UNIVERSITY OF CALIFORNIA AT LOS ANGELES The Movement of Wages In The Cotton Manufacturing Industry of New England Since i860 STANLEY E. HOWARD 8 46 t A DISSERTATION PRESENTED TO THE KACULTV OF PRINCETON l'NI\'ERSITV IN CANDIDACY lOR THE DECREE OF DOCTOR OK PIIILOSOPHV The Movement of Wages In the Cotton Manufacturing Industry of New England Since 1860 By STANLEY E. HOWARD Assistant Professor of Economics in Princeton University BOSTON NATIONAL COUNCIL OF AMERICAN COTTON MANUFACTURERS Boston, Mass. WAsiiiNcnoN, D. C. Chariotie, N. C. 1920 COPYRIGHT, 1920 By Stanley E. Howard Printed by E. L. Barry Waltham, Mass. u 'a- "O- *.« (V r CONTENTS Chapter Page I. The Scope and Method of this Investigation, ..... 5 II. Changes of Rates of Wages at Fall River, 13 Table I. Wages, prices ami margins at Fall River, 1875-1918, . . 17 Chart I. Weaving rates and margins at Fall River, 1875-1918, . 18 Table II. Hypothetical margins and wea\nng rates to illustrate the sliding scale agreement of October, 1905, ... 22 Table III. Actual margins and rates under the sliding scale agree- ment of October, 1905, 23 Table I\'. Schedule of margins and weaving rates pn^vided by the sliding scale agreement of May 14, 1907, 25 Table V. Schedule of margins and weaving rates pro\nded by the revised sliding scale agreement of Septembers, 1908, 27 Table \T. .\ctual margins and wages under the sliding scale \ agreement of May 14, 1907, (revised September 8, 1908,) 29 Tabic \'II. The correlation of wages and prices at Fall River, 1881-1904, 30 III. The Movement of Full Time Framings in the Massachu- setts Cotton Manufacturing Industry, 1860-1891, . . 41 Tabic \'III. Simple medians of relative full time earnings in the Massachusetts cotton industry, i860- 1880, .... 44 Chart II. The movement of full time earnings in the Massachusetts cotton industry, 1860-1880, as shown by the method of simple medians, 45 Table I\. The movements of full time earnings in Massachusetts cotton mills, and of prices, 1860-1880, 46 Chart III. I'he movements of full time earnings in Massachusetts cotton mills, and of prices, i86<')-l88o, 47 Chart I\". The movement of full time earnings in the Massachusetts^ cotton industry, 1860-1880; by sex groups, .... 48 Table X. Relative full time earnings in the Massachusetts cotton industry in the year 1880, 58 Table XI. The movement of full time earnings in the Massachusetts cotton industry, 1880- 1 89 1, 59 1 89555 4 WAC.ES IN COTTON MANUFACTURES Chapter Page Chart \'. The movement of full time earnings in the Massachusetts cotton industry, 1 880- 1 891, 60 IV. 'I'hc Movement of Full Time Earnings in the Cotton Manufacturing Industry of Massachusetts, Maine and Rhode Island, 1890-1916, 62 Table XII. The movement of earnings in the Massachusetts cotton industry, 1890- 1916, 66-67 Table XIII. The movement of earnings in the Maine cotton industry, 1890-1916, 68-69 Table XIV. The movement of earnings in the Rhode Island cotton industry, 1890-1916, 70-71 Chart VI. The movement of full time earnings in the Massachu- setts cotton industry, 1890-1916, 72 Chart VII. The movement of full time earnings in the Maine cotton industry, 1890- 1916, 73 Chart VIII. The movement of full time earnings in the Rhode Island cotton industry, 1890- 1916, 74 Table XV. The percentage distribution of employees in the Mas- sachusetts cotton industry on the basis of weekly earnings, 1889-1916, 80 Chart IX. The distribution of employees in the Massachusetts cotton industry on the basis of weekly earnings, 1889- 1916, 81 V. Fluctuations of Employment, 83 Table XVI. Fluctuations of employment in the Massachusetts cotton industry, 1889-1916, 84 Chart X. Fluctuations of Employment in the Massachusetts cotton industry, 1886- 1916, 86 Table XVII. Unemployment in the Massachusetts cotton industry, 1886- 1916. The deviation from the monthly average number of employees, 85 Table XVIII. Unemployment in the Massachusetts cotton industry 1886-1916. The operating time of the mills, ... 89 VI. Conclusions, 91 Table XIX. Wages in the Massachusetts cotton manufacturing industry and general prices, i860- 191 8, 93 Chart XI. Wages in the Massachusetts cotton manufacturing industry and general prices, i860- 191 8, 94 SCOPE AND METFiOD CHAPTER I The Scope and Method of this Investigation At a time like the present when there is at hand so much unavoidable and important work of economic and social read- justment, it seems particularly needful that there be on the part of the public both ability and willingness to view the problems that arise in their true perspectives. While in dealing with all economic problems a thoroughgoing knowledge of historical backgrounds is much to be desired, the importance of such knowledge is especially great when forces of great dynamic power have been and still are at work. We are in a period of upheaval in the industrial world. Under the spur of urgent war needs, production attained in terms both of monetary and of technical units unprecedented levels. Labor has been at a premium unprecedented in our industrial history. Money wages have in many industries been very markedly advanced, and that industry is rare in which there has been no advance at all. These wage changes can no more be thought of as standing by themselves than can the diplomatic and military events of 1914-1919. The present monograph is therefore offered in the hope that it may in some measure contribute to public knowl- edge of historical backgrounds in a particular portion of the field of industrial problems. There has been chosen as the subject of investigation the changes of wages in the New England cotton manufacturing industry since i860. Thus there is covered a period of nearly sixty years, embracing two important war periods, those of 1861-1865 and 1914-1919, and both the period of declining price movement preceding 1897 and that of rising prices since that year. As to form and content the present study is largely statistical. Two general problems have presented themselves : First, the 6 WAGES IN COTTON MANUFACTURES determination of the detailed facts regarding changes of wages; second, the discovery of the more important causal forces pro- ducing these changes. Of these two problems, the former has been stressed, and only in the discussion of the Fall River sliding scale experiments (in Chapter II) and in the analysis and explanation of the results of Professor Mitchell's work covering the greenback period (Chapter III) has there been any considerable attention given to descriptive and explanatory materials gathered from miscellaneous sources and serving to explain the evidence of the statistical exhibits. In Chapter II the analysis of market conditions in their relation to changes of rates of wages approaches more nearly than does any other part of the study the presentation of critical comment. As to the miscellaneous explanatory materials of Chapter III, their chief justification as a part of the present work lies in the light they throw upon the developments of the early years of the period covered — rather important developments in that not only was there in them a radical departure from the pre-war conditions of i860 so far as concerned cotton mill employees as a group, but also in that in these early years there was established a new relationship of the wage levels of the sexes. Of the purely statistical part of the subject there are three main divisions. First, there is that which relates to changes of rates of wages rather than to changes of earnings. This topic is treated particularly in Chapter II. Attention has been given to the history of the weaving piece rate at Fall River, Massa- chusetts, as being at once the best available and a fairly accurate criterion of changes of rates at least for the southern New England section of the cotton manufacturing industry. Second, there is that division which relates to changes in earnings, and more particularly to changes of full time earnings. It is obvious that in an industry in which the principle of the division of labor has attained a high degree of development, a development which has been accompanied by considerable resort to piece rate methods of payment, changes of rates may or may not result in proportionate changes of earnings, due to the complicating influence of changes in the technical conditions SCOPE AND METHOD 7 of production. And again, even if piece rates and hourly earn- ings do not change, weekly and annual relative earnings may be modified by the enactment of legislation limiting the hours of labor of certain groups of workers, as women and minors — a restriction upon industrial activity which from obvious considera- tions relating to skillful factory organization and management atifects all employees regardless of sex and age. The assembling and analysis of the facts relating to the movement of full time money earnings are presented in Chapters III and IV, the former showing for the period 1860-1891 the conclusions based chiefly, directly or indirectly, upon the materials of the Aldrich Report of 1893;^ the latter, covering the years 1890- 191 6, being based upon the work of the United States Bureau of Labor Statistics and of its predecessor