THE LIBRARY OF THE NEW YORK STATE SCHOOL OF INDUSTRIAL AND LABOR RELATIONS AT CORNELL UNIVERSITY h Cornell University J Library The original of tiiis book is in tine Cornell University Library. There are no known copyright restrictions in the United States on the use of the text. http://www.archive.org/details/cu31924001366479 National Industrial Conference Board IS BEACON STREET, BOSTON, MASS. THE National Industrial Conference Board is a co-operative body composed of representatives of national and state in- dustrial associations, and closely allied engineering societies of a national character, and is organized to provide a clearing house of information, a forum for constructive discussion, and ma- chinery for co-operative action on matters that vitally affect the industrial development of the nation. Frederick P. Fish .... Chairman Frederic C. Hood .... Treasurer Magnus W. Alexander . . Managing Director MEMBERSHIP American Cotton Manufacturers' Association American Hardware Manufacturers' Association American Paper and Pulp Association American Society of Mechanical Engineers Electrical Manufacturers' Club Manufacturing Chemists' Association of the U. S. National Association of Cotton Manufacturers National Association of Manufacturers National Association of Wool Manufacturers National Automobile Chamber of Commerce National Boot and Shoe Manufacturers' Association National Council for Industrial Defense National Electric Light Association National Erectors' Association National Founders' Association National Metal Trades Association Rubber Association of America, Inc. Silk Association of America The Railway Car Manufacturers' Association United Typothet.*: of America ASSOCIATE MEMBERSHIP Associated Industries of Massachusetts Associated Manufacturers and Merchants of New York State Illinois Manufacturers' Association HOURS OF WORK AS RELATED TO OUTPUT AND HEALTH OF WORKERS Metal Manufacturing Industries Research Report Number i8 July, 1919 Copyright 1919 PROPERTY CF LI3RARY NEW YOBK STATE SfilinOL IWOOSTRIAL m LABOR BELATIONS CORNELL UNIVERSITY National Industrial Conference Board 15 Beacon Street Boston, Mass. LIST OF TABLES PAGE Table 1: Geographical Distribution of Establishments and Employees Covered by the Investigation, 1917 and 1919 . . 3 Table 2: Establishments and Employees Covered by Replies to 1919 Questionnaire, Grouped According to Nominal Hours of Work per Week 6 Table 3: Summary of Changes in Output Accompanying Re- ductions in Hours of Work, by Hour Groups — 1917 Ques- tionnaire 8 Table 4: Summary of Changes in Output Accompanying Re- ductions in Hours of Work, by Hour Groups — 1919 Ques- tionnaire 12 Table 5: Percentage of Deaths of Males from Various Causes, for Certain of the Metal Trades and for Workers in all Manufacturing and Mechanical Pursuits, by Specified Age Groups, U. S. Census, 1909 32 Table 6: Percentage of Deaths of Males Due to Tuberculosis, of the Total Mortality for Certain Occupations and for All Manufacturing Pursuits, by Age Groups, U. S. Census, 1909 . 33 Table 7: Mortality from Tuberculosis of the Lungs in Oc- cupations Exposing Workmen to Metallic Dust — Males, 1907 to 1912 34 Table 8: Mortality from Pulmonary Tuberculosis among Tool and Instrument Makers, Industrial Experience of Prudential Insurance Company, 1897 to 1914, compared with that of All Males in United States Registration Area, 1900 to 1913, by Age Groups 35 Table 9: Distribution of Causes of Deaths among Members of Brass Workers' Union of North America, 1909 to 1914 ... 36 Table 10: Distribution of Total Time for- which Sick Benefits were paid by the International Molders' Union, 1913 to 1916 36 Appendix Tables 42 Foreword THIS report is the fifth of a series issued by the National Industrial Conference Board as the result of an investigation into the relationship between different hours-of-work schedules, efficiency of production and health of workers. The basis of this study is the experi- ence of employers in the principal industries of the country. In the reports previously issued — those on the cotton manufacturing, wool manufacturing, silk manufacturing and the boot and shoe industries — the information collected by questionnaire was supplemented by extensive field investigation and study of statistical records. In the case of the metal trades, for reasons set forth in this report, it was found impracticable to assemble extensive statistical data. It should be understood, therefore, that the present report is primarily based on the collective judgment of manufacturers as contained in replies to the questionnaire. These replies were checked by extensive correspondence, and by some field investigation, and the Board is satisfied that the conclusions drawn are trustworthy. Conclusions as to the proper length of the workday must take into account not only the effect on output and on health, but also the need of the worker for a reasonable period of leisure for recreation, home life, and self-development. The problem varies so widely in different industries — because of differences in type of worker, in character of product, in management, and particularly in the extent to which automatic machine processes influence results — - that the experience of any one industry must not be taken as representative of industry in general. Hours of Work as Related to Output and Health of Workers METAL MANUFACTURING INDUSTRIES INTRODUCTION This study of the hours-of-work problem in the metal manufacturing industries was conducted with the same general purpose as the Board's earlier studies of the cotton,^ boot and shoe,* wool,' and silk* manufacturing industries, viz., to assemble the results of actual experience with reductions in hours of work as reflected in output and in the health of workers. The present report includes the metal trades proper, foundries, automobile, hardware, electrical equipment, and some miscellaneous establishments. Because of the large number of metal trades establish- ments which have in recent years made substantial reductions in work schedules, interest now centers in experience with a 50-hour and a 48-hour week. Analysis of the evidence submitted by establishments working on these schedules indicates that a universal reduction to a 50-hour week in the metal manufacturing industries would involve a loss in output; nevertheless, the number of establishments maintaining production with such a schedule is sufficiently large to indicate that a 50-hour week could be rather generally introduced without seriously curtailing production. In the case of a 48-hour week a smaller percentage of establishments reported production as maintained, and several of these, because of unusual size, exacting nature of the work, or other reasons, were not representa- • Research Report No. 4, "Hours of Work as Related to Output and Health of Workers — Cotton Manufacturing." March, 1918. • Research Report No. 7, "Hours of Work as Related to Output and Health of Workers — Boot and Shoe Industry." June, 1918. ' Research Report No. 12, "Hours of Work as Related to Output and Health of Workers — Wool Manufacturing." December, 1918. • Research Report No. 16, "Hours of Work as Related to Output and Health of Workers — Silk Manufacturing." March, 1919. 1 tive organizations. Notwithstanding the fact that under certain conditions reductions to such a week have proved satisfactory from a production standpoint, the weight of the evidence leaves little ground to doubt that a general reduction to a 48-hour week at the present time would mean a serious loss in production. Ability to maintain production when work-hours are reduced to 50 or 48 per week is, largely influenced by the amount of handwork. Where production is practically dependent on highly automatic machinery such schedules cannot be expected to maintain maximum output. Output and health are, of course, not the only tests of the justification of any hours-of-work schedule. The broad social aspects of the problem as to the proper work schedule are of great importance. The social factors may, however, be more intelligently considered in the light of reliable information concerning the relation of health and output to hours of work. The purpose of this and the previous reports of the series is to determine the facts on these fundamental points, leaving the social features for later discussion. STATISTICAL BASIS OF THIS REPORT In the course of the investigation two questionnaires were sent out, one in the latter part of 1917, the second in March, 1919, primarily intended to secure data regarding changes in hours made during the intervening period. Replies were received from 1,252 establishments employing 753,561 workers. Of these, 413 establishments,' employing 358,336 workers had reduced hours, and furnished data as to the effects of such reductions.^ A classification, by states, of establishments furnishing definite replies as to the effects of reductions in hours follows : 'Throughout this report the word "establishment" is used to designate a corporation, partnership, or individual submitting a schedule. When a single management operating several plants returned a consolidated schedule, this was necessarily treated as one establishment. Owing to the fact that during the interval between the time of filling out the first and second questionnaires some establishments replying to the first inquiry had again reduced hours, there is some duplication in the number of establishments in these figures; this is indicated in connection with the summary comparisons given in the appendix of this report. Obviously, however, the experience of one establishment with two successive reductions in hours of work is quite as valuable as that of two separate establishments with a single reduction. ' This total includes factory employees only. 2 TABLE l: GEOGRAPHICAL DISTRIBUTION OF ESTABLISH- MENTS AND EMPLOYEES COVERED BY THE INVESTIGATION, I917 AND I9I9 (National Industrial Conference Board) 1917 Inquiry 1919 Inquiry State Establish- ments Employees Establish- ments Employees Number Per cent Number Per cent Total Z81 266,776 100.0 1S2 101,661 100.0 California 6 1,499 0.6 1 117 0.1 Connecticut 34 16,432 6.4 13 19,276 19.0 Illinois ... 25 28,880 11.2 13 9,614 9.5 Indiana . . 9 5,497 2.1 11 5,486 5.4 Maine ... 3 1,669 0.6 Massachusetts 28 23,428 9.1 15 5,821 5.7 Michigan . . 29 79,267 30.8 9 9,531 9.4 Missouri . . 4 1,252 0.5 2 737 0.7 New Jersey . 20 9,597 3.8 7 3,076 3.0 New York . . . . 34 15,528 6.1 17 6,145 6.1 Ohio 42 20,210 7.9 16 7,443 7.3 Pennsylvania 26 39,041 15.2 10 18,789 18.5 Rhode Island 3 6,379 2.5 4 7,025 6.9 West Virginia 2 536 0.2 Wisconsin . 6 5,085 2.0 9 6,754 6.7 All others . . . . 10 2,475 1.0 5 1,747 1.7 The geographical distribution of the establishments covered by this investigation is fairly representative of the several metal working regions of the United States. The results of reductions in hours here presented are based chiefly on the judgment of manufacturers as expressed in their replies to questionnaires. These judgments, which were in many cases verified by extensive correspondence or by field work, were in turn based on plant records or personal observation of manufacturers, though such information was not always sufficiently definite to permit precise measurements. Owing to the great variety of work in the metal working industry, the frequent changes in raw materials and in other factors which - introduce important differ- ences in tasks ordinarily regarded as substantially the same, it was fouiid impracticable to secure detailed statistical comparisons of output analagous to those presented in previous reports in this series. In many cases such statistical results for the metal trades would be applicable only to small groups of workers. In view of the large number of establishments covered by the metal trades study, however, and the definite 3 character of many of the statements made, showing^gn their face that they were the careful expression of mature judgment, sound conclusions may be drawn from the collective experience of manufacturers thus secured. This can be done with the more confidence in view of the fact that extensive field studies in the other industries thus far investigated substantially corroborated the evidence obtained by the questionnaires. Moreover, this was true of the limited amount of field work done in the metal trades. Indeed, where conditions present such a wide range of variation as often occurs in the metal trades, careful judgment of a large number of manufac- turers as to the effect of changes in hours-of-work schedules on output may be a safer criterion than statistical com- parisons for particular occupations or for particular groups of workers. General Features of Metal Manufacturing Machine Time Feature of the Work. Metal working, on the whole, unlike the textile industries previously reported on, cannot be classed as dominated by either machine or hand work. In many metal manufacturing establishments skilled handwork controls production; in others a large part of the work consists in tending automatic machines in consequence of which the produc- tion rate of the plant is governed largely by machine operations rather than by the efforts of the operatives. In general, however, a large amount of skilled handwork, either with or without machines, is involved in the several branches of the industry. Wages. According to data thus far assembled in a study by the Board of changes in wages in wartime, average hourly earnings of both male and female workers in the metal trades were higher in 1914 than were those in any other of eight major industries included in that investigation.' Low Percentage of Women and Minors. The number of women and children employed in the industry is distinctly lower than in the textile industries, and lower than the average for manufacturing in general.* Com- ' These were: Metal Trades, Cotton, Wool, Silk, Boots and Shoes, Paper, Chemicals, Rubber. 'The U. S. Census of Manufactures of 1914 showed percentages of women and minors under sixteen years of age in certain occupations as follows: cotton manufacturing, women 38.2, minors 8.4; wool manufacturing, women 41.5, minorii 3.6; silk manufacturing, women 54.3, minors 7.2. pilations based on data for 281 establishments included in this investigation showed the following distribution: Total ... ... 