The 1965 accident data to which I refer (with the exception of January and February) are based on the preliminary reports of the Interstate Commerce Commission. These data are subject to upward adjustment because of late reporting and corrections. However, differences between the preliminary reports and the final reports in the past have been greater with respect to the train accident data than the casualty data. This is primarily because of the fact that final determination of whether or not a train accident is reprotable depends upon an arbitrary amount of damage to railroad property. In 1964 the difference between the preliminary and final number of reported train accidents amounted to 3.3 percent. The difference in total casualties was 2.3 percent, and the difference in casualties to employees on duty was 1.9 percent. Adjustments of this magnitude in the 1965 preliminary figures would not appreciably change the percentage relationships or trends, and would not impair the validity of the conclusion that there is a downtrend in the number and rate of casualties.

For several reasons, the casualty statistics (fatalities and injuries) to which I have referred are a more important and significant measure of changes in relative safety than train accident statistics. Many train accidents occur without any personal injuries. While, of course, we are concerned with damage to equipment and property, our first concern is with the personal safety of individuals. The reportability of a train accident is determined solely by the amount of damage to railroad property. Currently the dividing line is $750. If an accident results in damage of $750 or less it is not a reportable train accident. If the damage is more than $750 it is a reportable train accident. The increasing cost of railroad equipment and the increasing cost of repairs because of higher costs of materials and supplies and increased wages could, without regard to any other factor, result in an increase in reportable train accidents without any change in the over-all frequency of accidents, fatalities, or injuries in the railroad industry. The changing composition of railroad service, resulting from a decline in the volume of passenger service and the increase in the volume of freight and yard services has contributed to the increase in train accident frequencies over the past several years. This is explained in somewhat more detail in the attachments. The question of the impact, if any, of the removal of firemen in freight and yard service, commencing in May of 1964, on the safety of operations of the American railroads was thoroughly aired during the hearings before the Senate Committee on Interstate Commerce in the summer of 1965. It was also given considerable attention in the January 1966 report of the National Joint Board. This Board was created and functioned pursuant to the Award of Arbitration Board No. 282, which, as you know, was established by Public Law 88-108. The four-man National Joint Board was composed of one member each from the Engineers' and Firemen's Unions and two inembers from the carriers. The evidence adduced at the Senate hearings and the facts included in the Joint Board Report, which are corroborated by the recently available 1965 preliminary statistics included in the attached compilation, strongly indicate that there is no relationships between increases (or decreases) in casualty and accident frequencies and the removal of firemen in freight and yard service.

In conclusion, I wish to express my appreciation for your interest in this very important subject. It has given me an opportunity to present to you what we consider to be a comprehensive review of the casualty experience of the railroad industry. If you desire additional information, or there is anything in this presentation that requires clarification, do not hesitate to call upon me, because it is my sincere desire to make it absolutely clear that we have found, and the available data confirms, that railroad accidents and casualties are not attributable to the absence of firemen.

Very truly yours,

J. E. WOLFE, Chairman.

ANALYSIS OF CURRENT SAFETY STATISTICS

A. EFFECT OF THE CHANGES IN THE LEVELS OF COSTS IN REPORTING OF TRAIN ACCIDENTS

A factor which has contributed to the higher train accident rates reported by the Interstate Commerce Commission during the last several years is the increase in the cost of labor, materials and equipment. The Interstate Commerce Commission defines a train accident as one in which damage to property or equipment exceeds $750. In computing such damages the railroads take into consideration the cost of equipment damaged and the cost of repairs. Due to higher wage rates and increases in the cost of materials and equipment, many accidents which were

formerly unreportable, because the damage involved was less than $750, are now reportable accidents. Train accident rates reported by the ICC may thus be increased without any increase in the actual number of such accidents.

The following table (A-1) contains significant cost statistics for the Class I Line-Haul Railroads for the years 1961 through 1964.