the United States Bureau of Labor, and upon the work of the Massachusetts Bureau of Statistics. Finally, as is necessary in any statistical examination of wage movements, consideration has been given to the occurrence, regularly or irregularly, of conditions of unemployment. Such conditions frequently modify in a substantial degree the showing made by statistics of relative full time earnings. This topic has been treated in Chapter V. In Chapter VI there is presented a summary of the statistical analysis just outlined. The curves of relative full time earnings in the Massachusetts section of the cotton manufacturing industry for 1860-1891 and 1890-1916 are united into a single curve for the entire period, using the year i860 as the base. Second, comparison is made of the curve of relative full time earnings thus constructed with a similar curve of relative wholesale prices of commodities in general, to reveal the movement of real, in addition to and in comparison with that of money, wages. Finally, comparison is made between the movements of piece rates and of full time earnings, to secure some sugges- tion as to the effects of technical developments upon wage con- ditions. It should be clearly understood with respect to the last mentioned topic that there is here presented only a sugges- I. Senate Committee on Kinance, Kt-port on WholesaU Prices, on Wa^cs, onJ on Transportation. Senate Reports, 521! Congress, 2d Session. Report 1394, pts. 1-4. 8 WAGES IN COTTON MANUFACTURES tion regarding possible interpretaions of the statistical evidence available. It is to be hoped that this topic may sometime be investigated by someone who not only has ready and constant access to the necessary sources of information, that is, to the strictly private records of the cotton mills, but is also thoroughly familiar by reason of long experience with the complicated technique of the industry. There has been no attempt made here to study, with a thoroughness exceeding that already applied by students of economic and industrial history, the history of the labor supply in the New England cotton industry. Obviously the supply has two fundamental aspects, quantity and quality. As regards the former, the compilation and analysis of population and occupation statistics could add little of value for present pur- poses, since, as we have seen in the recent war period, within limits of very general principles of social and industrial stratifi- cation there is no such thing as an arbitrary or permanent division of the industrial population into non-interchangeable classes, and munitions factories have drawn heavily on textile mills for a labor supply. Quantitative changes of population, from a positive point of view, have not been the chief factor in indus- trial adjustments, and to compile elaborate figures showing such changes would add little to the value of this study. In general, this quantitative change of labor supply has been a growth from about 122,000 cotton mill employees in the United States in i860, to 378,000 in the United States in 1910, of whom 189,700 were in New England. Within the New England group of states Massachusetts has always been in the lead, showing a growth of from 38,500 in i860 to 108,900 in 1910. These are United States Census figures. More detailed figures for Massachusetts are presented in Chapter V. Following Massachusetts in the order of numerical importance, the other New England states are Rhode Island, New Hampshire, Maine and Connecticut. The development of the industry in Vermont is so small as to be almost negligible. The second aspect of the labor supply is the qualitative one. In this connection it is assumed that the reader is familiar with SCOPE AND METHOD 9 the fundamental facts: First, tlie increasing resort by mann- facturers, willingly or unwillingly, to individuals representing the successive waves of luiropean immigration into the United States; second, the fact that the above mentioned change has been combined with the constant improvement of machinery and processes to make relatively more economical the employ- ment of certain immigrant males in place of former female employees, particularly in those departments in which the physical severity of the work to be performed has increased as machinery has increased in size and weight; and third, the fact of relatively decreasing employment of children due chief!)' to restrictive legislation by the states, in which matter Massachu- setts has consistently been in the lead of the other New England states. If the reader is unfamiliar with these facts let him consult, for information relating to the cotton manufacturing industr)', Professor M. T. Copeland's volume descriptive of the American cotton manufacturing industry.^ Another point of omission in this study is the detailed con- sideration of problems connected with the technique of the textile manufacturing processes. Most readers will be suffic- iently familiar with the fundamental textile processes to follow intelligently the work of statistical compilation and anah'sis in the succeeding chapters. To any who wish more detailed information than is here assumed, there is recommended the reading of Chapters IV and V of Professor Copeland's book above referred to, or of the descriptive materials of Bulletin 239 of the Bureau of Labor Statistics.- So far as we are concerned in this study with the technicalities of the occupational division of labor, the fact of chief interest relates to the influence of manufacturing technique upon the conditions of collective bargaining and upon the significance of statistical data covering the matter of piece rates. It is a fundamental prerequisite of successful collective bar- gaining concerning wages that there be some considerable 1. Copeland, M. T., The Cotton Manuftiituring Industry of the Vnitdi Stiitis. 2. United States Bureau of Lalxir Statistics, Hulletin 239. H'ti^a attil Hours of Labor in Cotton Gooih Manufacturing and Finishini^, 1916. I'p. 140-205. 10 WAGES IN COTTON MANUFACTURES degree of standardization of working conditions or of working rules. And it is a regrettable weakness in the American cotton manufacturing industry that no such standardization has as yet been attained. In the British cotton manufacturing industry there has gradually been developed a set of standardized wage scales, that is scales of compensation in the piece working crafts which adjust themselves automatically to changed condi- tions of production or to changes of products. For this reason most" English weavers are protected somewhat against the risk of accidental or intentional reductions of earnings caused by incorrect readjustment of piece rates as the width of the cloth, the fineness of the warp and weft threads, the number of threads per square inch, and the speed of operation of the machine are changed. Notwithstanding intermittent agitation for the construction of similar standard lists in the United States, par- ticularly in New Bedford, the leading center of the American fine goods industry, where such standardization is most needed, it has not been possible to secure agreement thereon. In fact, opposition of employers has gone so far that it has not been possible to secure for the Massachusetts laws requiring the attachment to all weaving work of cards of detailed specifica- tions an amendment to the effect that these specifications be not written or printed in a code necessarily unintelligible to the workers. Under such conditions collective bargaining whether by individual unions or by groups of unions can not attain a very high degree of development. The nearest approach to a real standard list of piece rates in the New Eng- land cotton industry is the mule spinners' list at Fall River, agreed to by the union of mule spinners and the local manu- facturers in 1889. From the standpoint of the statistician the absence of highly standardized conditions in the matter just discussed is of prime interest in that it partially explains the lack of close correlation between the movement of piece rates and that of full time earnings. Regarding union organization of cotton mill employees something is said in Chapter II in connection with the sliding SCOPE AND METHOD 1 1 scale experiments at VM River. In general, the New England textile industr)' is one in which craft organization has never been very strong, a fact due largely to the low degree of skill required of the majority of the workers and to the presence of large groups of immigrants of many nationalities. In recent years there has been developing considerable antagonism between the old craft unions and the industrial organizations which lean toward the philosophy and the policy of the Indus- trial Workers of the World. The character of the work done in the preparation of the succeeding chapters and the manner of presentation of results render superfluous the publication of a formal bibliography. For the purely statistical work the chief sources have been the so-called Weeks Report of the Tenth Census,^ the Aldrich Report of 1893, Professor Wesley C. Mitchell's excellent volume on Go/d, Prices, ami Wages under the Greenback Standard, the Annual Reports and Bulletins of the United States Bureau of Labor and of its successor, the Bureau of Labor Statistics, and the reports of the Massachusetts Bureau of Statistics, particularly the Reports on the Statistics of Manufactures. Throughout the following pages care has been taken to give specific references to these sources and to explain as clearly as possible the modi- fications and rearrangements of materials made for the purpose in hand. Frequent specific references are made in the text and in foot- notes to contemporary accounts in commercial and financial journals descriptive of industrial conditions, and there are also frequent references to books and pamphlets. It is needless to present a general bibliography of useful materials in view of the existence in print of a very complete and well arranged biblio- graphy covering the whole subject of the manufacture of cotton goods. ^ There is also an excellent select bibliograph)' in the 1. Weeks, Josejih D., Report on iJ'e Statistics of liases in J/nnu/.iitnrin^' InJustries. Tenth Census of the United States, Vol. XX. The cotton manufacturing industry is covered in pp. 327-367. 