256,775 100.0 % Men .... 239,341 93.2 Women. . 16,325 6.4 Boys ... 844 0.3 Girls 265 0.1 Size of Establishments. No other industry is character- ized by so wide a range in size of plants. The largest establishment covered by the report employed more than thirty-four thousand workers, while the smallest employed less than fifty. The capital invested in metal manufacturing industries is likewise subject to great variations. These variations are sometimes a result of the great variety in products, but more frequently they are the reflection of different degrees of develop- ment of particular enterprises. The machinery and general equipment of necessity range from the simplest to the most complicated. All of these considerations obviously have a bearing on the hours-of-work problem. PRESENT HOURS IN THE METAL MANUFACTURING INDUSTRIES With respect to present hours of work, a large propor- tion of metal manufacturing establishments fall in the 48, 50, 54, and 55 hour groups. Of the 279 establishments replying to the second ques- tionnaire, over one-third were on schedules of 48 or 50 hours per week. About 35 per cent were working on schedules of 54 or 55 hours per week. Less than 10 per cent were on schedules above 55 hours per week. Nearly two-thirds of the workers in establishments replying to the second questionnaire were employed in plants' operating on schedules of 50 hoprs per week or less. For the industry as a whole this ratio is doubtless too high, since many establishments, not having reduced hours within recent years, did not reply to' the ques- tionnaire. It is clear, however, that a considerable pro- portion of the industry is working on schedules close to 5 50 hours per week.' A detailed summary is given in Table 2: TABLE 2: ESTABLISHMENTS AND EMPLOYEES COVERED BY REPLIES TO I9I9 QUESTIONNAIRE, GROUPED ACCORDING TO NOMINAL HOURS OF WORK PER WEEK (National Industrial Conference Board) Hours per Week Establishments Employees Number Per cent Number Per cent Total 279 100.0 147,795 100.0 44 44K 8 2 18 1 3 45 2 7 52 2 4 10 2 53 45 3 2 4 1 9 1 5 2.9 0.7 6.5 0.4 1.1 16.1 0.7 2.5 18.6 0.7 1.4 3.6 0.7 19.0 16.1 1.1 0.7 1.4 0.4 3.2 0.4 1.8 6,968 2,842 20,239 95 1,009 37,729 454 1,395 27,537 950 11,273 3,059 275 10,989 16,389 1,007 913 1,196 65 2,090 20 1,301 4.7 1.9 45 13.7 47 0.1 47K 48 4:9b 0.7 25.5 0.3 49K 1.0 50 51 f 18.6 0.6 52 7.6 52>^ 2 1 63 0.2 64 7.4 S5 11 1 56i 0.7 57e 57K 0.6 0.8 58 69 0.1 1 4 &9}i 60 0.0 9 a In the case of some establishments running less than 48 hours per week, the reduction was intended to be temporary; a few establishments which had for extraordinary reasons reduced to less than 44 hours per week are omitted. b One of these establishments was on a 485i hour schedule. c One of these establishments was on a 51^ hour schedule. d One of these establishments was on a 56J^ hour schedule. e One of these estabhshments was on a 57^ hour schedule. ' The National Metal Trades Association under date of May 24, 1919, stated the weekly work-schedules of 498 establishments replying to a questionnaire as of May 15, 1919, as follows: 50 hours 89 18% 51 hours 10 2% 32 hours 63 11% 53 hoiu:s 12 2% 54 hours 105 21% 55 hours 149 30% 56 hours 3 7/10% 57 hours 1 2/10% 58 hours 2 4/10% 69 hours 23 4 7/10% ■rrr .. . . *47 90% Workmg less than 50 hours per week 61 10% Total number replying 498 100% 6 Saturday Half-Holiday The Saturday half-holiday is quite generally observed in the metal manufacturing industries. Thus, of 600 establishments reporting on this point in the 1917 ques- tionnaire, 455 did not work Saturday afternoon. Like- wise, of 265 reporting in the 1919 questionnaire, 175 closed at about noon on Saturday, while 24 others worked only five days per week. A considerable number of the establishments which reduced hours effected such reductions by inaugurating a Saturday half-holiday. While a majority failed to maintain production, there is little evidence that such block reductions, as distinguished from a distribution of the reduction throughout the week, had any signi- ficant influence on output. Of 65 such establishments replying to the 1917 questionnaire, 26 maintained, two increased, and 39 lost production, while of 23 such establishments replying to the 1919 questionnaire, three maintained, one increased, and 19 lost output. Percentage of Open Shop Establishments The replies to the second questionnaire indicate that approximately 93 per cent of these establishments were open shops. Following is a summary for 268 establish- ments reporting on this point: Establish- ments Per cent Employees Per cent Totals 268 248 12 8 100.0 92.5 4.5 3.0 117,984 115,055 1,812 1,117 100.0 Open shop 97.5 C osed union 1.5 Closed non-union 1.0 As the term "open shop" is sometimes loosely used, it is possible that in this summary there are included as open shops a few establishments which are in reality closed non-union shops. There can be little doubt, however, that a great majority of the metal trades establishments of the country are at present conducted on the open shop basis. HOURS AND OUTPUT As stated on page 2 the data for this report as to the effect of reductions in work-hours on output were chiefly obtained through two questionnaires, one sent out the latter part of 1917 in the midst of the war, the other in March, 1919, in the midst of the readjustment period. Since conditions in the two periods differed to an extent which might affect the ef&ciency of production, it was deemed advisable to present the results of the two studies separately before drawing conclusions. Results of 1917 Inquiry Altogether 281 establishments which had reduced hours submitted the results of this reduction on output in reply to the 1917 schedule.^ These establishments had 256,775 employees in 1917. There is therefore a broad basis of experience for purposes of analysis. A summary follows: TABLE 3 : SUMMARY OF CHANGES IN OUTPUT ACCOMPANY- ING REDUCTIONS IN HOURS OF WORK, BY HOUR GROUPS — I917 QUESTIONNAIRE (National Industrial Conference Board) Production 1 Production Maintained or Increased 1 Decreased EBtabllsh- Previous EatabUsh- Hours ments Employees Hours ments Employees Reduced to 45 Hours 50 1 278 1 69 60 2 1 1,543 2,609 60 1 80 Reduced to 48 Hours 55 2a S 4 644 34,9S1 4,493 55 6 3,296 54 54 10 3,702 53 1 325 53 1 13 52^ 1 385 62J^ 2 1,021 50 3 1,777 50 49M 2 1 944 149 1 148 Reduced to 49i Hours 55 52 1 1 488 186 55 54 3 1 1,328 100 a One establishment with 44 workers reduced to a 47i-hour week. Italics denote increases, ' Six establishments which had reduced hours below 44 per week, evidently because of some extraordinary conditions, are excluded from this compilation. In all but one of these cases output was reduced. 8 TABLE 3 : SUMMARY OF CHANGES IN OUTPUT ACCOMPANY- fc ING REDUCTIONS IN HOURS OF WORK, BY HOUR GROUPS J — 1917 QUESTIONNAIRE — Continued (National Industrial Conference Board) Production Maintained or Increased Production Decreased Previous Hours EstabUsb- ments Elmployees Previous Hours Establish- ments Employees Reduced to 50 Hours 60 59 57 56M 55 54 53 52J^ 52 1 2 1 1 12 7 1 2 1 320 1,342 10,030 172 S58 14,433 3,166 2,802 112 3,832 306 60 57H 55 54 53^ 53 52H 52 51M 2 1 13 13 1 1 1 1 1 66 32 8,998 8,210 569 196 1,868 13,075 12 Reduced to 61 or 61\ 55J^ 54 1 2 126 974 54 3 Hours 1,453 Reduced to 6S or 52| 57 561^ 55 54 S3 1 1 S 11 2 1 20 36 1,970 5,850 24,534 1,032 60 58 57J^ 55 54 1 1 1 11 2 140 225 Hours 87 5,350 4,132 Reduced to 53 or SS\ 59 57H 54 1 2 1 1 71 403 500 210 57H 1 Hours 153 Reduced to 54 Hours 60 59 56 • 55 13 1 2 1 17,410 8S5 237 232 4,306 61 60 59 58J^ 58 57 56 55 1 21 2 1 3 2 2 1 76 6,222 1,107 361 2,310 943 1,871 1,307 Reduced to 55 Hours 61 60 59 58 57M 1 8 2 8 4 1 3 363 3,234 10,594 2,666 1,859 130 356 60 59 58 57Ji 12 15 6 2 10,070 12,210 560 830 Reduced to 56\ Hours 60 1 231 Reduced to 57 Hours 60 1 100 Reduced to 68 Hours 60 1 118 Reduced to 69 Hours 60 1 500 60 1 250 Italics denote increases. SUMMARY FOR PRINCIPAL HOUR GROUPS Establishments Employees Increased Main- tained Decreased Increased Main- tained Decreased Reduced to 48 hours Reduced to 50 hours Reduced to 52-52^ hours Reduced to 54 hours Reduced to 55 hours 5 4 4 2 3 13 28 15 17 24 23 34 16 33 35 37,499 3,424 2,006 825 10,724 9,167 33,349 31,436 22,185 8,478 9,238 33,026 9,934 14,197 23,670 The summary" for the principal hour groups shows that of 41 establishments which reduced to a 48-hour week, five reported an increase in production, 13 that production was maintained, while 23 reported a loss. The group of establishments reporting an increase in production had much more than two-thirds of the total number of workers in the 48-hour group. One of the establishments increasing production was exceptionally large, which gives a disproportionate weight to the "increased" group from the standpoint of numbers employed. In this large establishment, moreover, various changes were made at the time the 48-hour week was adopted and the increase in output cannot be attributed to the reduction in hours alone.^ The management, however, expressed its conviction that the shorter day was conducive to better output. It should be noted that 16 establishments, none of which were of exceptional size, reported either an increased or a maintained output under such a schedule. Furthermore, in four establish- ments reporting a decrease in output the loss was com- paratively small. Of 66 establishments which reduced to a 50-hour week, four reported production as increased, 28 as main- tained, and 34 as reduced. The groups maintaining or increasing production represented approximately half the number of establishments in this group, and these two sub-groups together had rather more than half the total number of workers included in the 50-hour group. Moreover, in three of the establishments reporting production as decreased, pieceworkers maintained their ■ output, while in five others the loss was comparatively small. ' See p. 30. 10 For the 52-hour and 52J/^-hour groups experience was more evenly divided — four establishments reporting production as increased, 15 that it was maintained and 16 that it was reduced. The group maintaining produc- tion had a large majority of the total number of workers included in these hour groups. In the case of the 54-hour group, the number of estab- lishments maintaining or increasing production was only a little over half as large as that of establishments reporting a decrease. These two sub-groups, however, together had much more than a majority of the total number of workers in the 54-hour group. Of 62 establish- ments in the 55-hour group, three increased production and 24 maintained; these two sub-groups had over 40% of the total number of workers in the 55-hour group. The relatively large number of establishments which failed to maintain production on such schedules appears surprising in view of the substantial number of concerns which reported production as increased or maintained on much shorter hours. This matter is referred to later. (See page 28.) Results for the other hour groups presented in the table included too small a number of establishments to warrant extended analysis. Results of the 1919 Inquiry Replies to the questionnaire sent out in 1919 were filed by 132 establishments which had reduced hours following reductions reported in their replies to the 1917 questionnaire, or by establishments which were not included in that inquiry. These establishments together had 101,561 workers and, in general, represented the same branches of the industry as those replying to the earlier questionnaire, except that only a few auto- mobile plants are included. A summary of output results similar to that already given for the 1917 inquiry is given in Table 4: 11 TABLE 4: SUMMARY OF CHANGES IN OUTPUT ACCOM- PANYING REDUCTIONS IN HOURS OF WORK, BY HOUR GROUPS, I9I9 QUESTIONNAIRE (National Industrial Conference Board) Production Maintained or Increased Production Decreased Previous Hours Establish- ments Employees Previous Hours Establish- ments Employees Reduced to jU- or m 59 55 . 52J^ 48 2 3 3 1 691 4,468 Hours 702 538 Reduced to 1^ Hours 60 54 50 2 1 6 1 2 446 161 15,242 209 1,320 Reduced to i7 or 47i 521^ 1 378 52}^ 2 678 Hours Reduced to 48 Hours 55 54 52J^ 52 .50 1 2 1 1 1 1,985 232 360 3M1 415 60 59 56 55 54 52J^ 51 50 2 11 1,225 42 252 10,731 8,326 170 6,510 988 1,119 Reduced to 49i Hours 56 54H 1 1 430 185 55 52}^ 52 210 60 300 Reduced to 50 Hours 60 59 57 55 52^ 1 1 1 / 7 1 60 901 19 669 2,005 250 60 59 55 54 52}i 14 78 2,966 10,404 1,102 100 Reduced to 61 Hours 54 1 800 Reduced to 62 or 6Si 55 54 3 1 1,680 537 55 54 152 Hours 10,400 Reduced to 63 Hours 58 32 Reduced to 64 Hours 59 58 2 1 2,060 260 60 59 58 57 55 295 26 1,094 105 223 Reduced to 66 Hours 60 2 159 60 59 15 876 2,614 60 1 65 Italics denote increases. 12 SUMMARY FOR PRINCIPAL HOUR GROUPS Establishments Employees Increased Main- tained Decreased Increased Main- tained Decreased Reduced to 48 hours Reduced to 50 hours Reduced to 52-52i hours Reduced to 54 hours Reduced to 55 hours 1 2 5 10 4 3 2 34 25 2 6 6 3,241 1,570 2,992 2,334 2,217 2,320 159 29,363 14,650 10,552 1,743 3,505 The results of the 1919 inquiry are considerably less favorable to the practicability of a 48-hour or a 5Q-hour week than are those obtained as a result of the 1917 inquiry. So far as the 48-hour group is concerned, the number of establishments maintaining or increasing production on such a schedule was small, likewise the total number of workers included in them. In the case of the 50-hour schedule, the number of establishments maintaining or increasing production was 12, as against 25 reporting a decrease, while they had only about one-fourth the number of workers included in the 50- hour group. The basis of experience is therefore much more limited than that covered by Table 3. In this connection it should be pointed out that practically all of the experience obtained through the 1919 question- naire occurred under highly abnormal conditions. During the greater part of 1918, when approximately one-third of the reductions covered by this questionnaire took place, wages were exceptionally high and the demand for labor was extraordinary. On the other hand, after the signing of the armistice, business in many branches of the metal trades was more or less demoralized and frequently there was a shortage of orders, so that workers had less incentive to maintain maximum efficiency, while the mere reaction from the extraordinary stress of war in itself often had a tendency in the same direction. For all these reasons, the evidence gathered by the 1917 ques- tionnaire is a safer Icriterion of the practicability of a shorter hours-of-work schedule than is that obtained through the 1919 questionnaire. A few establishments included in this summary were operating on a two-shift system, but the results with respect to the efficiency of a given hours-of-work schedule were not significantly different from those for plants operating on a single shift basis. 