, for instance, and greater impact on track and roadway. These circumstances create need for additional maintenance attention-perhaps for more careful inspection, perhaps for more maintenance employees. But, one defective wheel on a train can cause a train accident to the whole train, and to the crew members on that train. The number of cars and wheels increases the number of wheels that may be defective, but the exposure to casualties is not increased in the same sense. Only one crew on a train is exposed even though the likelihood of an accident is greatly intensified on longer trains. Mr. Daulton's standard for computing accident rates may be of great interest to those who are in charge of maintaining wheels, track or other items, but it does not measure the potential for train accidents or for crew casualties. In fact, as his factors increase (carmiles or gross ton-miles) the computed rate will correspondingly decline. Thus, given two accidents-one on a short train killing five men and one on a long train killing five men, the accident rate of the long train will be much lower than the rate of the short train, using his standard. But the same number of persons are killed on each. Using either number of train miles, or number of persons exposed, the rates would be the same.

The Daulton statement also criticizes the use of locomotive and motor trainmiles as an exposure factor for collisions. Here, no better alternative basis for such comparisons is suggested. Collisions can only be measured with reference to a standard which varies directly with train and locomotive movement, irrespective of train size, length or weight. Locomotive and motor train-miles alone meet this requirement. In his comment with respect to collisions, Mr. Daulton emphasizes the fact that many collisions occur in yard operations and that rates based on locomotive and motor train-miles which include all of the millions of miles covered by road locomotives introduce a degree of distortion into the comparisons made. Here, Mr. Daulton could have made a more precise computation to test his thesis as to whether the factor used in the Homer memorandum was actually a distorting element. The Interstate Commerce Commission has reported the number of switching collisions occurring in each of the years 1961 to 1966, and it is possible to estimate the number of yard switching locomotive miles from the data reported for locomotive engineers in yard service (I.C.C. statement M-300). From these figures one can derive the following rates for switching collisions alone based on yard operations:

As these figures indicate, the increase in the switching collision rate is also very striking and is comparable to that shown for total collisions on Table 21 of the Homer memorandum. Mr. Daulton's criticism, thus, has no valiidty.

Mr. Daulton criticizes Mr. Homer's use of certain I.C.C. data on motor carrier fatalities. It is conceded that the data used are not all-inclusive-that thousands of accidents occur affecting non-regulated carriers and private automobiles. However, the data used were those published by the Commission for the carriers subject to its safety controls. These data were not represented to include anything else, and no apology for their use is appropriate.

One comment is called for by a remark made by Mr. Moloney on page 22 of his statement in references to Table 20 of Mr. Homer's memorandum which shows "employee error" accidents. It should be remembered that accidents and injuries See Annual Accident Bulletins of the Interstate Commerce Commission, Nos. 130-134, 1961 to 1965.