2. Woodbur)', C. J. H., Bibliography of the Cotton Manufacture. 2 vols. (1909 and 1910). Published by The National Association of Cotton Manufacturers of which Mr. Woodbury was then secretary. 12 WAGES IN COTTON MANUFACTURES appendix of l^rofessor Copeland's book. The reader may be interested, for purposes of comparison, in a study somewhat similar to the present, but covering the British section of the cotton manufacturing industry, by Mr. G. H. Wood of the Royal Statistical Society. ^ To make specific reference to all who have given assistance in the preparation of this monograph would be impracticable. The author wishes, however, to acknowledge his indebtedness to Professors F. A. Fetter and E. W. Kemmerer of the Depart- ment of Economics and Social Institutions of Princeton Univer- sity and to Mr. W. M. Adriance, formerly of that department, for advice and assistance in the preparation of the manuscript, and to Mr, A. F, Bemis of the National Council of American Cotton Manufacturers, for co-operation in securing publication at this time. I. Wood, George Henr)', The Statistics of Wages in the United Kingdom dur- ing the Nineteenth Century. Parts XV-XIX. 7^he Cotton Industry. Journal of the Royal Statistical Society, 1910, pp. 39-58, 128-163, 283-315, 411-434, 585-626. RATES OF WAGES AT FALL RIVER 13 CHAPTER II Changes of Rates of Wages at Fall River ' In an industry like that of cotton cloth manufacturing, organ- ized to a large degree upon the principle of the division of labor, and in which there is a great diversity of products and of technical conditions of operation from town to town, from mill to mill, and from one mill department to the next, the study of the movement of rates of wages is at best difficult. In the American cotton manufacturing industry changes in technical processes, equipment and products have been rapid, the opera- tive force has been continually changing as regards its racial composition, union organization and arrangements for collective bargaining, except in special cases, have been weak and some- what temporary, and there has consequently not evolved any very considerable standardization of rates of wages. The result is that materials for the study of movements of rates of wages are limited. However, at F"all River, Massachusetts, the leading cotton manufacturing city of the United States, collective bargaining has attained some strength, the representatives of the chief craft organizations^ being organized into a Textile Council, which deals on behalf of all the operatives, so far as occupa- tional classification is concerned, with the Cotton Manufacturers' Association of the city.'^ The latter organization represents and 1. This chapter is based upon an article by the author under the title The Fall River Slidini; Scale F.xperinuttt of iqoji-igio, appearing in the American Economic Review, Vol. VII, No. 3, pp. 530-552 (September, 1917). The materials there used have been revised and extended. Due acknowledgment is made to the editors of the American Economic Review for permission to use them in the present larger work. 2. The unions of mule spinners, weavers, slasher tenders, carders and lo 1. U c b UI vt u c« M 3 New weavini rates in force 4 Index numbe ofweavingrat Base (100) =21.00 cents u 10 c > 1 6 'Index numbe jofweavinerat Base (100) =21.00 cents Middling upland cotIo; j per pound. 8 28 inch print cloth per yard 38% inch print cloth per yard. 10 Margins. 11 Index numbe of margins Base (100) =108.52 cents (Cents) (Cents) (Cents) (Centb) (Cents) (Cents) 187s 21.00 100.00 15.46 5-33 8.71 140.43 129.40 1876 19.00 90.48 12.98 4.10 7.06 105.28 97.00 1877 11.82 4.38 6.77 116.05 106.93 1878 18.00 85.72 11.22 3-44 6.09 88.45 81.50 1879 10.84 3-93 6.25 105.17 96.90 1880 21.00 100.00 II. 51 4.51 7.41 132-07 1 2 1 . 70 I88I 100.00 21.00 100.00 12.03 3-95 7.00 108.52 100.00 1882 21.00 100.00 I 1.56 3.76 6.50 99-73 91.90 1883 21.00 100.00 11.88 3.60 6.00 85.79 79.06 1884 Feb. 4 18.50 88.10 18.71 89.10 10.88 3-36 6.00 87.89 80.99 1885 Jan. 19 16.50 78.58 16.78 79.91 10.45 3.12 6.00 85-93 79.18 1886 Mar. I 18.15 86.44 17.89 85.20 9.28 3-31 6.00 97-57 89.90 1887 18.15 86.44 10.21 3-33 6.00 92.58 85-31 1888 Feb. 13 19.00 90.48 18.90 90.00 10.03 3-8i 6-50 113.09 104.20 1889 19.00 90.48 10.65 3-8i 6.50 108.14 99.64 1890 19.00 90.48 11.07 3-34 6.00 85.92 79-17 I89I 19.00 90.48 8.60 2-95 6.00 96.90 89.29 1892 July 11 Dec. 5 19.60 21.00 93-34 100.00 19-39 92-34 7.71 3-39 6.25 118.10 108.83 '893 Sept. u 18.00 85-72 20.03 95-38 8.56 3-30 5-25 92.69 85.41 1894 Aug. 30 16.00 76.20 17-34 82.58 6-94 2.75 4.90 102.47 94-42 1895 Apr. 22 18.00 85.72 17-38 82.76 7-44 2.86 5-25 96.74 89.14 1896 18.00 85.72 7-93 2.60 4.66 74-21 68.39 1897 18.00 85-72 7.00 2.48 4.70 77.61 71-52 1898 Jan. 1 16.00 76.20 16.00 76.20 5-94 2.06 3-96 64.90 59.80 1899 Feb. 27 Dec. II 18.00 19.80 85.72 94.28 17-79 84-72 6.88 2.69 4.25 70.84 65.28 1900 19.80 94.28 9-25 3.21 5.00 81.00 74-64 I9OI 19.80 94.28 8.75 2.84 4.62 70.38 64.86 1902 Mar. 17 21.78 103.72 21.36 101.72 9.00 311 5.00 80.75 74.41 1903 Nov. 23 19.80 94.28 21.55 102.63 II. 18 3-25 5.00 66.46 61.24 1904 July 25 17-32 82.48 1S.55 88.34 11-75 3-44 5.00 66.18 60.98 1905 Oct. 30 1S.63 88.62 17-54 •83-52 9.80 3-13 4-75 70.66 65.12 1906 July 2 Nov. 26 19.80 21.78 94.28 103.72 19-39 92.34 11.50 3-63 5.12 74-44 68.60 1907 May 27 23.96 114. II 23.08 109.92 12.10 4.62 6.00 106.48 98.12 1908 May 26 19.66 93.62 21-39 101.86 10.62 3-5° 5-37 82.69 76.20 1909 19.66 93.62 12.68 367 5.06 64.90 59-80 I9IO 19.66 93.62 15. II 3-87 5.62 59.24 54-59 I9II 19.66 93.62 13.01 3-54 5.24 62.32 57-42 1912 Mar. 25 21.62 102.96 21.16 100.76 1 1.52 3-82 5.22 80.21 73-9' I9I3 21.62 102.96 12.80 3.86 5.40 73-85 68.05 1914 21.62 102.96 II. 13 3-48 5-'9 75-18 69-27 I9I5 21.62 102.96 10.14 3-27 5-«9 78-38 72.22 I9I6 Jan. 24 May I Dec. 4 22.71 24.98 27.48 108.15 118.97 130.86 24.36 116.00 14.45 4.60 7.20 107.10 98.68 I9I7 June 4 30-23 143-96 29.36 139.82 23-49 7.18 10.79 152.26 140.28 Dec. 3 34.02 162.00 1918 June 3 39-12 186.30 36.96 176.00 18 WAGES IN COTTON MANUFACTURES On ON > P^ en < .a s 3 C C S ^ C I I I I I I I I I I t I I I I 1 ri I I I I I I I I I I >»» 1 1 I I I I I M M ,cui o V O "» o ^-inch 64 x 64 prints) were to be considered. It was tacitly agreed that the sliding scale should be given a fair trial for one year.^ The following schedule of hypothetical margins and weaving rates under the agreement of October, 1905, may help the reader to visualize the plan of wage determination and to make comparison with later schedules. TABLE II Hypothetical Margins and Weaving Rates to Illustrate the Sliding Scale Agreement of October, 1905 Margins. Weaving Rates. Margins. Weaving Rates. (Cents) (Cents) (Cents) (Cents) "5 22.95 85 20.25 1 10 22.50 80 19-35 105 22.05 75 18.45 95 21.15 72.5 18.00 This first formal sliding scale experiment was of short dura- tion. For this there were two chief reasons. First, under this agreement, the average weaving rate in force was only 18.63 cents per cut of regular print cloth ; which constituted an average "wage premium" of only 3.5 per cent., or only about one-half of the desired restoration of wages to the level prevail- ing between November, 1903 and July, 1904. Second, the rates of wages paid fluctuated so violently from week to week (see Table III) that the operatives became much dissatisfied and I . Some sources for the terms of this agreement are : Chronicle, November 4, 1905, p, 1385; Lincoln, loc. cit., p. 456; Massachusetts Labor Bulletin, No. 38, p. 342, No. 41, pp. 192-196; Textile World Record, vol. XXX, pp. 361-363. RATES OF WAGES AT FALL RIVER 23 were led to question the validity of the theoretical foundation of the experiment.^ TABLE III •-' Actual Margins and Rates under the Sliding Scale Agreement of October, 1905 •0 •0 •0 c c vB bos be > i V •0 c c "bi Ms be 5 «>■ V ^1 V S5 is" 1905 (Cents) (Per Cent.) (CenU) 1906 (Cents) (Per Cent.) (CenU) Oct. 27 77.00 5 18.90 Mar. 3 82.85 10 19.80 Nov. 3 75-70 3 18.54 " 10 80.19 8 19.44 " 10 71.40 18.00 " 17 81.27 9 19.62 " 17 74.00 2 18.36 " 24 77.82 5 18.90 «« 24 73-50 I 18.18 " 31 75-42 3 18.54 Dec. I 75.20 3 18.54 Apr. 7 76.15 4 18.72 " 8 72.40 18.00 " 14 7";. 