13 Before drawing final conclusions, attention should "briefly be called to a few special factors in the problem and to further evidence gathered by correspondence and field inquiry. FACTORS BEARING ON EFFICIENCY Size of Establishment In general, the size of the establishment did not appear to exert a controlling influence on changes in production when hours of work were shortened. An examination of the appendix tables shows a wide range in the number of employees in the group of establishments maintaining production as well as in the group reporting a decrease, and this holds true of each of the principal hourly sched- ules for which replies were received. The same variety of experience was found in the case of several exceptionally large establishments, each em- ploying over 10,000 workers. One of these reported an increase in production in reducing hours from 54 to 48 per week. Another reported an increase in produc- tion when hours were reduced from 59 to 55, one that output was maintained when hours were reduced from 57 to 50, and another that it was maintained when hours were reduced from 54 to 52. The other three reported a loss in output, but in one of these (which ' reduced from 52 to 50 hours per week) the pieceworkers maintained their production, and the decrease in the output of dayworkers was not in proportion to the re- duction in hours. In another of these three estab- lishments, moreover, the change in hours was from 55 to 45 per week, an unusually large reduction. In the third large establishment reporting a decrease in output the reduction in hours was from 54 to 52 per week. While for this group of exceptionally large establish- ments experience with the shorter week was perhaps somewhat more favorable than the average experience for all establishments included in the appendix tables, the difference was not marked and cannot safely be attributed to the factor of size. Comparing the experience of establishments with from say 200 to 1,000 workers each there is little evidence to show that the size of the establishment was a controlling influence. 14 Influence of Management Undoubtedly a factor in the results in some cases was the improvement in methods of management which frequently are introduced when hours of work are short- ened. Obviously, if any major change in management or in machinery is introduced concurrently with reductions in hours of work it becomes exceedingly difficult, if not impossible, to determine how far any accompanying changes in production were due to the change in hours alone. A special effort was made to guard against this difficulty, however, as will be seen from the following excerpt from the questionnaire; What change, if any, was made at or about the time of reduction in hours: (a) In piece rates? (i) In hourly rates of time-workers.? (c) In method of wage payment? (d) In type or speed of machinery? {e) In number of machines tended by one operative? (/) In "standard" or required production? (g) In regulations affecting punctual or full-time attendance? (A) Other important changes ? What has been the effect of such reduction in hours on : (a) Total output per employee, percentage change?* (b) Labor cost per unit of product, percentage change?* (f) Amount of lost time, percent- age change?* (d) Tardiness? (/) General health of employees? (/) General attitude of employees? (g) Frequency of accidents? (A) Quality of work? (j) Other important changes? Approximately what proportion of your employees are on a piece-rate basis? If your wage system includes a bonus or premium feature, give details. •Please state whether change was an increase or a decrease. In a few cases where radical changes were made in type of machinery or in other major respects the evidence was excluded from the compilations. In most cases, however, the changes were of a character which did not discredit the evidence. 15 For Piecework For Daywork In this connection the following statement by »4he president of a steel tube company which reduced* from 50 to 45 hours per week may be cited : We introduced closer supervision and more foremen; we pushed them harder. Production has been maintained. Men can be pushed harder eight hours; after that they show the strain and lag. We would not consider the return to a ten- hour day. Influence of Automatic Machinery The amount of handwork as contrasted with automatic machine work exerts an important influence on output. In establishments where a. very high percentage of the work is performed by automatic machinery the evidence indicates that maximum production cannot, in general, be had on a schedule as low as 50 hours per week. Doubt- less a part of the differences in results reported by estab- lishments operating on the same schedule of hours is a reflection of the degree of machine domination in those plants. This is in substantial accord with results obtained in the four other industries studied by the Board. Thus in both the wool and northern cotton manufacturing industries, where machinery largely dominates production, the adoption of a 54-hour week generally resulted in decreased production. In the boot and shoe and silk manufacturing industries, on the other hand, where handwork plays a much more important part, the evidence indicated that maximum production could be maintained on a schedule substantially less than 54 hours per week, and for silk manufacturing it appeared that the point of maximum production was somewhat nearer 50 hours than 54 hours per week. In this connection, the following statement by the vice president of a large car building establishment may be cited : We have gone to 48 hours a week, and no reduction in out- put was experienced except on machines where the output is governed by the speed of the machine. The vice president of a sawing machinery company stated : We consider that the work is proportionate to the hours. We figure that the output of our shop, if we run eight hours per day, is only eight-tenths of what it is when we run ten hours a day. Much of our work is in large pieces and when put in the machine requires only slight supervision on the part 16 of the operator. He may for two or three hours absolutely do nothing, while his machine is in operation. If our work was laborious, and if it required activity and constant energy, then there is no question but what the last hours in the day would evidence the operator being tired and slower in move- ment by a decreased output per hour. Character of Work The maintenance of output in general did not appear to depend upon the character of the industry. For instance, as between automobile, hardware, and general machine shops, no significant differences in results when hours were reduced were apparent. Certain types of foundry work present exceptions. For instance, in the process of making some castings the workday cannot be shortened without interfering with production. In this connection it is to be observed that only a compara- tively small number of foundry establishments have reduced hours to 50 or 48 per week. Some foundry establishments which reduced to a 54 from a 60-hour week reported output as maintained. Some evidence was obtained in the course of the in- vestigation to the effect that in the case of work requiring unusual precision or which was for other reasons par- ticularly exacting, there was more opportunity to intro- duce the eight-hour day successfully than in ordinary metal trades establishments. Thus, the general super- intendent of a fine tool manufacturing establishment in which the work required particular skill stated: We changed from ten to nine hours at our own suggestion and later to eight, without pressure from the outside, as we believe that if a man works conscientiously for a shorter time he can do better work, and that we have a right to expect it. Our records prove the truth of this. We found that we could get the better quality for which we were looking, and that the men were in better condition and more willing to co-operate with us to get the quality. In addition we have found that our men were more loyal, even in time of strike which we went through last year, fewer of our men going out than the leaders expected; many of them were ready to come back within a few days. This establishment, however, further stated: Certain classes of work can be safely and properly con- ducted for more hours per day than others. 17 Another establishment making jigs, tools, dies, gai^s, cutters, forming tools and special machinery, all of which call for a high degree of precision in the work, stated : Our experience has been that the eight-hour day produces a better output together with many other improved condi- tions. Influence of the Piece-Rate System It was the experience of a considerable number of establishments in the metal trades, as in other industries thus far studied, that output of pieceworkers was less unfavorably affected when hours were reduced than was the output of dayworkers. In some cases pieceworkers maintained their previous output. The following sum- mary is fairly representative of the experience of a con- siderable number of concerns in this respect; other statements of similar import will be found in the appendix tables. An establishment reducing from 54 to 50 hours: Daywork output decreased about 5%. Piecework main- tained. An establishment reducing from 52 to 49J^ hours: Weekly output of dayworkers decreased 5%; that of piece- workers maintained. An establishment reducing from 55 to 48 hours : Weekly output of dayworkers decreased 20%; that of piece- workers 6%. In this connection, the following statement by the president of a large western foundry machine shop is of interest: Wherever the piece-rate system is used there is no great controversy as to the working hours. In fact, on piecework, men can produce much more in shorter hours tha:n they do under the ordinary day rate of longer hours. There also seems to be greater satisfaction among employees working under piecework as there seems to be greater incentive to work. In some cases, however, the same percentage reduction in total weekly output of pieceworkers as for dayworkers was reported. In many cases no percentage estimate was submitted. While the experience above cited is hardly typical of the metal manufacturing industries as a whole, it was reported in a substantial number of cases. In this connection, it is important to note that the proportion 18 of pieceworkers in the metal trades establishments reporting for this Investigation was on the average considerably smaller than in the case of the cotton in- dustry and the boot and shoe industry, as shown in reports already issued on those industries. Thus, for 253 metal manufacturing establishments giving the proportion of their employees paid by the piece-rate system, piece- workers represented on the average 38% of the total work force. In numerous establishments there were no pieceworkers. Influence of War Conditions In considering experience with reductions in work-hours since the signing of the armistice, it should be noted that in some instances business was slack and that establish- ments were compelled to reduce the number of employees. Naturally they were disposed to retain their most efficient workers. As a result the force working on the reduced schedule might have a higher average efficiency than the previous force working under a longer schedule. For example, a structural steel establishment which reduced from 523^ hours to 44 hours per week early in 1919 re- ported in substance as follows: A few months ago hours were reduced to 44 per week; at the same time there was a large reduction (for instance, per- haps one-third) in the number of men employees; the less desirable workmen were let go, and only the best retained. As a result of this discriminating process the average produc- tion per employee under a 44^hour week equaled the former production under a 52H-hour week. This judgment is based on tonnage production. A foundry establishment which reduced from 55 to 50 hours per week in January, 1919, also laid considerable stress on the unusual conditions prevailing at that time,- and on the factor of selection of employees. Several other establishments noted a similar selection of workers in the case of recent reductions in hours of work. Nevertheless, comparison of Table 3 and Table 4 sho^s that, on the whole, experience with a 50-hour schedule since the close of hostilities was rather less favorable than that reported in 1917. It may be that the general uncertainty of conditions during the past few months explains this; or that it was a natural reaction from the high pressure under which much work was con- ducted during the war. 19 SUPPLEMENTAL DATA As already noted, extensive statistical data covering different hours-of-work schedules were not obtainable. A brief field study for this purpose showed that the difficulties of such comparison, which have been pointed out in previous reports on other industries, were even greater in the case of the metal trades. The great variety of products in the same establishment, and frequent change in the character of the article produced or in pre- cision of work required often rendered exact measurement impossible. Another difficulty arose from the fact that in products made to different specifications, such as special machinery, the character of the work under two different hours-of-work schedules often could not be compared with any degree of accuracy. In view of these and numerous other practical difficulties in the way of securing statistical data which would really be significant, no detailed investigation of book records on a large scale was made. As an illustration of the difficulty of getting satisfactory statistical data for comparative purposes, the following statement by the general superintendent of a large electric company may be cited: We are satisfied that most of those employees who are working on piecework or bonus work are turning out as many parts in 48 hours as they formerly did in 60 hours, but since we manufacture over 25,000 different parts and our product is constantly changing, you can readily imagine that to thoroughly check this matter up would be quite difficult, and we have consequently never attempted to do so except in a few cases. The general manager of another machinery manufactur- ing corporation stated: So many conditions enter into our work to alter production, such as machine tools, jigs, special tools, that it is hard to de- termine the reasons for changes in output. However, the statements of employers as given in replies to the questionnaire or through correspondence and interviews were so specific as to give a sufficient basis for conclusions as to the result of changes in hours on production. Representative statements made by em- ployers with respect to their experiences are given below. 