Repairs to right-of-way and equipment are made by the maintenance of way and shop employees. From 1961 to 1964 there was a 7.8 percent increase in the average straight time hourly rates of pay of maintenance of way employees and a 6.0 percent increase in the hourly rates of pay of employees engaged in the maintenance and repair of equipment. In September of 1965 the average straight time hourly earnings of maintenance of way employees stood at $2.579, 8.0 percent above 1963 levels; and those of maintenance of equipment employees were $2.867, 7.1 percent above 1963 levels. In 1963 the average cost of freight cars purchased by the American railroads exceeded their average cost in 1961 by 23.9 percent. Equipment costs are known to have increased in 1964 and 1965, although published data is as yet unavailable for these years. The effect of these increases in costs upon the number of train accidents reported to the Interstate Commerce Commission are reflected in the fact that while the number of reported train accidents on all United States railroads increased 28.1 percent during the period 1961 to 1964, the number of casualties in train and train-service accidents increased only 4.6 percent.

Source: Interstate Commerce Commission statements M-300 and statements of railways.

B. EFFECT OF REDISTRIBUTION OF OPERATIONS AMONG THE SEVERAL CLASSES

OF SERVICE

Since 1961, as during prior years, there has been a decline in the volume of passenger service and a concomitant increase in the volume of freight and yard services. This redistribution of the operations among the several classes of services is shown on the following table (B-1). Passenger gross ton-miles declined by 9.3 percent over the period 1961 to 1964, while gross freight ton-miles increase by 12.4 percent; and net freight ton-miles increased by 16.2 percent. The effect of these changes in the redistribution of service on total accident and casualty levels may be judged from the accident and casualty rates shown in the second following table (B-2). Train accident rates in freight and yard service are from three to four times as high as in passenger service. Casualty rates in freight service are three times as high as in passenger service, and casualty rates in yard service are nearly seven times as high as in passenger service. With no change in the hazards of any of the three classes of service, levels of total accidents and casualty frequencies would have increased merely as a result of the redistribution of the service that has occurred since 1961.

TABLE B-2.-TRAIN ACCIDENT RATES AND CASUALTY RATES TO TRAIN AND ENGINE SERVICE EMPLOYEES IN TRAIN AND TRAIN-SERVICE ACCIDENTS PER MILLION LOCOMOTIVE AND MOTOR TRAIN-MILES, BY CLASSES OF SERVICE

1 Rates by class of service computed by dividing accidents on all railroads by train-miles for Class I Line-Haul Railroads. Rates for all services computed by dividing accidents on all railroads by train-miles on all railroads.

2 Not available.

8 Class I line-haul railroads, except 1964, for all services is for all railroads.

Source: Interstate Commerce Commission Annual Accident Bulletins, Statements M-400, and Statistics of Railroads.

C. CASUALTIES TO ALL CLASSES OF PERSONS AND TO EMPLOYEES ON DUTY Because of the factors discussed in Part A, casualty statistics (fatalities and injuries) are a more significant measure of changes in relative safety than train accident statistics.

The four tables in this series summarize the casualty experience of the United States railroads for the years 1961 through 1965, and for the eight months period May to December inclusive for the same years. For the years 1961 through 1964 casualties are segregated between those occurring in train and train-service and in nontrain accidents. The only statistics presently available showing 1965 casualties (except for January and February) are reported on ICC Statements M-450 entitled "Preliminary Report of Railroad Accidents and Resulting Casualties". This statement does not segregate casualties between train, train-service and nontrain accidents (except for casualties occurring to passengers on trains). Therefore, for 1965 only casualties in train, train-service, and nontrain accidents combined are available.

The first two tables show casualties to persons of all classes. This includes casualties to employees on duty, passengers on trains, all classes of persons in highway-grade crossing accidents, and other classes of persons. Casualties to

95-388-6826

persons of all classes in train and train-service accidents during the year 1963 were 3.3 percent higher than in 1962; in 1964 these casualties were only 1.6 percent higher than in 1963, although in 1964 20.1 percent of the freight and yard operations were conducted without firemen (Table C-1).