02 3 18.54 " 15 75.02 3 18.54 " 21 74.82 2 18.36 « 22 73-76 I 18.18 " 28 75-55 3 18.54 " 29 75-37 3 18.54 May 5 72.85 18.00 1906 " 12 71.40 18.00 Jan. S 76.41 4 18.72 " 19 70.68 18.00 " 12 78.09 6 19.08 " 26 71.14 18.00 «« 19 75-23 3 18.54 June 2 73.28 18.00 '« 26 75-56 3 18.54 " 9 75-56 3 1S.54 Feb. 2 79-49 7 19.26 " 16 76.67 4 18.72 " 9 81.16 9 19.62 " 23 71.17 18.00 " 16 81.45 9 19.62 " 30 71.16 18.00 " 23 82.75 10 19.80 Dissatisfaction caused by these conditions was intensified when, in May, 1906, voluntary wage advances of 10 per cent, were made by certain corporations in Massachusetts and Con- necticut engaged in the manufacture of fine goods. '^ Although the employers in Fall River were insistent that " increases in fine goods mills, where margins of profits were large, should have no bearing on the general situation in Fall River, "^ the 1. P'or an elaboration of this point see Lincoln, loc. cit., pp. 456-459. 2. This table is constructed on the basis of the table presented by Mr. Lincoln on page 457of his article. Computations for June, 1906 have been added; also weav- ing rates for the entire period. 3. Chronicle, June 23, 1906, pp. 1405, 1406. 4. Ibid., September 8, 1906, p. 531. 24 WAGES IN COTTON MANUFACTURES Textile Council, voicing the feeling of the operatives with regard to the sliding scale, strongly urged the abandonment of the experiment and the restoration of the rates of wages prevailing before July, 1904. This request was made on June 8. On the sixteenth of the same month the Manufacturers' Association made a counter proposal, offering to guarantee a wage premium (on the basis of the sliding scale) of 5 per cent, from June 12 to October i "with a further guarantee that if the margin in any one week exceeded this 5 percent, increase they would pay the additional amount."^ When this plan was rejected the manu- facturers yielded, promising to restore the rates desired {i.e., rates based on a weaving rate of 19.80 cents), at the same time protesting that the margin of profit warranted no such advance and claiming that the advance was made only in the interest of the general welfare of the city.- The new schedule became effective July 2, 1906. The weaving rate of 19.80 cents per cut thus established remained in force only a few months. A widening margin of profit and a demand for goods so large that, in view of a reported scarcity of labor, it could not be satisfied,'^ soon led to renewed demands for wage advances. This time the request was for a restoration of the rates prevailing before the reduction of 1903. After some parleying, a counter proposal by the employers of a compromise of 5 per cent, increase (with the promise of 5 per cent, more in February, 1907, if conditions should be favorable),"* and a threat of strike on the part of the employees, the manufacturers yielded, and a rate of 21.78 cents was announced to take effect November 26. The margin at this time was about 95 cents.' And it was a part of the settle- ment then made that the wage determined upon should be 1. Chronicle, June 23, 1906, p. 1406. 2. Ibid., p. 1448. Following the action of the Manufacturers' Association, the Fall River Iron Works increased the wages of its operatives 10 per cent., " giving them that advantage over the help in other mills of the city." Chronicle, September 8, 1906, page 531. 3. Ibid., January' 5, 1907, p. 60. Mr. Lincoln in his article (p. 459) says that the prosperity of the industrj- at this time was " greater than ever before known." 4. Chronicle, November 24, 1 906, p. 1 302. 5. Ibid., May 11, 1907, p. 1129. RATES OF WACJES AT FALL RIVER 25 effective for six months, at the end of which period there should be a readjustment with reference to the then existing margin. 1 Accordingly, near the end of April, 1907, discussion of the wage problem was reopened. The margin of profit was found to have widened quite substantially, and the advantages which might have accrued to the operatives had they not insisted upon abandonment of the sliding scale appeared great. The demand of the operatives was for a 10 per cent, advance. This the employers were disposed to refuse, but finally the advance was granted with a sliding scale agreement attached. In brief the provisions of this agreement were as follows i^ Eighteen cents per cut was established as the minimum weaving rate for 28-inch prints at a base margin of 72.5 cents. At a margin of 75 cents the weaving rate should be 18.68 cents, and above the margin of 75 cents rates were graded in accord with the margin changes indicated in the following tabulation until a maximum margin of 1 15 cents was reached at which the maximum weaving rate should be 23.96 cents. The scale of progression was slightly steeper than that provided in the agreement of 1905. TABLE IV Schedule of Margins and Weaving Rates Provided by the Sliding Scale Agreement of May 14, 1907 Margins, Weaving Rates. Margins. Weaving Rates. (Cents) I'S no 105 95 (Cents) 23.96 23-42 22.87 21.78 (Cents) 85 80 75 72.5 (Cents) 20.69 19.66 18.68 18.00 Second, the agreement provided that rates of wages should 1. Ibid., .Septemlier 7, 1907, p. 564. 2. The full text of this agreement is presented in the Appendix. 26 WAGES IN COTTON MANUFACTURES change only once in six months, calculations being made on the last Monday of May and November of each year, the rate to be effective in each six-month period being determined by the average margin of the preceding six months. And, quoting from section 2 of the agreement, " there shall be no change in prices on either the ascending or descending scale unless the margin has reached a point indicated in the . . . -. schedule." The new agreement went into force with a weaving rate of 23.96 cents, the highest rate allowed under the terms of the bargain, and also the highest rate which had up to that time been paid in the Falls River mills. But the extremely high margins of 1907 were of short duration, and 1908, 1909 and 1 910 witnessed successive downward movements. Not only was this true, but, in 1907, difficulties in the financial world found their way into the cotton industry, with the result that curtailment of production was considered and well under way before the end of the year.^ Corporations supposed to be filling large orders were unable to secure ready money for use in purchasing materials and in meeting payroll demands, as jobbers were embarrassed by very high interest rates in making payments on deliveries of goods. ^ Through the spring of 1908 curtailment in the form of a short working week was still general.'^ At points throughout New England other than Fall River reductions in wages were the order of the day. But at Fall River the sliding scale agreement for a time saved the operatives from this form of hardship. When, near the end of May, calculations of margins and of wages were made, they showed that the financial distresses had been accompanied by a marked decline of the margin (which in April and May had been between 50 and 60 cents) and that, in accordance with the agreement, the weaving rate must drop from 23.96 cents per 1 . For interesting information regarding the method adopted by the mill men throughout New England to secure uniformity of curtailmeut see the Chronicle of November 16, 1907, p. 1237; November 23, 1907, p. 1301; December 28, 1907, p. 1653. 2. Chronicle, November 23, 1907, p. 1301. 3. Ibid., March 14, 1908, p. 631; March 28, 1908, pp. 757, 758; May 23, 1908, p. 1293; May 30, 1908. p. 1308. RATES OF WAGES AT FALL RIVER 27 cut to 19.66 cents — a reduction of 17.95 P^"" cent. The spirit in which this, the largest single reduction of wages experienced by Fall River workers was accepted, deserves only the highest commendation. • The operatives as a whole were as yet un- willing to declare the experiment a failure, although the mule spinners led an attempt to have the agreement abandoned through the required process of three months' notice. Instead of abandonment, revision was resorted to in September, when the scale of margins and wages was graduated somewhat more nicely, the intervals between the margins named in the schedule being made uniform at two and one-half cents as shown in Table V. Another slight change was the provision for a delay of two weeks in the enforcement of new rates of wages at the end of each six-month period, ^ TABLE V Schedule of Margins and Weaving Rates Provided by the Revised Sliding Scale Agreement of September 8, 1908 Margins. Weaving Rates. Margins. Weaving Rates. (Cents) (Cents) (Cents) (Cents) JI5 23.96 92.5 21.50 112. 5 23.69 90 21.23 no 23.42 87.5 20.96 107.5 23->4 85 20.69 105 22.87 82.5 20. 1 S 102.5 22.59 80 19.66 100 22.32 77-5 19.17 97-5 22.05 75 18.68 95 21.78 72.5 18.00 In November, 1908, calculation of the margin showed another marked decline, according to which the rates of wages, under the automatic operation of the agreement, were to be expected 1 . Mr. Lincoln on pages 465, 466, of his article gives in full the statement issueil at the time by Mr. James Tansey, then president of the Textile Council. See also Massachusetts Labor Bulletin, No. 60, pp. 263-266. 