20 A large metal trades establishment manufacturing a variety of products, after making special effort to secure comparable data, made the following statement, in sub- stance: In all 387 lots of parts were considered, these being taken from five different departments of the shop, representing widely different classes of work, as, for example, hand oper- ated machines, assembling and fitting, hand scraping, and general machine work. In all of these cases the work was performed by the same operator before and after the change in time. Also, in some of the cases, the same operator had carried through more than one lot of parts either before or after the change, and the time of these lots was averaged so as to minimize irregularities. The reduction in hours was 9.1%. The foremen were asked in selecting these cases to be impartial and to select cases regardless of whether they showed greater or less production, unless there was some change of design, or for some special reason the results would be abnormal. AH these parts selected or operations analyzed were those which had been done on job or pieccfwork. Based on the 387 lots referred to, the number of cases considered was 161. These 161 cases represented the production of 116,842 pieces (or operations) in a week on a 55-hour basis, and a pro- duction of 114,228 pieces (or operations) per week, on a 60-hour basis, a net reduction in weekly production of approxi- mately 2.2%. The production per hour on a 55-hour basis was 13.2 pieces or operations, while on a 50-hour basis it was 14.3 pieces or operations, or a gain of 7.9% in production per hour for a 50-hour basis. Of the total 161 cases 125 showed greater production per hour on a 50-hour basis, 15 showed the same production per hour, while in 21 cases there was less production per hour. To show the unexpected variations which indicate that such figures as these are not final it may be noted that while in 65 cases of general machine work, the gain was 6% on the hourly basis, with a loss of 3% in production per week, in 25 cases of hand scraping the gain was only J^% on the hourly basis, with a loss of 7J^% in production per week. It will be noted that this experience relates to piece- work. With respect to daywork the following statement was made: There is a feeling among our foremen and executives that while we have nothing tangible to show what difference 21 54 50 260 245 321 304 730,080 637,000 Gain 8.5% 0.5%. there may be on daywork the percentage of gain would not be as high as is indicated by the above figures, and while we believe there is a gain per hour for the shorter hour week, it is felt that it would not be as great on the average as the figures here given. Following is a comparison of results under a 54 and 50- hour schedule furnished by a machinery establishment. The change in hours occurred in 1912. The same general type of machine was constructed in the two periods. The hourly rate of pay was 12 per cent higher under the 50-hour week. The change in hours was made by the introduction of a Saturday half-holiday, and the co- operation of the workers was requested by the manage- ment. 1911 1912 Hours per week .... ... Average number of workers Number of machines built Number of employee hours worked for the year . Increase in hourly output per employee . . Increase in weekly output per employee . . This comparison is, of course, distinctly favorable to the 50-hour week. The gain in hourly efficiency a little more than offset the loss in time, so that there was a small increase in total weekly output per employee. The superintendent of a large middle-western brass company stated: We are making a definite experiment with one department, the Pattern and Tool Department, in a reduction from 55 hours per week to 50 hours per week, with no reduction in wages per week, and we will say that so far very good results have been obtained. We are getting the same amount of pro- duction per week and we are also getting intensity of application and a spirit of co-operation from the men which we asked for, and which was a part of our agreement. The main question w'e are asking is whether this will continue after the cause which led to the arrangement has been forgotten. In other departments where we reduced the number of hours per week, because of our production requirements, from 55 hours to 40 and 45 hours per week, we have found that we are getting the same production per day in eight hours that we previously obtained in ten hours from most of the pieceworkers, except in a few cases where it was humanly impossible to do more, largely because of restrictions in the time element inherent in the process. 22 i^' It should be noted that In this instance the reduction in work-hours was due to a shortage of orders and was intended to distribute the work in hand over a larger number of employees. The following statement was submitted by a hardware establishment: This company in the past has operated on the basis of a 10-hour day, from time to time reducing to 9 hours. A number of years ago it established, with some misgiving, a Saturday half-holiday. These changes were voluntary, but each re- duction in the number of hours of labor was found to be attended with no reduction in the total production, which seemed to indicate that the energy of our employees was less impaired by the shorter hours, and the result was increased energy during the working period, so that they had the shorter hours and we had the production. Less favorable experience with reduction in work-hours was reported by other manufacturers. A large western machinery and tool-making establishment stated in part: Our change from 55 to 52 J^ hours per week was prompted merely by the feeling on our part that the very great amount of agitation regarding the so-called 8-hour day had created enough comment and attention that it might be only wise for us to plan to end the workday at 5 p.m. We have, however, found it necessary to run a considerable amount of overtime. On the other hand we believe that for continuous opera- tion the 8-hour shift is correct and proper; also for work requiring skill and concentration, heavy and continuous physical effort, etc. There is, however, certainly a very great difference between work that is largely dependent upon the skill, speed, agility, etc., of the employee, and work that is largely or wholly the product of machines where the speed of the machine determines the output. In the case of auto- matic machinery there can be no question, and in the case of semi-automatic, and even in the case of machines requiring feeding by hand and some attention, we do not believe that the same production (or more, as is often claimed) can be secured in 9 hours as 10, or in 8 hours as 9 or 10. We also believe that where the work is largely manu^il much of the increased production reported as taking place upon adoption of the new schedule is not permanent and that even on such work there is a decided tendency to drop back to as low a production per hour as the management will tolerate. An executive of a large concern making special ma- chinery stated: 23 A drop was made from 54 to 51 hours a week February 16, 1917. At first the hourly rate of production was slightly increased but later the employees drifted back Into their old speed. The superintendent and chief engineer of a large middle- western milling machine company made the following statement: When we reduced the hours in 1913 from 55 to 5214 per week there was acorresponding reduction in output for a time. However, as the result of more complete time studies, im- proved tools and special machinery, our output was later increased to above what we turned out in 55 hours. We have this month reduced our hours from 52 J^ to 48 per week, increasing the hourly rate so that the pay will be the same as for 52 }4 hours. Our output has again been cor- respondingly reduced but we hope to make some of it up, at least, by further improving our methods. The secretary of a malleable iron company: Relative to the effect experienced because of the reduction in the hours per week from 55 to 50, you no doubt realize that an answer to this inquiry is somewhat difficult and com- plicated for a variety of reasons, some of which are the various schedules under which we have operated within the past year, the character of the work we have been engaged on, the rate of output, etc., but after giving the subject careful thought and discussing it with our superintendent, we have reached the conclusion that under a 50-hour schedule the output per man has averaged a decrease of approximately 10%, while the cost has remained the same as it was previous to the change. An establishment which reduced from 54 to 50 hours: Our records show the output decreased in direct proportion with the decrease in the number of hours worked. An establishment which reduced hours in 1919: Output, in our business at least, is practically in proportion to the number of hours of work. This has been repeatedly verified. These various statements give a fair picture of the diversity of experience with and the diversity of attitude among employers toward a 48-hour and a 50-hour week in the metal manufacturing industries. 24 CONCLUSIONS AS TO OUTPUT For a 50-hour and a 48-hour week, in which interest chiefly centers, the evidence presented in the preceding sections may be summarized as follows: Effect on Output - - 48-Hour Group 1917 1919 Output Estab- lishments Per cent of Total Em- _ ployees in Group Estab- lishments Per cent of Total Em- ployees in Group Totals Increased Maintained .... . . Decrease, less than proportional . . . Decrease, about proportional ... Decrease, greater than proportional . . Decrease, amount not stated . . . 41 5 13 4 10 1 8 100.0% 67.1a 16.4 2.9 9.4 0.5 3.7 40 1 5 8 14 3 9 100.0% 9.1 8.4 21.2 44.0 1.5 15.8 Effect^on Output - - 50-Hour Group 1917 1919 Output Estab- lishments Per cent of Total Em- ployees in Group Estab- lishments Per cent of Total Em- ployees in Group Touds Increased ... ... Maintained Decrease, less than proportional . . . Decrease, about proportional ... Decrease, greater than proportional . . Decrease, amount not stated 66 4 28 12 10 2 10 100.0% 4.9 47.8 28.4 9.7 3.4 5.8 37 2 10 9 8 1 7 100.0% 8.5 12.6 42.0 11.0 1.6 24.3 aOver 66% of employees in the 48-hour group were in two establishments. Among the more important facts brought out by this comparison and the preceding discussion are the following: 1. A 50-hour week has proved efficient and practicable in a large number of metal manufacturing establishments. 2. A 48-hour week has proved practicable in a con- siderable number of establishments. 3. The piece-rate system is more conducive to current efficiency of production than is the day-rate system. 25 4. There Is no clear-cut line below which a reduction in hours brings a practically uniform change in efficiency of production in different establishments. A majority of establishments changing to a 50-hour schedule lost production, but the number and collective importance of those maintaining output is sufficient to demonstrate that a 50-hour week is a practicable schedule for a large proportion of metal manufacturing establishments. It is true that many manufacturers reported a loss in output even with a 54-liour or a 55-hour schedule. It must be remembered, however, that either because of some disadvantage with respect to location, class of labor, quality of material, nature of the work, faulty management, or other factor, some establishments will fail to maintain maximum output under almost any reduction in hours. Aside from the actual maintenance of output reported in many cases is the favorable attitude toward a 50-hour schedule by manjr employers who have introduced it. Thus of 40 manufacturers who had reduced to a 50-hour week and who replied to a query on this point in the 1919 questionnaire, a substantial majority preferred such a schedule, as the following summary shows: 29 preferred a 50-hour week 3 preferred a 54-hour week 8 preferred a 55-hour week Experience with a 48-hour week likewise was favorable in a sufficient number of cases to invest such a schedule with a high degree of interest, and there can be no doubt that a 48-hour schedule will under certain conditions yield maximum efficiency. Several of the establishments which reported favorable experience with such a schedule were, however, either so distinctive as to organization, or other conditions, that their experience cannot be fairly regarded as typical. Where, as in some plants, the work is largely of an assembling nature and can be standardized to a very high degree, and where workers are driven at high pressure, or where the work calls for unusual precision, the results secured may be quite different from those reasonably obtainable in an ordinary metal trades establishment. But where production is practically dependent on highly automatic machinery, a 48-hour week cannot be expected 26 to maintain maximum output. Furthermore, con- clusive