Table 1.-Accident Rates in 30 Manufacturing Industries, 1933 and 1934-Con.

A comparison of the 1933 and 1934 data shows that in 15 manufacturing industries including such major groups as automobiles, boots and shoes, chemicals, flour, glass, leather, paper and pulp, and slaughtering and meat packing-accidents in 1934 were proportionally more frequent and more severe then in 1933, a fact which is reflected in higher frequency and severity rates. On the other hand, only 4 industries showed decreases in both frequency and severity rates, namely automobile tires and rubber goods and the three industries in the lumber group: logging, planing mills, and sawmills. In the case of automobile tires and rubber goods, the decreases were small. But in the three lumber groups, the decreases were significant. In logging, the average number of disabling accidents per million manhours worked decreased from 103.60 to 79.90, and the average days lost per thousand man-hours decreased from 21.56 to 15.94. Expressed in percentages, the accident frequency decreased by about 30 percent and the accident severity by 35 percent. For planing

mills, the frequency rate declined from 39.15 to 33.23, and the severity rate from 6.66 to 5.36. The rates for sawmills were appreciably higher than for planing mills, but also declined-in frequency from 62.03 to 55.40, and in severity from 8.38 to 7.31. Logging operations, however, easily stand out as the most hazardous and dangerous of all the 30 industries covered in this survey.

Some of the individual industry experiences warrant special mention because of the degree of change in the accident rates from 1933 to 1934. In the brick, tile, and terra cotta industry, the average number of disabling accidents per million man-hours worked increased from 37.81 to 42.76. But at the same time the average days lost, as expressed in the severity rate, were nearly halved-dropping from 6.49 to 3.75, indicating that while there were proportionately more accidents in 1934 than in 1933, they were, on the average, of a much less serious nature. In the chemical industry, a rise in the frequency rate from 12.69 to 15.81 was accompanied by a more than doubling of the severity rate, which jumped from 1.95 in 1933 to 4.02 in 1934. This experience was closely paralleled in the fertilizer industry in which an increase in the frequency rate from 41.39 to 53.82-about 30 percentcarried with it almost a doubling of the severity rate, which increased from 4.93 to 8.03. Another industry with the same type of experience is slaughtering and meat packing, in which the frequency rate rose 26 percent, and the severity rate 75 percent. Industries with appreciable but less drastic increases in rates are leather and paper and pulp.

In nine industries the frequency rates declined, but the severity rates increased, and in two industries frequency rates increased but severity rates decreased. The furniture industry averaged a decrease of more than four accidents per million man-hours, but an average. increase in the severity rate of more than half a day. In shipbuilding, accidents per million man-hours dropped from 20.31 to 16.34, but days lost rose slightly, from 2.98 to 3.16. A relatively sharp increase in the severity rate took place in the stamped and enameled ware industry, from 1.95 to 3.00, while the frequency rate dropped from 19.31 to 17.59.

Several limitations of the data require special mention: (1) The accident rates of individual industries are more reliable than those of totals of all industries combined, because these latter rates have not been weighted to reflect the relative size and exposure hazards of the component industries; (2) the injuries reported are considerably less than those which occurred, because, in addition to reportable accidents not reported to the proper State agency, the data exclude injuries which were not considered compensable and therefore either were not reported at all to the State authorities, or if reported did not enter into the tabulations of these authorities; (3) finally, the figures do not cover injuries involving lost time but occurring in

establishments excluded from coverage under workmen's compensation acts either because of exempted size or because of election. It may therefore be safely concluded that the statistics given in the table tend toward understatement, and therefore toward conservative accident rates more favorable to the industries than they should be. On the other hand, the establishments reporting were of all sizes, subject to the exceptions noted above, and were selected regardless of quality of accident experience.

Rank of Industries

THE relative rank, both as to frequency and severity rates, of the 30 industries, listed alphabetically, is shown in table 2. The accompanying chart shows not only the rank but in addition the relative size of frequency and severity rates for each of the industries for 1933 and 1934. The industry with the lowest frequency rate was ranked first, the next lowest second, etc.

Table 2.-Industry Rank, by Accident Frequency and Severity Rates, 1933 and 1934 1

12 industries tying for the same rank were assigned the same rank number, but the next number was omitted to avoid distortion of subsequent rank numbers.

Analysis of the chart reveals some interesting facts. For one, the relation to each other of the frequency rates of individual industries did not change much from 1933 to 1934. Seven industries maintained exactly the same rank in both years, and 15 others did not change by more than 2 grades in rank. Twenty-two of the industries, then,

indicated a fairly steady relationship to each other in terms of the relative frequencies with which accidents occurred, in spite of changes in both the frequency rates and exposures. Of the seven industries which did not change rank, all except the machine-tool group belonged to the upper half of the array-i. e., had rankings above sixteenth. In fact, the flour, feed, and other grain-mill products industry stayed twenty-fifth, and fertilizer industry twenty-eighth, sawmills twenty-ninth, and logging thirtieth-indicating the unvaryingly high accident hazards in each of these groups. These hazards are still further emphasized by the relatively high severity rates in 1934 in each of these groups-5.92, 8.03, 7.31, and 15.94 respectively. One of the outstanding facts to be read from the chart and table 2, is not only that slaughtering and meat-packing plants, flour mills, fertilizer plants, planing mills, sawmills, and logging establishments had high proportions of accidents in relation to exposure, but that these accidents were also of a very serious nature-a fact clearly revealed in the high severity rates.

On the other hand, such industries as carpets and rugs; boots and shoes; electrical machinery, apparatus, and supplies; cotton goods; pottery; and automobiles had low frequency and severity rates in both years. It is to be noted, however, that the rates in 1934 in this less hazardous group generally are above those for 1933.

In both years petroleum refining stands out as having a relatively low accident frequency, but a relatively high accident severity. Ranking sixth and fourth in frequency in 1933 and 1934, the industry was twenty-third and twenty-second, respectively, in severity, with rates of 3.79 and 3.97.

A similar position was occupied by the chemical industry in 1934. With a frequency rate increase from 12.69 to 15.81, or almost 25 percent, over the preceding year, the severity rate more than doubled (from 1.95 to 4.02), changing the industry's severity rank from thirteenth to twenty-third. The man-hour exposure which accompanied this unfortunate experience increased slightly over 18 percent.

Effects of Changes in Exposure

IN TABLE 3 are given the increases or decreases in frequency and severity rates, which accompanied upward or downward changes in exposure. The table clearly demonstrates that changes in exposure do not result ipso facto in uniform changes in accident rates. Increasing exposures resulted in increases in both frequency and severity in 11 industries which included, among others, such basic and large industries as automobiles, chemicals, iron and steel, and slaughtering and meat packing. On the other hand, the same results followed in four industries with decreasing exposures, including boots and shoes, leather, and paper and pulp. In five industries decreased frequencies and increased severities accompanied increased exposures. In four