2. The whole amended agreement is found on pages 16-18 of the handbook of the Carders' Union. 28 WAGES IN COTTON MANUFACTURES to drop to the minimum of i8 cents. Instead, the manufacturers waived their rights and allowed the existing rates to continue in force. The situation can best be made clear by quotation from the letter of the Manufacturers' Association to the Textile Council under date of November 19, 1908.' Believing that indications point to a prosperous season, the Manufacturers' Association is inclined to waive for the present occasion its right under the contract to reduce wages and to suggest, if it be agreeable to the Textile Council, that the present rate of wages be maintained for the ensuing period of six months, it being thoroughly understood that the extra wage thus paid is something over and above what is required by the contract, which still remains binding on both parties and is offered as in some way a substantial recognition of the good faith of the operatives in remaining true to their con- tractual obligations. If, therefore, the members of the Textile Council approve this suggestion, and thus indicate their understanding that no precedent is hereby established and that the textile agreement still remains in full force, the arrangement outlined above will become effective. The Commercial and Financial Chronicle gives an additional suggestion of what the conditions were which made this action of the manufacturers possible."^ It is explained [by the manufacturers] that supplies of cotton were secured much below recent quotations ; had the manufacturers been forced to buy raw materials and sell pro- ducts at current quotations, the waiving of the rights would not have been possible. As is shown in Table VI, the succeeding periods of wage 1. Handbook of the Carders' Union, p. 19. See also Massachusetts Bureau of Statistics, Annual Report, 1909, pp. 44-47; Textile World Record, Vol. XXXVI, pp. 314, 315. The Textile World Record said with reference to the action of the manu- facturers: "They do this undoubtedly because they believe that insisting on the reduction would be an injustice. This is the practical recognition of the important fact that the wages of textile mill operatives are below the American wage standard." (Vol. XXXVI, p. 315.) 2. Chronicle, December 11, 1909, p. 1509. RATES OF WAGES AT lALL RIVER 29 calculation witnessed repetitions of the employers' action of November, 1908. In 1^'cbruary, 1910, the operatives, keenly dis- appointed at the failure of wages to slide upward, voted to withdraw from the agreement. This gave rise to fresh discussion of future rates of wages, and tentative plans were taken up. The proposal of the employees was that of a sliding scale in which 19.66 cents per cut, the rate then being paid, should be the minimum rate, this to be paid at a margin of 67.5 cents or lower. The maximum rate proposed was 26.03 cents at a margin of 1 1 5 cents. Such a scale, if adopted, would have meant an advance all along the line of between 8 and 10 per cent. The manufacturers would not agree to such a marked change, being willing to make only a smaller advance on the sliding scale basis. Consequently the sliding scale experiment was abandoned.^ TABLE VI Actual Margins and Wages under the Sliding Scale Agreement of May 14, 1907 (Revised September 8, 1908) Dates of Average margins for the preceding six- month periods. Weaving rates for the the succeeding six-month periods. wa(;e changes. Rates called for by the agreement. Rates actually in force. May 27, 1907 Nov. 25, 1907 May 26, 1908 Nov. 26, 1908 May 28, 1909 Nov. 29, 1909 May 30, 1910 (Cents) 115.60 2 130.89 79.00 60.82 74.01 66.90 66.15 (Cents) 23.96 23.96 19.66 18.00 18.00 18.00 18.00 (Cents) 23.96 23.96 19.66 19.66 19.66 19.66 Agreement abandoned 1. Chronicle, September lo, 1910, p. 622. 2. The margin of profit upon which the weaving rate for the six months following May 27, 1907, was based was the average margin for a period slightly shorter than six months, ending Ajiril 30, 1907 (Chronicle, May 11, 1907, \i. 1129I. The other s2-inch prints. 2. Average annual prices of middling upland cotton. RATES OF WAGES AT FALL RIVER 31 between weaving rates and raw cotton prices of the same year is the only coefficient of the six greater than -|--5 (and if from this coefficient the probable error be subtracted the coefficient falls below the -I-.5 mark) ; singularly, because, if one assumes a direct causal relationship between margins and wages, one would expect that the increase of cotton prices would tend to lower wages, not to raise them. Whatever be the explanation of the relative values of these coefficients, we must conclude that the sliding scale agreements were founded upon no adequate basis of experience ; that they cannot be regarded as providing a mechanism to secure without friction the results of the operation of the competitive forces through which wages in this industry had previously been adjusted. Second, the most elaborate and nicely adjusted system of determination of rates of wages offers no guarantee against un- employment ; and we have seen that the Fall River operatives, in their brief experience with the sliding scale, found that cur- tailment of production by means of short-time operation was still an unsolved problem. Under unsatisfactory conditions of manufacture the employer has two means of securing relief so far as his relations with his employees are concerned ; namely, the reduction of rates of wages, and partial or complete stoppage of production. It is evident that the method of a horizontal reduction of wages, particularly of piece rates, presents to the manufacturer the most direct way of reducing costs per unit of product. On the other hand, the successfulness of an attempt to lower rates of wages when the margin is small and when, from the standpoint of the manufacturer, a direct lowering of the cost per unit seems most to be desired may be limited, even though indefinitely, by the approach of real earnings to the minimum of the workers' standards of living. As to the conditions of trade, these may be such that the method of rate cutting tends to defeat its own purpose. Rate cutting, by decreasing costs, tends to encourage large production. At a time of over-production, when the mills have on hand large stocks of goods unsalable except at low 32 WAGES IN COTTON MANUFACTURES prices, the obvious remedy is not further manufacture but short- time operation or dismissal of part of the operative force. For a low cost of production in the immediate future cannot remedy the hardships resulting from high cost of goods held in store- houses and not marketable at a profit. On the other hand, the partial suspension of work has the advantage of actually cur- tailing output and also of being able to be effected with less open opposition on the part of employees. Systematic and combined curtailment of output, although difficult of successful enforcement, has been sufificiently preva- lent in the New England cotton manufacturing industry to have an important bearing upon the soundness of the sliding scale plan. It is of considerable importance that this curtailment and the unemployment resulting therefrom have occurred with no appreciable seasonal regularity. While it is true that the months of July, August and September (when mill stocks are approaching exhaustion and managers are waiting for the mar- keting of the new crop of cotton) are the months which most frequently show unemployment in excess of the average for the year, it is true that this seasonal swing is very slight. Unem- ployment that occurs irregularly offers greater difficulties to be overcome in wage agreements than that which is regularly recurring. It appears from examination of the materials relating to the history of unemployment in the Massachusetts cotton manufac- turing industry, presented elsewhere in this monograph, ^ that in the period of thirty-one years from 1886 to 1916 inclusive there were 10 years ( 1886, 1888, 1893, 1894, 1896, 1897, 1898, 1903, 1904 and 1908) in which the number of cotton textile operatives reported to have been unemployed at some time during the year constituted at least 10 per cent, of the average number of employees for the year. Of these ten years, two (1886 and 1888) were years in which this unemployment existed while wage advances were being made at Fall River. In two other of these ten years (1896 and 1897) rates of wages at Fall River were unchangad. While in the remaining six I. Chapter V. RATES OF WAGES AT FALL RIVER 33 years (1893, 1894, 1898, 1903, 1904, and 1908) unemployment existed in the face of very substantial wage reductions. In other words, in these six years wage reductions were supple- mented either by part-time operation or by temporary dismissal of a part of the operative force. With the particular form of curtailment we are not here concerned except to note that at Fall River the tendency has been for the manufacturers to act collectively, to run on part time, and thus to retain the advan- tages of the existing factory organizations of the