judgment as to the merits of a 48-hour week cannot fairly be reached until the effect of speed on the fatigue and health of workers has been more thoroughly studied. It may be noted that many employers who had adopted the 48-hour week were satisfied with such a schedule. Thus of 40 establishments employing 25,583 employees working under a 48-hour schedule 20 preferred a 48-hour week 10 preferred a 50-hour week 5 preferred a 54-hour week 4 preferred a 55-hour week 1 preferred a 60-hour week Taken in connection with the favorable attitude to- ward a 50-hour week by employers who had reduced to such a schedule, it appears that of 80 establishments which i^educed either to 50 or to 48 hours, only 21 ex- pressed a preference for a week longer than 50 hours. It is true that establishments operating on a 54 or a 55 hour week in nearly all cases expressed a preference for the particular schedule they were using. The experience of these manufacturers who had not reduced hours below 54 per week cannot, however, be given equal weight in reaching conclusions as to the practicability of a shorter week with the experience of employers who had adopted shorter schedules. Distinction should be made between what can be done and what as a practical matter will be done. Much depends upon the degree of co-operation secured between the management and its workers. If full co-operation to attain the highest reasonable efficiency could be secured there can be little doubt that a 50-hour week could be generally adopted in the metal trades without serious loss of production. Several of the statements given in this report by manufacturers who maintained production when hours were shortened refer to such co-operation. Likewise, the evidence indicates that under such conditions a 48-hour week could be made an economic schedule in a much larger proportion of estab- lishments than is now possible. No single factor could do more to accomplish this result than the recognition by workers of the indisputable principle that their real interest and that their real opportunity for bettering 27 their condition lies in increasing the efficiency of indivi4ual production. But unless both management and workers actively co-operate to this end, there can be little question that the general adoption of a 48-hour week in the metal trades would involve a serious economic loss to the nation. The evidence clearly indicates that piecework is more conducive to efficiency than is daywork. In this con- nection, the rather low percentage of pieceworkers in the industry as a whole as compared with the other industries thus far investigated suggests that there is opportunity to increase efficiency by increasing the number of workers on a piece-rate basis. The difficulties in the way of this are often great, especially where quality is the prime consideration. Extension of the piece-rate system, moreover, imposes a greater burden on management. The fact, however, that only about one-third of the workers in the metal trades establishments covered by this report were on a piece-rate basis suggests that there is opportunity for improvement in this direction. It is of course essential to guard against a system of piece- rates which leads to overspeeding. A striking feature of the evidence brought out by this study, and one which has been noted in the reports on several of the other industries included In the Board's investigation of the hours-of-work problem. Is that there is no clear-cut line below which a reduction In hours brings a practically uniform change in results. Thus, despite the fact that a substantial number of establish- ments maintained production on a 48-hour week and a rather large number on a 50-hour week, a majority of establishments reducing only to 54 or 55 hours per week nevertheless reported that production was decreased. This diversity of experience is characteristic of industries where production Is not virtually controlled by ma- chinery and must constantly be kept In mind. It means that reductions In work-hours are only one factor in the problem. Unquestionably, the difference in results under different hours-of-work schedules Is attributable In many cases to differences in efficiency of management. On the other hand. It may be that an establishment which failed to maintain production on a 54-hour week may be quite as efficient as one which maintained production on a 50- hour week; the difference in result may be due to differ- 28 ences in the nature of the process, in the character of the raw material, in the type of machinery, or in the type of worker employed. Exact mathematical measurement of the effect of changes in hours alone is not possible. Conclusions must necessarily be drawn in a broad way. The value of this study of experience in the metal trades lies in its positive evidence that it is possible not merely for oc- casional and exceptional establishments but for a con- siderable proportion of establishments to maintain production on a schedule of 50 hours per week, and that a 48-hour week is a practicable one for a limited proportion of establishments in these industries. Whether such schedules are practicable in individual cases is a problem for the particular establishment to determine. It is obviously to the interest of employers to make a thorough study of conditions in their respective plants in the light of the evidence presented in this report. 29 HEALTH OF METAL WORKERS Conclusions as to the effects of reductions in hours of work on health of workers in the metal trades industries cannot be reached at this time, for the reason that data covering a sufficient period of time to permit satisfactory study are not in existence. Of those establishments replying on this point, 8 reported an improvement in health conditions following reductions in hours to 50 or 48 a week; 84 others reported no notice- able change. In many instances, however, the state- ments were based on opinion only. In the absence of a definitely recognized standard for the measurement of health conditions such opinions cannot be given the same weight as those concerning output. Furthermore, it is necessary to take into account the effects of a great number of collateral factors which unquestionably have an important bearing. As was stated in an earlier report on the hours-of-work problem,^ "such influences as nationality, inheritance, personal hygiene and habits have a very important bearing on health, so that even though a change in hours of work may have a significant effect, this may be overshadowed by the collective effect of other factors." Thus, arepresentativeof alargeestablishment conducting practically all branches of metal manufacturing stated: Inasmuch as several changes were made at the same time, it is impossible to say how much of the benefit which followed was due to any one matter. It is impossible to say how far we can attribute better conditions to the introduction of the eight-hour day. We are convinced, however, tha|t the shorter day does conduce to larger output, better quality of work, better health conditions, to the decrease in number of acci- dents, and to the contentment of our workers. As noted in its report on the hours-of-work problem in the silk industry, the National Industrial Conference Board secured the co-operation of a large number of manufacturing establishments for the keeping of absence records in industry for a 24-day period in 1918 and a similar period in 1919. While such records will throw little light on the effect of hours of work on health until kept over long periods, they should be of value in arriving •Research Report No. 16, "Hours of Work as Related to Output and Health of Workers — Silk Manufacturing." 30 at broad conclusions regarding general health conditions in industry. Furthermore, it is hoped that the keeping of these records for specific periods may lead to keeping them permanently by leading manufacturing establish- ments and thus provide a basis for determining the underlying causes of sickness and result in preventive measures. Only by careful compilation of such data showing the amount and the causes of absence in a large number of industrial establishments will it be possible to present valid conclusions as to the effects of hours of work or other specific industrial factors upon the health of workers. Such data should, however, be supplemented by intensive study of other factors as well.^ General Health Conditions in the Metal Manu- facturing Industries Factory conditions in metal manufacturing establish- ments vary greatly, a reflection, in part at least, of differ- ences in the legal requirements of the various states. Some establishments are models of modern sanitary engineering and go far beyond requirements of the law, while other establishments are rendered more or less hazardous by dampness, metallic dust, extremes of temperature, insuffi- cient ventilation and poor lighting, the latter feature being perhaps most common. Statements of manufacturers furnished in the course of this investigation indicated that, on the whole, health conditions in metal manufacturing compare favorably with those in other manufacturing industries. Certain occupational diseases, such as lead and brass poisoning, are prevalent in the metal trades; lead poison- ing, however, is not characteristic of the particular branches of the metal manufacturing industry included in this investigation. In some processes acid and other fumes introduce an occupational hazard.^ Tuberculosis among Metal Trades Workers However, the death rate from tuberculosis in certain branches of the metal manufacturing industry is high. This is clearly shown by Table 5, compiled from data collected in the United States Census of 1909. ' See footnote, p. 41. Research Report No. 16. Thus E. R. Hayhurst, in "Industrial Health -Hazards and Occupational Diseases in Ohio" (1915), reports after a study of 52 forging and blaclcsmithing plants: "Poisoning by fumes or otherwise was a serious hazard in 23 plants, fairly so in 6 more, and negligible in the remaining 23. The chief danger was from chronic gassing, due to the escape of fuel fumes and smoke. In addition, tempering and case hardening with lead, potassium cyanide and oil, were carried on in a number of blacksmith shops." 31 iJ Q < W r^ E Sg , w w lu a CO h ix fe m o d ^ H m P< P< w a U fc »< J ^d »> >-* CO g w <; CO M 5 ^' < w ^ S S o §§ Pi . <1 ^ l-H M (J I- << S !^ CO a is H ^ < rn U «j w w Ui 5; PC O o W^ So S 1^ P^ n ro u 5 Hi O ^ CO P^ M o •• O r^ M O W H < n u o sa 5g gcog cC>l>'«*«DCOOt>-QO'^i-0 (M CD <© OS o gtcS " f,3, pen c A P. a w Q h O D O 10l>.CCD'-l.-(01MO«>>'53 CM '^'~' 0000Ttl'-IIN"5OQ0Qq00C0t>^Ol>;>n T-itooat-eoMOiOiMtoeoooqiMcooq ui^'^^(M,-iu5-stidd»-l"5O>0«5 1-*^ T-l T— 1 l0-*ON'*CD05CD0s0lT-tO'-]00»-^_"^ Ki^.^l>;(NU51>!odt-Hd'-IINO(M'-jt~; •^ «-4 T^ I— t 00i-i-*«3c> u 5 y 2 u u oj (=^ u u 3'ir«-^'S^-5 3'r. .3 « 3 u "S o. t3 rt c Ct: .i; -JS >> u 5 u <3 PlI|0 Dr. E. R. Hayhurst, "Industrial Health-Hazards and Occupa- tional Diseases in Ohio." — 1915. 36 that during the five-year period ending in 1913 the Union recorded 204 deaths for which benefits were paid. The chief causes of death were as follows, in order: pneumonia, 30; heart trouble, 30; tuberculosis, 27; violence, 18. It will be seen that this comparison shows an unusually high death rate from pneumonia as well as from tuberculosis. One explanation of this is the change from extremes of heat and cold which characterizes the foundry industry, it being not unusual for workers after laboring in the presence of excessive heat to go immediately into cold air. While the basis of experience included in the foregoing tables is doubtless too limited to permit of definite con- clusions as to the ratio of deaths from tuberculosis in the several branches of the metal manufacturing industry, the evidence is clear that for a part of the industry the death rate from tuberculosis is exceedingly and unneces- sarily high. It should, however, be pointed out that in general the branches characterized by exceptionally high death rates from this disease employ a relatively small proportion of the workers in the industry. It is obviously incumbent upon industrial managers to make intensive study of the conditions peculiar to these branches of the industry, with a view to devising some means for bringing about a reduction in this excessive rate. Furthermore, the rate is so high for the industry as a whole as to call for careful study of general working conditions and more exact knowledge of the causes contributing to tuberculosis. Until this is done no definite constructive policy can be laid down. In view of the fact, however, that tuberculosis is a disease more or less common to the human race in all walks of life, such a study must embrace conditions outside of the factory as well as those within the factory walls. As an illustration of the importance of studying these other factors as well as conditions within the factory, the following statement may be cited •} The familiar Berlin statistics showed that 42 per cent of all the cases of tuberculosis occurred in families occupying but one room, 40 per cent where they occupied two rooms, 12 per cent where they occupied three rooms, and only 6 per cent where they occupied four or more rooms. Williamson pre- sented English statistics which showed that 60 or 70 per cent of the cases of tuberculosis occurred in houses of three 'British Journal of Tuberculosis, IX, pp. 111-117, July, 1915. 37 rooms or less and that the number of cases was larger in two- room houses than in three, and larger in one-room houses than in two. In connection with this statement the question should, however, be borne in mind as to how far these crowded housing conditions may have been due to a low wage. There is considerable evidence to indicate that there is a definite connection between the prevalence of tuber- culosis and inadequate nourishment and poor housing conditions. Racial characteristics and personal habits must also be carefully considered. Accident Hazard Owing to the great variety and intricacy of much of the machinery used, the high speed at which it is often operated, the use of acids and poisons, and the intense heat characteristic of certain operations, the accident hazard in the metal manufacturing industry is compara- tively high. Thus in the State of Ohio the insurance rate per $100 of wages paid, as indicated by the rate manual of that state, shows a distinctly higher accident hazard for metal manufacturing than that for several other leading industries.