For the eight months May to December inclusive, casualties to all classes of persons in train and train-service accidents in 1963 were 5.8 percent higher than in 1962, whereas in 1964 these casualties were only 5.1 percent higher than in 1963. During the 1964 period, 31.4 percent of the freight and yard service was operated without firemen, whereas in 1963 practically all of the freight and yard locomotives were manned with firemen (Table C-2).

As heretofore indicated, 1965 data is available only for all types of accidents (train, train-service, and nontrain accidents). Casualties to all classes of persons in all types of accidents were 2.1 percent higher in 1963 than in 1962. 1964 was only 1.5 percent higher than 1963, but 1965 was 7.6 percent lower than 1964, although 47.7 percent of the freight and yard locomotives in the latter year were operated without firemen. The 1965 preliminary total of 27,751 casualties to all classes of persons is lower than the final total for any of the years 1961 through 1964 (Table C-1).

Total casualties to persons of all classes in accidents of all types for the eight months May through December for each of the years 1961 through 1965 are shown on Table C-2. The increase between 1963 and 1964 was less than the increase between 1962 and 1963, and the preliminary figure for 1965 reflects a decline of 9.3 percent, although 51.3 percent of the freight and yard service was operated without firemen.

The last two tables in this series (C-3 and C-4) are identical to the preceding tables except that they cover only casualties to employees on duty. In all substantial respects, they reflect the same reversal or standoff in the comparative increases in casualties in the years 1962 to 1963 and the years 1963 to 1964, and the same decrease in the years 1964 to 1965. For the entire year 1965 these preliminary data reflect a decrease in casualties to employees on duty under 1964 of 8.3 percent, and for the eight months May to December a decrease of 10.1 percent.

1965 preliminary casualty figures included on the tables in this series are subject to upward adjustment because of late reporting or corrections. In 1964 differences between the preliminary and the final figures amounted to 2.3 percent in total casualties and to 1.9 percent in casualties to employees on duty. Changes of a similar magnitude in the final 1965 results would not affect the conclusions which are self evident from an examination of these four tables.

It is clear that the removal of freight and yard firemen commencing in May of 1964 did not have an adverse effect on the casualty record of the United States railroads. Increases between 1963 and 1964 were less than the increases between 1962 and 1963. The available data for 1965, the first full calendar year of substantial operations without firemen, show a substantial improvement over 1964, which should not be substantially altered when the final 1965 figures are compiled by the Interstate Commerce Commission.

1961.

1962.

1963.

1964.

1965.

TABLE C-1.-CASUALTIES TO PERSONS OF ALL CLASSES, U.S. RAILROADS, YEARS 1961-65

1 Class I line-haul railroads.

2 Not available.

• Preliminary.

Source: ICC Statements M-300, M-400, and M-450.

TABLE C-2.-CASUALTIES TO PERSONS OF ALL CLASSES, U.S. RAILROADS, MAY-DECEMBER, INCLUSIVE, 1961-65

1961.

1962.

1963.

1964.

1965.

TABLE C-3.-CASUALTIES TO EMPLOYEES ON DUTY, U.S. RAILROADS, YEARS 1961-65

1 Class I line-haul railroads.

2 Not available.

3 Preliminary.

Source: ICC statements M-300, M-400, and M-450.

TABLE C-4.-CASUALTIES TO EMPLOYEES ON DUTY, U.S. RAILROADS, MAY TO DECEMBER, INCLUSIVE, 1961-65

8 months, MayDecember

Percent of freight and yard operations without

Casualties in nontrain
accidents

firemen 1

year

D. CASUALTY RATES OF TRAIN AND ENGINE SERVICE EMPLOYEES IN TRAIN OPERATIONS The quarterly casualty frequency rates tabulated on Table D-1 show no relationship between increases in rates and the elimination of firemen's jobs in freight and yard service. The increases in such rates generally were no greater in 1964 than in 1963 or in 1962. Notwithstanding rising traffic levels and the redistribution of the service (see Table B-1), the increases in the majority of the casualty rates were smaller for the year 1964 than for the preceding two years.

That there is no ascertainable relationship between casualty rates and freight and yard operations without firemen is apparent when these data are examined.