operative force. In an industry in which such conditions as these prevail it cannot be ignored that a sliding scale plan may be ever so finely elaborated solely with reference to rates of wages and still offer no guarantee that the wages thus determined shall actually be paid ; for the sliding scale does not assure employment. It merely provides that if there is employment to be had, com- pensation shall be on a prearranged plan. If, however, the sliding scale system does sometimes tend to reduce unemploy- ment by bringing about almost frictionless reduction of wages, then it devolves upon the laborer to weigh for himself the alter- natives of steady work at low pay or more fluctuating employ- ment at a higher rate of compensation per piece or per hour. Third, the so-called margin of profit is not adequate to indi- cate the wage-paying ability of an industry, assuming that the true unit margin of profit can be ascertained and that no diffi- culty exists to hinder or prevent the collection and utilization of price information in computing this margin. The philosophy of the sliding scale appears to be this: that it will eliminate from industry the friction of industrial warfare and its consequent social waste, at the same time giving reason- able assurance that the individuals and corporations concerned shall lose nothing which the operation of competitive forces would secure for them. In a sense the sliding scale is a pros- perity-sharing plan. Now it is quite obvious that the prosperity of the cotton manufacturing industry is not a thing to be meas- ured by the manufacturer's gross margin of profit. This method of prosperity measurement takes no account of the 34 WAGES IN COTTON MANUFACTURES changing vohinic of trade under changing margin conditions. It is entirely conceivable that the textile mills of New England may secure larger sums of money available for dividends under conditions of narrow margins and large sales than under con- ditions of wide margins and small sales. To these considerations bearing upon the inadequacy of the margin as a basis of wage calculation add another — that the margin as calculated at Fall River was a crude one, which ignored entirely not only the volume of trade but also the existence of heavy fixed charges upon the business. As concrete evidence of the inadequacy of the margin basis of wage adjustment witness the failure of the Fall River manu- facturers to insist upon rigid adherence to the terms of the agreement of 1907, when, in 1908, "believing that indications point[ed] to a prosperous season" they made slight recognition of existing low margins. That the mill owners, through their representatives, were scrupulously insistent that their considerate action should not be regarded as a precedent is beside the point. A schedule of margins and rates of wages had been agreed upon, and within eighteen months the schedule was found to be not at all indicative of wage-paying ability. Fourth, it is doubtful whether it is practicable to find the real margin of profit. The sliding scale agreements are to be criticised on the ground that in the calculation of the margin of profit only two grades of cloth were considered. Although at Fall River there is a greater degree of homogeneity of the cloth products of the mills than is found in many of the New England textile centres, it is true that here at any one time thousands of employees are engaged in the production of cloths other than the 28-inch and 383^-inch prints. If the wages of these operatives are to slide in accordance with changing margins of profit, it would seem no more than reasonable to expect that the margin be calculated with reference to more than two kinds of cloth. For there can be no assurance that the prices of these two cloths are indicative of the cloth market so far as these particular mills are concerned.^ I . It is interesting to note that the 28-inch 64 x 64 prints are no longer regarded as the most typical mill products. The best statistical tabulations of cloth prices now give instead quotations on 27-inch 64 x 60 prints. RATES OF WAf;ES AT FALL fvIVER 35 A similar criticism might be made of the use of cconil)cr, 1907, p. 26; ciiition of December, 1910, p. 26. 36 WAGES IN COTTON MANUFACTURES of the dates of margin and wage calculations to meet this seasonal fluctuation is not easy for the fluctuation is not regular. To make a more positive statement of this criticism — the calculation of margins, if made at all, should be made with a knowledge of the prices at which the mills concerned have actually bought supplies of cotton, and with a knowledge ako of the prices actually received from time to time for the mill products. Only in this way can the real margin be ascertained. Probably this criticism is destructive rather than constructive. It is doubtful whether it would be feasible to make a sliding scale agreement which would reveal not only to employees but also to competing enterprisers the marketing policies of each corporation.^ The probable result of such an attempt to eliminate labor disputes would be the bargaining of each mill with its own employees and the elimination of collective bar- gaining on a large scale. It is reasonable to suppose that mill owners would resort to pure profit-sharing devices rather than to accept a system which would disclose methods by which profits are secured. Fifth, this experiment was based upon the ungrounded assump- tion that wages in any one industry can be determined by rules which relate only to the internal affairs of that industry. The ability of cotton textile operatives to secure an advance or to resist a reduction of wages cannot be measured solely by market conditions within the industry. Margins of profit have to do with the demand for labor ; but the prosperity of the cotton manufacturing industry, however indicated, does not constitute the sole demand for the services of cotton mill operatives. Of this witness the attraction of mill workers into munitions factories. Also margins of profit are not directly related to the supply of labor, which cannot be regarded as a fixed or even steadily growing quantity, but which varies with population changes, both quantitative and qualitative. That labor supply tends to conform to labor demand is a mere truism, which means that I. See the statement of Mr. Edward Stanwood, secretary of the Arkwright Chib of Boston, in a review of Professor Copeland's book in the American Economic Re\4ew, vol. Ill (June, 1913), p. 372. RATES OF WAGES AT FALL RIVER 37 the number of goods or services sold must be equal respectively to the number of goods or services bought. It must be evident that any formal agreement, however elaborately drawn, if based upon intra-industrial conditions only, cannot be expected to guarantee to either workers or employers all that could be secured by hard bargaining under competitive conditions. That is, at comparatively frequent intervals the whole sliding schedule would have to be considered anew. Ipso facto the scale would not slide ; it would become a new scale. This brings us to a final critical comment upon the sliding scale plan : That the detailed construction of a schedule of margins and wages, in the last analysis, is a matter of the com- parative bargaining powers of employers and employees. As we have seen, the sliding scale experiment was abandoned in 1910; first, because the employees were dissatisfied with the failure of rates to slide upward, and, second, because they were unable, in considering the extension of the agreement, to raise the whole scale of rates to a level satisfactory to themselves. In spite of the fundamental unsoundness of the whole scheme, it is not improbable that it would have been continued beyond the spring of 191 o, if both sides had been able to agree upon a new schedule. The difficulty appears to be that there is no criterion by which the economic correctness of any proposed schedule maj' be judged. What would be an economically correct schedule? Presumably such would be a schedule which would accord to either party all that could be secured by the most vigorous and keen bargaining. Or, to put the question in another form, what is the criterion of the wage-pa\'ing ability of an industr)-? We do not know. The answer is still in process of being evolved in the cases of public and quasi-public utilities. Governor Douglas, in 1905, when arbitrating the wage dispute at Fall River, made certain allowances for dividends and fordepreciation. But other authorities might make different allowances for the same items. This is a difficult problem, especially in the case of the Fall River cotton mills, regarding wiiose capitalization comparatively little is publicly known. ^ I. Cf. Copeland, op. cit., ch. 15. 