^ As noted on page 33 however, census returns show a lower average death rate from accident for the industries covered by this report than the average for all manufacturing and mechanical industries combined. 1 The following table shows the rates for various industries as quoted in the Ohio Rate Manual in effect July 1, 1918: Boot and shoe manufacturing ... , $ .24 silk manufacturing . .25 Wool combing . . , 25 Printing 26 Textile dyeing, finishing, and printing 30 Rubber garment manufacturing . . 50 Rubber tire manufacturing 65 Wool spinning and weaving 65 Shoe stock manufacturing . .70 Blacksmlthlng exclusive of shoeing . . .... .80 Electric apparatus manufacturing ... .80 Cotton spinning and weaving 85 Brass foundry ... 90 Automobile manufacturing 95 Earthenware and clay pipe manufacturing 1.10 Pulp manufacturing, no saw or barking mills 1.10 Blacksmithing and shoeing ... 1.15 Auto parts (sheet metal), etc 1,16 Can manufacturing 1.20 Machine shop, without foundry 1.25 Pattern making 1.25 Molding 1.35 Machine shop, with foundry 1.45 Wool pulling 1.50 Iron Foundry 1.55 Rolling mills 1.75 Machinists 2.10 Boilermaklng 2.20 Steel castings 2.40 Paper manufacturing, except writing and bond and ledger 2.55 Blast furnace operatives 3.25 Sawmills 3.85 38. Conclusions as to Health While definite conclusions as to health hazards in the metal trades cannot be reached from the evidence at present available, on the whole it appears that the worker in these industries occupies a fairly favorable position, judged by the average hazards of industrial employment. The tuberculosis death rate for certain occupations, such as brass working, is exceptionally high. This unfavorable condition is, fortunately, modified by the fact that those occupations having an unusually high tuberculosis death rate include a relatively small number of workers. The tuberculosis death rate for the industry as a whole, however, as for many other industries, is excessive, but until much more is known as to the influence of housing conditions, personal habits and hygiene, and racial or family predisposition, the responsibility of industry alone cannot fairly be measured. As noted in a similar report on the cotton manufacturing industry there is an increasing tendency among those making scientific study of the tuberculosis problem to attach more weight to these outside factors than to the occupation itself.^ This does not relieve the industry, or such a specific factor as the length of the work-week, from its full share of responsibility. But any constructive policy for reducing the prevalence of this disease among industrial workers must give due consideration to the other factors in the problem. * National Industrial Conference Board. Research Report No. 4, "Hours of Work as Related to Output and Health of Workers — Cotton Manufacturing," p. S5, footnote. 39 Appendices Tables A and B epitomize the important data sub- mitted by those establishments which reported the effects on output accompanying reductions in work-hours. For convenience, the information is assembled by hour groups and the establishments classified according to the effect on production. A separate classification by branches of the industry was not made, as no characteristic differences in experience in this respect was apparent. In previous reports of this series changes in rates of wages accompanying reductions in hours were indi- vidually noted in the appendix. In the case of the metal trades, rates of wages were almost invariably increased at the time of the reduction in work schedules and in a great proportion of cases were increased to a compensating degree. A detailed statement of changes in wage rates would, therefore, throw little additional light on the problem and in order to economize space this information has been omitted. 41 Appendix A Results Reported in Replies to the 1917 Questionnaire REDUCED TO 45 HOURS Production Maintained Establishments ... 1 Employees 278 Date of Previous Number Change. Hours Employees Product Statement of Management 1917 50 278 Steel tubes. Production has been maintained. REDUCED TO 48 HOURS Production Increased Establishments 5 Employees 37,499 Date of Previous Number Change Hours Employees Product Statement of Management 1915 60 69 Small tools and 5 to 10% increase. machinery. 1915 55 2,509 Machine parts. Production increased slightly. 1916 54 124 Machinery. Increase of output due to improved methods. 1914 54 34,571 Automobiles. Production increased between 15% and 20%. 1916 54 226 Automobiles. Quality of work improved and quan- tity increased. Production Maintained Establishments 13 Employees 9,167 Date of Previous Number Change Hours Employees Product Statement of Management 1913 60 1,270 Hardware. Records do not show any appreciable change. 1907 60 273 Farm tools. Production almost as great. Men speed up. 1914 55^ 1917 50 1917 50 Product Statement of Management Small tools. Daily rate of output the same. Machine parts. Normal output retained by slightly higher pressure. A few automatic machines were added. Output maintained; quality im- proved. Telephone appli- Satisfied that most employees are ances. turning out as many parts now in 48 hours as they formerly did in 50. 618 Machine parts. 736 Production Decreased Establishments Employees Number Employees Product 80 Machinery and tools. 154 Meters. 30 Automatic apparatus. Silverware. 23 9,238 1917 4SH 640 325 1,900 247 172 363 609 550 554 89 529 284 362 200 46 250 771 904 40 149 Metal products. Automobiles. Machinery. Machinery. Machinery. Machinery. Machine tools. Machiner}'. Machinery. Steel tanks. Machine tools. Machinery. Chains. Machinery. Hardware. Magnetos. Auto starters. Machine parts. Steel tanks and pipes. Statement of Management Decrease in output. Less production. Slightly less output. A decrease in output less than pro- portional to the reduction in hours. Production correspondingly de- creased. Proportional reduction in output per operator. Less work. Corresponding reduction of output. On the whole a loss in output. A few workers succeed in doing 9 hours work in 8 hours. 5 to 10% decrease in daywork. Weekly output decreased propor- tionally. 10% decrease. Output decreased in proportion to time. Output decreased in proportion to time. Less work done. Loss of production. Reduction in day and piece work. We were already at our limit as far as speed of machines and operator* was concerned. Production per hour remains same as before. Decrease in day and piece work of 10%. Weekly output decreased. Piece and day work decreased 4%. ■ Decrease in output. More overtime required to complete contracts. Diminished production. a This eBtablishment reduced to 47H hours. 43 Date of Change 1916 1917 Previous Hours 55 52 REDUCED TO 49}^ HOURS Production Increased Establishments Employees . . Number Employees 148 186 2 334 Product Electrical appli- ances. Special machinery. Statement of Management Better feeling, also increased pro- duction. A greater production in the shorter hours. Date of Change 1915 Previous Hours 55 Production Maintained Establishments Employees Number Employees Product 488 Automobiles. Statement of Management No appreciable reduction in output. Date of Change 1917 1902 1902 1917 Previous Hours 55 55 55 54 Production Decreased Establishments 4 Employees 1,428 Number Employees Product 385 Automobiles. 776 Steel castings. 167 100 Iron castings. Iron castings. Statement of Management Small decrease in output. Production of daywork decreased proportionally. Proportional reduction in output. Decrease of about 5% in production. Date of Change 1917 1913 1916 1916 Previous Hours 55 55 54 54 REDUCED TO 50 HOURS Production Increased Establishments 4 Employees 3,424 Number Employees Product 189 Silverware. 3,126 40 Machinery. Gifice machinery. Hardware. Statement of Management Records show an increased produc- tion. Increased production. We experienced a decided increase in production. Daywork increased 10%. Piecework increased 5%. Increase probably 20% in piecework > 75% of workers are on piecework. Production Maintained Establishments 28 Employees . 33,349 Date of Previous Number Change Hours Employees Product Statement of Management 1916 60 320 Machinery. No material change. 1916 59 965 Automobiles. Increase in output per hour 15%. No piecework. 1910 59 377 Machinery. Production maintained. 1913 67 10,030 Automobiles. No effect on output, change affecting only Saturday afternoon. 44 Date of Previous Number Change Hours Employees Product Statement of Management 1916 56K 172 Machinery. Output about the same. 1916 55 101 Steam blowers and iron castings. Weekly output maintained. 1917 55 447 Small tools. Production maintained. Employees attend more strictly to biisiness and there is less absence. 1917 55 450 Small tools. Output is maintained. 1915 55 2,760 Machine parts. Maintained for the most part. 1917 55 154 Machinery. Output about the same. 1917 55 146 Machine parts. Output maintained. 1908 55 8,070 Steel castings. No apparent effect in production or costs. 1916 55 600 Alloy castings. No important effects noticed. 1917 55 287 . Automobiles. Total weekly output maintained. 1916 55 482 Automobiles. Same output; better satisfied men. 1913 55 737 Instruments and transformers. Believed we would get equal amount of work and better quality in 9 hours than in 10. This has proven true. 1916 55 199 Electrical appara- tus. Output unchanged. 1914 54 514 Engines. By the Emerson standard we get as much from the 50-hour as we did from the 54^hour schedule. 1917 64 750 Electric motors. Speeded up production; concluded we could get about all there was in the employees out of them in 50 hours per week. 1916 54 504 Machine accessories. No appreciable reduction in pro- duction. 1912 54 289 Farm implements. No noticeable change. 1917 54i 175 Special hardware. Output same as before change. 1917 54 268 Steel tanks. Output maintained. 1912 54 302 Machinery. Perceptible increase at first but grad- ually returned to normal condi- tions. 1915 53 112 Machine tools. No noticeable effect anywhere. 1915 52}4 336 Plumbers' tools. No appreciable loss in production by the change. 1916 52K 3,496 Machine parts. No change. in output. 1916 52 306 Office appliances. Output maintained. Production Decreased Establishments . . . 34 Employees . 33,026 Date of Previous Number Change Hours Employees Product Statement of Management 1912 60 37 Machinery. Decrease of 10% in output. 1907 60 29 Machinery. Decrease of 10% in output. 1917 57M 32 Hardware. Piecework maintained; daywork de- creased 11%. b Reduced to 50H hours. 45 Date of Previous Change Houis 1917 55* 1915 55 1916 1916 1915 1914 1916 55 55 55 55 1916 55 1915 55 1915 55 1917 55 1916 55 1902 55 1902 55 1917 54 1916 54 1917 54 1901 54 1904 64 1901 54 1916 54 1916 64 54 Number Employees 707 236 394 3,819 1,101 455 340 460 118 97 593 Product Machinery. Hardware. Hardware. Automobiles. Automobiles. Valves and hy- drants. Machinery. Machinery. Office appliances. Wire products. Sheet metal. 578 Marine engines. 100 1,034 163 56 723 10 244 49 149 Machine parts. Farm implements. Welding apparatus. Office appliances. Fire fighting apparatus. Machinery. Meters. Special machinery. Wire products. 1,808 Machinery. 1916 1915 54 54 631 2,760 Machinery. Automobiles. 1915 1914 54 54 314 269 Automobiles. Electrical appli- ances. 1916 53^ 669 Electrical appara- tus. 1916 1916 1917 63 52Ji 62 196 1,868 13,075 Architectural iron. Wires and cables. Electrical appara- tus. 1916 51M r. 60 to 56. 12 Electrical appara- tus. •Foundij Statement of Management Proportional loss in output. In a few cases some improvement, but in most cases less work. About 10%Jless output. Effort to get same^production only partly successful. Production somewhat reduced. Records kept for three months before and after change show that amount of work per man per hour re- mained same. Less than proportional decrease for whole mill. Pieceworkers decrease 6%. Proportional decrease in output. Less production per man. Weekly output decreased. Production decreased. Decrease in production; somewhat better attendance. Output decreased more than in pro- portion to reduction in hours. Output decreased 10% per man. Daywork^output decreased about 5%. Piecework maintained. Proportional decrease. Output decreased; no piecework. Proportional decrease. A reduction in output. Output reduced; percentage not known. Output decreased. Output decreased; men speeded up somewhat. Decrease greater than proportional to the reduction in hours, partly on account of inexperienced help. Slight decrease in output. Approximately 7% decrease in out- put. Small reduction. Proportional decrease. When our machinery stops, production also stops. Increased production through greater force, etc., offset result of lesser hours. No changejin hourly production. Output decreased. Pieceworkers maintained, dajr- workers decreased, but not in proportion. Slight decrease in output. 46 REDUCED TO 51 HOURS Production Maintained Establishments 3 Employees . 1,100 Date of Previous Number Change Hours Employees Product Statement of Management 1904 55}4a 126 Hardware. Tried it out and found we produced as much in 5J^ days. 1917 54 474 Gas appliances. Output remains the same. 1917 54i 500 Auto trucks. No decrease in production. Production Decreased Establishments 3 Employees . 1,453 Date of Previous Number Change Hours Employees Product Statement of Management 1917 54 155 Machinery. Less production. 1916 54 631 Machinery. Decrease in production. 1916 54 667 Valves. /Slight decrease in output. Date of Change Previous Hours 55 Date of Change 1915 Previous Hours • 54 REDUCED TO 52 HOURS Production Increased Establishments 1 Employees . 262 Number Employees Product Statement of Management 262 Auto trucks. Output increased. Most noticeable in the assembly departments. Production Maintained Establishments 1 Employees 23,850 Number Employees Product Statement of Management 23,850 Electrical appara- Our output remained practically the tus. same. Date of Change 1917 1915 Previous Hours 55 54 Production Decreased Establishments 2 Employees 3,503 Number Employees Product Statement of Management 250 Sheet metal Estimated decrease of 15% in out- products, put. 3,253 Railroad supplies. Pieceworkers increased output. Date of Previous Change Hours 1917 56J^ 1915 55 REDUCED TO 52J^ HOURS Production Increased Establishments 3 Employees 1,744 Number Employees Product 36 Machine parts. 