38 WAGES IN COTTON MANUFACTURES The answer to the last two criticisms (that adjustments of any sUding schedule must be made periodically to meet changing industrial conditions and that there is as yet no criterion by which the sliding schedule may be made accurately to reflect the state of the competitive markets for labor) may be of this nature: That great nicety of adjustment is not to be expected in such a schedule; that some gain will be effected if only the wastes of industrial conflict can be avoided ; and that "in the long run" one side will gain through this inaccuracy of adjust- ment as much as the other. Of this last contention, however, there is no certain proof. While it might be safe to rely upon this supposed tendency to equalize advantages and disadvantages thus arising, if the sliding scale scheme were fundamentally sound, we have seen that in the Fall River case, at least, this soundness did not exist. From May 26, 1908, the last date in the sliding scale period on which the piece rates actually changed, until March 25, 1912, the weaving piece rate remained 19.66 cents. On the latter date, at the conclusion of the textile strike at Lawrence arising out of the legislation reducing the length of the working day for women and children, the weaving rate became 21.62 cents per cut, involving an advance of 10 percent. Then there followed a period of nearly four years with no change in piece rates, culminating January 24, 1916, when under the combined operation of war demands on the cotton industry itself and of demand for workers in munitions factories the textile operatives of Fall River, as well as those of all New England, entered upon a period of frequent and considerable wage advances. The weaving rate determined upon January 24, 1916 was 22.71 cents, involving an advance of 5 per cent. On May i of the same year another advance of 10 per cent, brought the rate to 24.98 cents. On December 4, 1916, the rate was again advanced 10 per cent, to 27.48 cents. Thus in 1916 alone after a long period of stable rates the Fall River operatives received advances which brought the rates at the end of 1916 to a level over 39 per cent, above those prevailing in 1908 under the sliding scale agreement, and 27 per cent, above those effective in March, 1912. RATES OF WAGES AT FALL RIVER 39 At the time of the advance of wages effective December 4, 1916, it was agreed that there should be no change for six months thereafter.' On May 4, 191 7, the Textile Council notified the manufacturers that another advance of 10 per cent, was expected effective June 4 for the succeeding six months.- In the meantime the United States had become a participant in the world war, heavy demands had been and were being made upon the cotton manufacturing industry for army and navy supplies, the already prevailing scarcity of textile operatives caused by the drain to the munitions factories had been increased by voluntary enlistments for military service, and the operation of the machinery for the enforcement of the selectixe draft act held out the prospect of an even greater dearth of labor. Mills could ill afford stoppage of production, even if such stoppage had been permitted by the government. Consequently the advance requested was granted without delay, the new weaving rate being 30.23 cents. In October, 191 7, under the continued stimulus of war con- ditions the Textile Council at Fall River again requested an advance of wages, this time of 15 percent. The manufacturers countered with an offer of 10 per cent, which was not accepted. Finally a compromise was effected at an advance of I2^j per cent, subject to the approval of a representative of the United States Department of Labor. The new weaving rate, 34.02 cents, became effective December 3, and finally was approved in February, 19 18. The Government representati\e urged that the rate be continued until the end of the customary six-month period, about June i.'' With the approach of the latter date the operatives asked for an advance of 25 per cent., but compromised at 15 per cent., bringing the weaving rate to 39.12 cents, effective June 3.'* Thus up to the end of the first half of 1918 advances of wages were secured at Fall River aggregating an increase over 1. Chronicle, November iS, 191O, p. iSiS; August 25, 1917, i». jSj. 2. Chronicle, August 25, 1917, p. 7S2. 3. Ibid., August 31, 191 8, p. 869. 4. Idem. 40 WAGES IN COTTON MANUFACTURES the rates fixed May 26, 1908, of 98.98 percent, and an increase over the immediately pre-war wages fixed March 25, 1912, of 89.01 per cent. Since the entrance of the United States into the war the gains in wages have aggregated 42.35 per cent, of the rates effective December 4, 191C. Conditions in the industry from June until the first of November, 191 8, were such as to give rise to serious expecta- tion of further advances on December first. Unprecedented prosperity, in spite of governmental price control since the early summer, existed contemporaneously with an ever increasing shortage of labor caused not only by the extension of the draft law but also by the epidemic of influenza. But with the signing of the armistice there came an abrupt change in the form of stoppage of production. The Fall River Textile Council late in November presented a request for another advance of rates. Refusal by the employers was prompt and firm, and was acceded to with little delay by the operatives. From that time until the completion ofthis monograph in May, 1919, operation continued on the basis of the 39.12 cents weaving rate, modification of working conditions caused by the new market conditions taking the form chiefly of the forty-eight hour week. FULL TLME EARNINGS, 1860-189I 41 CHAPTER III The Movement of Full Time Earnings in the Massa- chusetts Cotton Manufacturing Industry, i 860-1 891 The statistical materials assembled and analyzed in the pre- ceding chapter relate directly to rates of wages. Next, logically, come earnings, that is, the sums of money actually received by operatives per day or per week ; and since this study is primarily historical, relative earnings or "index numbers" of earnings will first claim our attention. Furthermore, we are here concerned with full time earnings. Unemployment and its effects upon pay envelopes in different years constitute a topic for separate consideration, reserved for Chapter V. In the present chapter and the one immediately succeeding, the assumption is made that the quotations of daily and weekly earnings selected from the available statistical raw materials were originall}' compiled as being typical of conditions prevailing in the years or other periods in question. Hence the significance of "full time earn- ings" in the titles of Chapters III and IV. For the study of relative full time earnings of employees in the cotton manufacturing industry of New England since i860, the chief sources of statistical information are such as to make advisable a three-fold division of the period. For the years 1 860-1 880 there exist two comprehensive bodies of statistical raw materials — the so-called Weeks Report in Volume XX of the reports of the Tenth Census, and the well known Aldrich Report of 1893.' These sources have been ably examined by Professor Wesley C. Mitchell in his Gold, Prices, a>td Wai:;cs under the Greenback Sta)idard. From the results of his study of general wage movements there have been selected for presentation here the facts of importance relating to the cotton mills of New I. Senate Committee on Finance, Report on WhoUsaU Prices, on Wages, and on Transportation. Senate Reports, 52(1 Congress, 2d Session. Report 1394, pts. i -4. 42 WAGES IN COTTON MANUFACTURES England. These facts, supplemented by explanatory material, serve here as the foundation for the earnings statistics covering later years. For the years 1 880-1 891 the most nearly complete and satis- factorily comparable earnings statistics are found in the volumes of the Aldrich Report. These materials have not heretofore been examined for the purposes of such a study as the present one, and it has been the task of the present writer to apply to certain of these series of wage quotations statistical methods calculated to secure continuations of the curves of relative full time earnings constructed by Professor Mitchell for the years 1860-1880. Third, for the period since 1890 there are available in the Reports and Bulletins of the United States Bureau of Labor and of its successor the United States Bureau of Labor Statistics materials for the construction of nearly complete series of full time earnings for five occupational groups of cotton textile employees. These series have been modified somewhat for purposes of comparative analysis, with special reference to the effects upon earnings of legislation reducing the hours of labor. The present chapter contains an analytical and explanatory discussion of the first two periods described, i860— 1880 and 1880— 1 891. The next succeeding chapter covers the period 1890-1916. The exhibits of the Aldrich Report include wage data for five establishments manufacturing cotton cloth, of which four were in Massachusetts. From the original exhibits for these four mills Professor Mitchell constructed 135 occupational series of relative earnings for the years 1 860-1 880, using as a base (100) for each series the actual wage reported to have been paid in January, 1860.