600 Machine tools. Statement of Management Output slightly increased. Output increased; greater satisfac- tion and loyalty among employees. a Hours in this establishment, 51)^ per week. b Hours in this establishment, 51M per week. 47 Date of Previous Number Change Hours Employees Product 1913 55 1,108 Machinery. Statement of Mana^ment Proportional reduction for a time. However as a result of more com- plete time studies, improved tools and special machinery, our output was later increased to above what we turned out in 55 hours. Production Maintained Establishments Employees . Date of Change 1910 1916 1914 Previous Hours 57 55 55 Number Employees 20 786 603 Product Ironwork for buildings. Machinery. Machine tools. 1916 55 411 Machine tools. 1917 55 40 Machinery. 1916 65 44 Architectural iron. 1916 55 111 Meters. 1913 55 1,198 Engineering specialties. 1915 55 649 Machinery. 1917 55 1,161 Machine parts. 1917 65 711 Machinery. 1913 55 136 Engineering specialties. 1916 64 684 Appliances. 1914 63 1,032 Special hardware. 14 7,586 Statement of Management Output about the same. Output maintained. Output maintained. Output maintained. Output maintained; better time kept among employees. Output about the same as before change in hours. Output maintained; both piece and day workers producing more per hour. Output maintained; increase in effi- ciency. Change in output not noticeable. About the same output. No reduction in output. Output maintained. Output maintained; more steadiness in attendance. Same output as before change in hours. Production Decreased Date of Change 1913 1907 Previous Hours 60 58 1916 1916 571^ 65 1917 1913 55 55 1913 1913 65 65 55 Establishments Employees . . Number Employees 140 225 87 70 831 155 246 46 Product Machinery. Machinery and castings. Machinery. Special railroad equipment.. Machinery. Motor fiTe appara- tus. Machine tools. Machinery. Machine tools. 48 14 6,431 Statement of Management Diminished output of machine tools having a fixed rate of speed. Otherwise no effect observable. Production reduced. Less work turned out. Output decreased. Proportional decrease. Output decreased 5%. Output decreased 5%. Output decreased. Less production. Date of Previous Number Change Hours Employees Product Statement of Management 1917 55 1,500 Electrical appliances. No change in piecework; a slight decrease in daywork. 1916 55 632 Machinery. Output decreased. 1916 55 972 Machinery. Decrease in weekly output 5%. 1916 55 550 Machinery. Decrease of 5% in daywork; no change in piecework. 1915 54 879 Machine parts. Output decreased 3%. REDUCED TO 53 HOURS 1 Production Increased Establishments . . 1 Employees ... 71 Date of Previous Number Change Hours Employees Product statement of Management 1915 59 71 Sheet metal products. Production of previous year was not only equaled but considerably in- creased. Production Maintained Establishments 3 Employees . 613 Date of Change 1917 1917 1910 Previous Hours 59 59 54 Number Employees 334 210 Product Machinery. Electrical appliances. Household and ofSce appliances. Statement of Management Output maintained. Production did not change appre- ciably. Production maintained. REDUCED TO 53Ji HOURS Production Maintained Date of Change 1916 Previous Hours 67}^ Establishments Employees Number Employees 500 1 500 Product Auto horns, fans, etc. Statement of Management Production remains about the same, Date of Change 1917 Previous HoiuB 57}^ Production Decreased Establishments . ... Employees ... Number Employees Product 153 Machinery. 1 153 Statement of Management Decreased output. Date of Change 1913 1916 Previous Hours REDUCED TO 54 HOURS Production Increased Establishments 2 Employees 825 Ntmiber Employees Product Statement of Management 729 Special vehicles. Output increased. 96 Marine hardware. More production. Production Maintained Date of Change 1917 1902 1907 1917 1910 1916 1913 1917 1915 1913 1902 1914 Previous Hours 60 60 60 60 60 60 60 60 60 60 60 1914 60 1907 59 1916 56 1916 56 1913 55 Establishments Employees . . 17 22,185 Number Employees Product 466 Hardware. 7,493 Railway supplies. 134 Iron castings. 820 Stoves. 427 462 205 670 1,406 Steel castings. Iron castings. Machinery. Farm implements. Machinery. 402 2,128 Machinery and castings. Office appliances. 64 Machine parts. 2,733 237 Machine parts. Architectural iron. 200 32 4,306 Power transmitting machinery. Small tools. Automobiles. Statement of Management Output maintained; a merit bonus system installed. No important effects. Output maintained. Output maintained by both day and pieceworkers. Output maintained. No change in output. Production about the same. Almost same production. Maintained. Partly due to speeding up machines. Output maintained. The employees seem to be better satisfied and the company gets just as much work. Approximately as much output in 9 hours as was formerly gotten out in 10. Output maintained; men speeded up. No noticeable change. No difference in output. No noticeable change. Same amount of work accomplished in 9 hours as formerly in 10. Production Decreased Date of Change 1903 1912 1914 1914 1912 1906 1903 1912 1916 1916 1900 1911 1904 1916 Previous Hours 61 60 60 60 60 60 60 60 60 60 60 60 60 Establishments Employees . . Number Employees 76 294 160 60 79 112 525 33 14,197 316 254 768 249 126 74 818 Product Plumbers' supplies. Plumbers' supplies. Iron and alumi- num castings. Iron castings. Iron castings. Machinery. Machine appli- ances. Machinery and tools. Machinery. Edged and farm tools. Boilers. Machinery. Machine parts. Machine parts. 50 Statement of Management Total weekly output decreased. Total weekly output decreased. Production reduced proportionally. Production decreased proportionally. Production largely curtailed. Proportional decrease. Proportional decrease. Production reduced 10% for day- workers; 5% for pieceworkers. Decreased production. Day work decreased 10%; piecework maintained. Output decreased proportionally. Output decreased 10%. Output decreased 10%. Output decreased. D»te of Previous Number Change Hours Employees Product 60 20 Machinery. 1915 60 82 Hardware special- ties. 60 670 Plumbers' hard- ware. 1912 60 720 Architectural steel and bronze. 1901 60 282 Agricultural imple- ments. 19&2 60 225 Light structural iron. 60 396 Plumbers' hard- ware. 1917 60 92 Stoves. 1906 59 59 Machinery. 1907 59 1,048 Twist drills and reamers. 1901 58H 361 Machine appli- ances. 1917 58 85 Machinery. 1917 68 150 Machinery. 1916 58 2,075 Brass goods. 1904 57 96 Machinery. 1915 57 847 Wire products. 1912 56 266 Machinery. 66 1,605 Machinery. 1916 55 1,307 Machinery. Statement of Management Output decreased proportionally. Daywork output decreased. Piece- work output maintained. Output decreased but not in propor- tion to the time. Output decreased 9%. Output decreased 5%. Decrease in output. Reduction of output. Pieceworkers maintained. Day- workers lost 5% in output. Proportional decrease. Proportional decrease. Decrease in output about 7% for piecework; about 5% for day- workers. Slight decrease in output. Output decreased. Output decreased proportionally. Output decreased. Loss in production. Loss in production. Loss in production. Loss in production. Date of Change Previous Hours 1915 59 1914 59 1915 57J^ REDUCED TO 55 HOURS Production Increased Establishments 3 Employees 10,724 f Number Employees Product 372 Machine parts. 10,222 Automobiles. 130 Automobiles. Statement of Management Production increased. Daywork not affected; piecework in- creased 8%. Increase in total output per em- ployee. Date of Change 1917 1909 1916 Previous Hours 61 60 60 Production Maintained Establishments 24 Employees 8,478 Number Employees Product Statement of Management 363 Insulation No piecework. Reduction in hours products. increased eiHciency and, by im- proving quality, lowered percent- age of spoiled work, thus main- taining average of effective pro- duction. No noticeable change. Jactu^^ th^ame output. 59 76 Steam engines Date of Change 1913 1912 1901 1902 1911 1916 Previous Hours 60 60 60 60 60 60 Number Employees 1,237 127 350 536 105 744 1 Product Machine needles. Machinery. Wire products. Machine parts. Machinery. Brass and iron castings. 1915 59 199 Hardware. 1917 1916 1916 59 59 59 264 231 343~ Machinery. Plated silverware. Plated silverware. 1916 59 94 Marine engines. " 1916 59 59 38 775 Machinery. Machinery. 1916 1917 1915 59 58 58 722 86 240 Machinery. Hardware and castings. Machinery. 1916 68 1,458 Machine parts. 1909 58 75 Machines. 1916 1917 57}^ 57y2 57H 230 17 109 Machinery. Machinery. Wire goods. Statement of Maaagemeat No effects noticeable. No effects perceptible. Output maintained. No noticeable change in production. No perceptible change. Obtain as great a production in the shorter hours as on former sched- ule. Also think change has been beneficial to the employees. We get just the same results in 55 hours as in 59. Output maintained. Output practically the same. Production about the same. By judicious arrangement of ma- chines we have 'observed no de- crease in production. No noticeable effects. Output maintained; men keep better time. Production maintained. Production nearly normal. Employees more contented. Out- put maintained. Output maintained. Attendance better. Output maintained. Machinery speeded up. Apparently no effects. Output maintained. Estimated that change in hours caused no reduction in output. Production Decreased Establishments . .... 35 Employees . Number . . . . 23,670 Date of Previous Change Hours Employees Product Statement of Management 1916 60 683 Iron castings. Output reduced. 1916 60 446 Silverware. 10% reduction. 1916 60 3,199 Iron and steel castings. Records show reduced output. 1916 60 1,500 Metal castings. Decreased 4 to 5%. 1917 60 236 Silverware. Output decreased proportionally. 1916 60 201 Silverware. Output decreased. 1914 60 1,800 Shipbuilding. Proportional decrease. 1908 60 254 Sheet metal. Production decreased. 1917 60 576 Engines and boilers A decrease in output. 1902 60 910 Machinery. Proportional decrease in output. 1915 60 169 Machinery. Decreased 8%. 1907 60 96 Electrical ma- chine parts. 52 Weekly output decreased 7%. Date of Previous Number Change Hours Employees Product Statement of anagement 1900 59 148 Machinery. Output decreased. 1913 59 1,018 Machinery. Less production. 1917 59 95 Machine parts. Output decreased. 1916 59 1,273 Silverware. Curtailment of production. 1916 59 200 Silverware. Output decreased 5%. 1907 59 5,963 Machine tools. Proportional loss for dayworkers; pieceworkers (90% of the force) eventually maintained. 1915 59 63 Hardware. Less production. 1905 59 231 Small tools. Proportional decrease. 1915 59 42 Machinery. Output decreased. 1913 59 696 Machinery. Reduction in output less than pro- portional. 1902 59 913 Machinery. Reduced output 7%. 1916 59 608 Machine parts. Output diminished somewhat. 1917 59 697 Machinery. Reduction in output. 1910 59 44 Machinery. Less production. 1917 59 219 Machines and castings. Piecework practically maintained; daywork reduced proportionally. 1916 58 50 Iron and brass castings. About 10% less. 1910 58 125 Automobiles. 2% decrease. 1907 58 66 Machinery. Slightly decreased. 1916 58 121 Heating appara- tus. Less than proportional decrease on the whole output. 1914 58 132 Machinery. Proportional decrease. 1916 58 66 Machinery. Proportional decrease; no piecework. 1917 57>^ 360 Hardware. Output decreased less than propor- tionally. 1913 57H 470 Machinery. A decrease in output. REDUCED TO 56^ HOURS 1 E>roduction Decreased Establishments . . . 1 Emolovees . . ... 231 Date of Previous JmJ^A^^A^Jj ****** A Number Change Hours Employees Product Statement of Management 1911 60 231 Machinery. Proportional decrease. Date of Change 1916 Date of Change 1916 Previous Hours 60 Pre^ous Hours 60 REDUCED TO 57 HOURS Production Maintained Establishments Employees Number Employees Product 100 Iron castings. 1 100 Statement of Management Slight decrease in production at first, followed by a slight increase. REDUCED TO 58 HOURS Production Maintained 1 118 Establishments Employees Number Employees Product Statement of Management 118 Machine accessories. No important effects. 53 REDUCED TO 59 HOURS Production Maintained Establishments 1 Employees 500 Date of Change 1916 Previous Hours 60 Number Employees Product Statement of Manageme 500 Edged tools. Output maintained. • Production Decreased Establishments 1 Date of Change Previous Hours 60 Number Employees Product Statement of Managemen 250 Machinery. Output decreased. 54 Appendix B Results Reported in Replies to the 1919 Questionnaire REDUCED TO 44 HOURS Production Decreased Establishments 7 Employees . . . . . 3,557 Date of Previous Number Change Hotirs Employees Product Statement of Management *1919 59 191 Machine parts. Output decreased. *1919 59 500 Machinery. Daywork decreased proportionally; piecework nearly maintained. •1918 55 1,968 Steel boats. Daywork decreased 20%; piece- work decreased 10%. •1919 55 100 Machinery. Day and piece work decreased 10%. •1919 52}^ 204 Machine tools. Output decreased in proportion to the decrease in hours. •1919 52^ 56 Machinery. Proportional loss in weekly output. •1919 48 538 Automatic machine Proportional decrease. products. REDUCED TO 44Ji HOURS Production Decreased Establishments Employees . ... Date of Previous Change Employees Product •1918 55 2,400 Ships and machinery. •1919 52i4 442 Machine tools. 2 2,842 Statement of Management Daywork and piecework decreased. Weekly loss in daywork about propor- tional to reduction in hours; no piecework. REDUCED TO 45 HOURS Production Decreased Establishmen- Employees . Cs Date of Change Previous Hours Number Employees Product •1919 60 196 Machinery. 1918 60 260 Hardware. •1919 55Ji 161 Heavy sheet me •1918 55 111 Engines. •1919 55 220 Fire appliances. 12 17,378 Statement of Management Output decreased proportionally. Daywork and piecework decreased proportionally. Weekly output decreased; percent- age not known. Output decreased for both day and piecework. Decreased in same proportion as hours. •These establishments changed hours after November 1, 1918. 55 Date of Change *1919 Previous Hours 55 Number Employees 531 Product Fire appliances, 1918 55 1,066 Machinery. *1919 55 40 Machinery. *1919 55 13,274 Manufactured metals. *1919 *1919 54 50 209 721 Machinery. Machinery. *1919 50 599 Machinery. Statement of Management Daywork decreased in same propor- tion as hours. Piecework de- creased, but less proportionally than hours. Proportional reduction in daywork. No piecework. Cannot say just what the exact per- centage of reduction in production would amount to. Output reduction about proportional to decrease in hours. Output decreased 16?^%. Daywork and piecework decreased 10%. Daywork decreased 5%. Piecework maintained. Date of Change *1918 REDUCED TO 47 HOURS Production Decreased Previous Hours 52}^ Establishments Employees . . Number Employees 95 1 95 Product Statement of Management Machine tools. Daywork less in proportion; no piecework. Date of Change 1917 REDUCED TO i7}4 HOURS Production Increased Establishments Employees 1 378 Previous Hours 52J^ Number Employees Product 378 Edge tools. Statement of Management Slight Increase in daywork piecework. and Production Decreased Date of Change •1918 Previous Hours 52J^ Establishments Employees . . Number Employees 583 1 583 Product Statement of Management Machine tools. Estimated decrease of 15% for day- workers and 5% for pieceworkers. Date of Change ♦1919 Previous Hours 52 REDUCED TO 48 HOURS Production Increased Establishments Employees . . Number Employees 3,241 Product Railway appliances. 1 3,241 Statement of Management Output increased 10%. *These establishments changed hours after November 1, 191S. 66 Production Maintained Establishments .... 5 Employees . .... 2,992 Date of Previous Number Change Hours Employees Product Statement of Management *1918 55 1,985 Magnetos. Output maintained. 1917 54 63 Iron castings. Output maintained. *1918 54 169 Machinery. Daywork maintained; no piecework. 1918 52^ 360 Machine tools. No noticeable effect on output of dayworkers; no piecework. 1918 50 415 Office appliances. Practically maintained output under shorter schedule. 1918 Production Decreased Establishments 34 Employees 29,363 Statement of Management Proportional decrease in output. Decrease of 20% in output; no piece- work. We consider that our work is pro- portional to the hours. Daywork decreased 25%; no piece- work. Daywork decreased 20%; piecework 6%. Daywork and piecework decreased 12.7%: Production cut down in strict prxj- portion to reduction in hours. Output decreased; no piecework. Output decreased 20%. Decrease proportional to decrease in hours. Output decreased for both day and piecework. Decrease proportional to loss of time. Daywork decreased 10%; piecework decreased 5%. Output decreased 11%; no piece- work. Output was maintained on part of work but on greater portion it decreased. Decreased 10%. Output has been reduced but not in same proportion as reduction in hours. No reduction in output was experi- enced except on machines where the output is governed by the speed of the machine. Daywork decreased slightly; nt piecework. *^cu Mtabliihmcnti changed houri after November I, 1918. 67 Date of Change 1918 1918 Previous Hours 60 60 Number Employees 1,044 181 Product Machinery. Machinery. •1918 59 42 Machinery. 1915 56 252 Machinery. ♦1918 55 1,022 Machinery. •1918 •1918 55 55 1,654 6,000 Manufactured metals. Ships. •1918 1918 •1919 56 55 56 265 169 1,621 Iron castings. Machinery. Machinery. •1918 •1919 •1918 54 54 54 433 1,643 117 Plumbers' brass supplies. Machinery. Stoves, •1919 54 70 Machinery. •1919 54 170 Architectural ire •1919 •1918 54 54 160 79 Iron and metal castings. Light castings. •1919 54 4,825 Railway cars. 54 170 Machinery. Date of Change 1918 Previous Hours 54 Number Employees 172 Product Hardware. *1919 54 487 Stoves. 1918 *1919 1918 53M 52J^ 52i4 170 616 826 Machinery. Machinery. Machinery. *1919 52J^ 1,400 Machine parts, 1918 *1919 *1919 52J^ 52}^ 52H 521^ 544 1,318 698 1,108 Power engines. Electrical appliances. , Electrical appliances. Machinery. 1918 51 85 Machinery. *1919 51 103 Stoves. 1918 51 800 Machinery. ♦1918 50 920 Machinery. 1918 50 199 Electric supplies. Statement of Management Daywork decreased 5%; piecework decreased 10%. Daywork and piecework decreased 10%. Less output; no piecework. Output reduced; no piecework. Output decreased for both day and pieceworkers. Output reduced; better quality of production. Proportional decrease. Daywork decreased 4%; piecework decreased 2%. Daywork decreased; piecework maintained. Per cent not known. Output has been correspondingly re- duced, but we hope to make some of it up at least, by further im- proving our methods. Output correspondingly less; no piecework. Output decreased 10%. Curtailed production. Output decreased. Output decreased. REDUCED TO 49}^ HOURS Production Maintained Establishments Employees . . 2 615 Date of Change ♦1919 ♦1918 Previous Hours 55 54^ Number Employees Product 430 Sheet metal products. 185 Automobiles. Statement of Management Output maintained. Practically maintained. Production Decreased Establishments Employees 3 570 Date of Change *1919 1918 1918 Previous Hours 52y2 52 Number Employees 210 60 300 Product Statement of Management Small tools. Daywork decreased 5%; piecework maintained. Automatic machine Daywork decreased 6%; no piece- products, work. Sheet metal Daywork decreased 5%; piecework products. maintained. *These establishments changed hours after November 1, 1918. 58 Date of Change 1916 1917 Previous Hours 59 55 REDUCED TO 50 HOURS Production Increased Establishments 2 Employees 1,570 Number Employees Product 901 Farm machinery. Automobiles. Statement of Management Daywork increased 10%. Piece- work maintained. Weekly output of day and piece- workers increased. Production Maintained Date of Change 1918 Previous Hours 60 1917 57 *1919 55 ♦1919 55 1915 55 1918 55 *1919 55 *1919 *1919 55 55 Establishments , Employees Number Employees Product 60 Stoves. 19 Machinery. 40 Machine accessories. 250 Machines and tools. 10 2,334 386 52 450 675 152 Machine accessories. Utensils. Electrical railway supplies. Automobiles. Toys. 1913 521^ 250 Machinery. Statement of Management Output maintained. Output maintained. Our output was maintained. So far no apparent difference and we hope to maintain our output. Daywork and piecework maintained. Output maintained. Getting same production per week; also intensity of application and co-operation from men, which was part of our agreement. Daywork and piecework maintained. Output practically maintained. Sev- eral up-to-date machines added. Output maintained. Production Decreased Date of Change *1919 1918 Previous Hours 60 59 59 59 •1919 59 •1918 59 55 •1918 55 •1919 55 •1919 55 Establishments Employees . . 25 14,650 Nimiber Employees 78 2,406 387 631 160 300a Product Plumbers' supplies. Agricultural implements. Farm implements. Hardware. 53 Farm tools. 22 Farm tools. 683 Iron castings. Machines. Machinery. Sheet metal products. Statement of Management Output decreased. Output reduced for both day and piecework. Daywork reduced 10%. Piecework not reduced. Dayworkers accomplish proportion- ally less; pieceworkers' output re- mains practically the same. Output decreased 10%. Daywork decreased 10%; piecework maintained. Output per man has averaged a de- crease of approximately 10%. Output decreased. Output decreased approximately 10%; no piecework. Daywork reduced in proportion; output of pieceworkers same. *These establishments changed hours after November 1, 1918. a Change applied to press shop only. 59 Date of Change *1919 Previous Hours 55 Number Employees 210 Product Structural steel. ♦1919 55 200 Machinery. *1919 *1919 55 55 5,811 534 Machinery. Electrical appliance *1919 55 512 Steel castings. *1919 1918 *1919 65 55 55 386 116 278 Machinery. Machinery. Iron castings. ♦1919 1908 *1918 55 55 54 463 120 305 Iron castings. Brass castings. Iron castings. 1918 *1918 1916 54 54 64 462 35 300 Valves and hydra ntSi Heavy sheet metal products. Engines. ♦1918 62J^ 100 Machinery. Statement of M^^jsagement Daywork decreased 10%; no piece- work. Output in our business is practically proportional to hours worked. Output slightly decreased, i. Daywork decreased 10%; piecework maintained. Output decreased for both day and piecework. Output decreased. Decreased 10%; no piecework. Piecework maintained, daywork de- creased. Output decreased. Output cut down in proportion. Piecework decreased 4%; daywork decreased 7.4%. Output decreased proportionally. Daywork decreased; no piecework. Daywork and piecework decreased 9%. Practically the same hourly produc- tion. REDUCED TO 51 HOURS Production Maintained Establishments 1 Employees 800 Date of Previous Number Change Hours Employees Product Statement of Management 1918 64 800 Office machines. No eifect on weekly output. REDUCED TO 52 HOURS Production Decreased Establishments 2 Employees 10,552 Number Employees Product Statsmant of Management 152 Electric motors. Output decreased i}4%. 10,400 Electrical Daywork decreased; no effect on apparatus. piecework. REDUCED TO 52}^ HOURS Production Maintained Establishments 4 Employees 2,217 Date of Previous Number Change Houn Employees Product Statement of Management •1919 65 451 Machinery. Daywork maintained; no piecework, *1919 56 1,214 Machinery. Output maintained for daywork. Ne piecework. 1918 65 16 Electrical Weekly output about the same. appliances. 1918 64 637 Edge tools. Weekly output maintained. *Theie citablishmcnts changed hours after November 1, 1918, 60 Date of Change Previous Hours ♦1919 55 1915 54 Date'of Change 1918 Date of Change *1919 *1918 1918 Previous Hours 58 Previous Hours 59 59 58 REDUCED TO 53 HOURS Production Decreased Establishments Employees Number Employees Product 32 Castings. ... 1 ... 32 Statement of Management Output decreased. REDUCED TO 54 HOURS Production Maintained Establishments Employees . . Number Employees 660 1,400 2,320 260 Product Farm implements. Farm implements. Automobiles. Statement of Management Output maintained. Output maintained and general quality of work better. Output maintained. Date of Change *1919 n919 1917 1919 1917 1917 Previous Hours 60 60 59 58 57 55 Production Decreased Establishments Employees Number Product Iron castings. Light iron castings Iron and brass castings. Automobiles. 6 1,743 Employees 210 85 26 Statement of Management Output decreased 10%. Output decreased proportionally. Decrease in output; no piecework. 1,094 Automobiles. Both day and piece work decreased 7%. 105 Heating apparatus. Daywork decreased proportionally; piecework maintained. 223 Sheet metal Amount of tonnage turned out was products. slightly less. REDUCED TO 55 HOURS Production Maintained Establishments . Employees 2 159 Date of Change Previous Hours Number Employees Product Statement of Management *1919 60 144 Electrical motors. Output maintained. 1914 60 15 Machines. Weekly output about the same. Date of Change 1918 Previous Hours 60 Production Decreased Establishments 6 Employees 3,605 Number Employees Product Statement of Management 15 Plumbers' supplies. Output reduced just in proportion to reduction in hours. 1918 *1919 59 59 289 332 Iron castings. Agricultural implements. Output decreased proportionally in the machine shop. Maintained in the foundry. Output decreased, but not in propor- tion to decrease in working hours. ♦These establishments changed hours after November 1, 1918. 61 Date of Previous Number Change Hours Employees Product Statement of Management 59 55 Machine tools. Daywork decreased 7%; piecework maintained. *1919 59 200 Machine tools. Output decreased proportionally. *1918 57J^ 2,614 Railway cars. Our production has decreased, partly because of the shorter day. REDUCED TO 57}^ HOURS Production Maintained Establishments Employees . . . . Date of Change *1919 Previous Hours 60 Number Employees Product 65 Iron and steel wire. 1 65 Statement of Management Former schedule too long a period without rest, resulting in greatly decreased efficiency. Weekly out- put maintained under shorter schedule. *These establishments changed hours after November 1, 1918. 62 Publications of the National Industrial Conference Board 15 Beacon Street, Boston, Mass. Research Report No. 1. Workmen's Compensation Acts in the United States — The Legal Phase. April, 1917. Revised April, 1919. Research Report No. S. Analysis of British Wartime Reports on Hours OF Work as Related to Output and Fatigue. November, 1917. Research Report No. 3. Strikes in American Industry in Wartime. March, 1918. Research Report No. 4- Hours of Work as Related to Output and Health of Workers — Cotton Manufacturing. March, 1918. Research Report No. B. The Canadian Industrial Disputes Investiga- tion Act. April, 1918. Research Report No. 6. Sickness Insurance or Sickness Prevention? May, 1918. Research Report No. 7. Hours of Work as Related to Output and Health of Workers — Boot and Shoe Industry. June, 1918. Research Report No. 8. Wartime Employment of Women in the Metal Trades. July, 1918. Research Report No. 9. Wartime Changes in the Cost of Living — July, 1914, TO June, 1918. August, 1918. Research Report No. 10. Arbitration and Wage-Fixing in Australia. October, 1918. Research Report No. 11. The EiGht-Hour Day Defined. December, 1918. Research Report No. IS. Hours of Work as Related to Output and Health of Workers — Wool Manufacturing. December, 1918. Research Report No. 13. Rest Periods for Industrial Workers. January, 1919. Research Report No. I4. Wartime Changes in the Cost of Living — July, 1914, TO November, 1918. February, 1919. Research Report No. IS. Problems of Industrial Readjustment in the United States. February, 1919. Research Report No. 16. Hours of Work as Related to Output and Health of Workers — Silk Manufacturing. March, 1919. Research Report No. 17. Wartime Changes in the Cost of Living — ^July, 1914, to March, 1919. May, 1919. Research Report No. 18. Hours of Work as Related to Output and Health of Workers — Metal Industries. July, 1919. INDUSTRIAL NEWS SURVEY Important industrial news in concise form. Weekly. $2.00 per year. Cornell University Library H0 5119.M5N3 Hours of work as related to output and h 3 1924 001 366 479