^ These series represented all of the occupations reported for the mills in question, including several "non-textile" or " non-operative " groups of workers, such as overseers, mechanics, yard hands, sweepers and scrubbers. From these series there were then constructed other series presenting general I. See Table 5 in the appendix of his volume. FULL TIME EARNINGS, 1860-1891 43 conclusions regarding all the employees and different groups of them.^ The materials of the Weeks Report embrace tabulated returns from thirty-five cotton manufacturing establishments, of which thirty were located east of a line drawn north and south at the eastern boundary of Ohio, and of which twenty-two were in New England. Of the twenty-two New England mills, six were in Connecticut, three in Maine, six in Massachusetts, six in New Hampshire and one in Rhode Island. Professor Mitchell studied as one group the thirty mills east of Ohio, so that exact com- parison of the Aldrich and Weeks materials is not possible. Unlike the Aldrich Report, the Weeks Report made no record of the number of employees receiving each designated wage. For general purposes, therefore, as well as for the purposes of the present investigation, the information of the Weeks Report as analyzed by Professor Mitchell is not as satisfactory as that of the Aldrich Report ; but it is useful chiefly to serve as a check upon the latter. That there is little difference between the con- ditions depicted by the two sets of figures is shown by the accom- panying Table (VIII) and Chart (II) of relative earnings computed by the method of simple medians. I. For a detailed description of the methods used, see .Mitchell, op. cit., pp. 94-97. 44 WAGES IN COTTON MANUFACTURES TABLE VIII ■ Simple Medians of Relative Full Time Earnings in the Massachusetts Cotton Industry, i860- 1880 (Computed from the Aldrich and Weeks Reports) Year. Aldrich. Weeks. Year. Aldrich. Weeks. i860 100 100 1871 163 157 1861 100 100 1872 164 160 1862 100 100 1873 163 160 1863 104 ICX) 1874 154 152 1864 118 112 1875 147 150 1865 131 129 1876 141 144 1866 153 149 1877 140 140 1867 160 155 1878 138 13s 1868 158 150 1879 133 i3> 1869 157 151 1880 136 139 1870 160 15s I. This table has been compiled from the table on pp. 208-210 of Professor Mitchell's volume. The semi-annual Aldrich figures have been averaged to obtain one relative wage for each year. The simple rule of calling an average involving the decimal .5 or over equivalent to the next higher unit raises slightly in a few cases the level of the relative wage as compared with the correspondmg figure of the Weeks Report. • For four years of the twenty-one the median relatives of the two sets of quotations are identical, for each of three years there is a difference of two points, for five years a difference of three points, for three years four points, for two years five points, for three years six points, and for one year eight points. The differences of movement shown by the two curves are differences of magnitude rather than of general direction. FULL TLME EARNINGS, 1860-1891 45 CHART II The Movement of Full Time Earnings in the Massachusetts Cotton Industry, 1 860-1 880, as Shown by the Method of Simple Medians . Aldrich Report Weeks Report — i—r T I II- 1 1 r 1 1 1 1 1 1 1 1 i" 1 I - - -200 200- - - -180 180- - - -160 -140 J (^- ^^"^ >^^ ^^ 160- 140- -120 / 120- _1 AA J' - - eo — 100- ao- - 60 60- - 40 40- - 20 S T 1 1 10 1 1 1 1 1 1 1 1 1 1 1 1 2 1 1 1 1 1 i n 1 -^ ao- Therefore, using the results of Professor Mitchell's work with the data of the Aldrich Report, Table IX is presented showinfj the inovements of the Greenback prices of gold, of relative wages in the four Massachusetts cotton mills (by the methods of weighted averages and weighted medians), of relative whole- sale and retail prices of cotton textiles, and of general retail prices in the eastern part of the United States. On Charts III and IV are curves plotted on series of relatives selected from Table IX. 4H WAGES IN COTTON MANUFACTURES TABLE IX ' The Movements of Full Time Earnings in Massachusetts Cotton Mills, and of Prices, 1860-1880 I i Wages in cotton mill s. Commodity prices. All employees. Females. Males. Cotton textiles. "IcS "ec^ ■* m •0 • •0 . ■w • JU •S"£2 Prices i green Ho 1 .2^ "a "(5 4> g 2 " «» 1" £ ^ a i s-2 = i860 100 lOI 100 100 TOO 100 TOO ■ 100 I86I 100 97 102 103 100 98 124 112 1862 H3 98 lOI 104 99 161 197 129 1863 145 108 106 107 103 292 287 149 1864 203 127 117 112 120 462 414 179 1865 157 140 139 143 136 411 373 189 1866 141 161 170 172 166 320 304 181 1867 138 167 175 182 168 227 236 172 1868 140 165 170 174 164 181 205 169 1869 133 169 175 177 167 174 173 162 1870 115 166 174 179 172 159 158 152 I87I 112 177 186 194 171 146 13s 140 1872 112 183 192 209 174 151 130 136 1873 114 178 190 205 173 140 123 '37 1874 III 163 182 199 169 124 116 135 1875 "5 150 169 181 161 112 109 130 1876 112 145 163 174 157 93 102 125 1877 105 142 151 160 147 87 97 122 1878 lOI 145 151 168 147 79 92 116 1879 100 144 143 163 141 79 91 114 1880 100 153 151 165 141 90 92 115 Base (100) ^quotations for i860 (January quotations where such exist) I. Compiled from materials presented in Mitchell, op. cit., pp. 91, 103-118, 278. FULL TIME EARNINGS, 1860-1891 47 CHART III The Movements of Full Time Earnings in Massachusetts Cotton Mills, and of Prices, i860- 1880 Base (100) = quotations for January, i860 Weighted averages of relative full time earnings in Massachusetts Cotton Mills. Relative wholesale prices of cotton textiles. .X...X... Relative general retail prices in the East. . — 1 1 1 1 ■ 1 ■ > ■ ■ 1 t 1 — . 1 ^ • i\ ■ • • '400 400- 1 1 • 1 t f ; • ■ 1 1 t 1 • 1 1 ■ •500 1 \ 300- 1 \ ' * " • 1 \ • ■ 1 \ ■ •200 1 1 1 \ 200- • 1 l*"'^'0 C- t' oo - ' 00 CO CD 00 «0 .-1 fH M eH »-< ■ 48 WAGES IN COTTON MANUFACTURES CHART IV The Movement of Full Time Earnings in the Massachusetts Cotton Industry, 1 860-1 880; by Sex Groups Base (100) = quotations for January, i860 Weighted averages of relative wages — all employees. Weighted medians of relative wages — all employees. « Weighted medians of relative wages — females. X — X — Weighted medians of relative wages — males. 1 — 1 — I — I — I — I — I — I — I — I I I I — I — I — I — I — I — I — r ■Z40 220 200 180 160 140 120 100- - 60 - 60 - 40 - 20 1 ' "i* ' ' J L I'll'' 'I T 240- 220- 200' 180 160- 140' 120 •100' 80- <0- 40- 20- Examination of Table IX and of Charts III and IV shows that, for the employees of these textile establishments considered as one group, wages (as measured by the method of weighted averages) fell slightly from the level of i860 in 1861 and 1862, but from 1863 to 1867 rose steadily to a point 67 per cent, above the level of January i860. In the three-year period 1868- 1870 FULL TLME EARNINGS, 1860-189I 49 inclusive, full time earnings fluctuated somewhat, but after 1870 rose again until in 1872 they were at a point 83 per cent, above the level of January, i860. After i 872 there occurred a decline of wages, slight at first, but after 1873 rapid, reaching its low point in 1877; but this low point was still 42 per cent, above the level of the wages of January, i860. The period ends, in 1880, with the wage level 53 per cent, above the base line. Comparison of the series of weighted averages and of weighted medians for all employees shows that the use of the method of weighted averages as compared with that of weighted medians does not magnify the upward change of wages after i860. And the differences between the two sets of relative figures in these two cases are of negligible importance. While for nearly all of the period the weighted medians show advances of wages some- what greater than do the weighted averages, the two curves are approximately parallel. The curve of weighted averages is therefore selected as the more conservative and for the additional reason that the method of weighted averages tends to eliminate overstatements of wage movements which might be due to unrepresentative distribution of wage quotations in the tally sheets for any one year. The results of the use of the weighted medians have been presented here to make possible comparison of wage changes applying to the sex groups. Chart IV shows clearly that the women in these cotton mills shared in the wage advances of the greenback period to a greater degree than did the men, their advances being about 50 per cent, greater than those of the men. The table and the charts just referred to present in statistical form the chief facts relating to wage and price movements in the greenback period. It now remains to suggest some of the forces motivating these textile wage movements and to present such pieces of evidence bearing upon them as have been possible of collection from contemporary records. Some reference is undoubtedly due to the displacement of gold money as the circulating medium and standard of value during the major part of this early period ; to the effects of paper currency inflation upon price movements. Our interest here, 50 WAGES IN COTTON MANUFACTURES however, is in special markets for labor and for commodities. Conditions existing in the cotton and cotton cloth markets during the Civil War were somewhat peculiar to these markets, the prime factor at work being the so-called "cotton famine." The then fabulous prices of raw cotton, which would normally enter into manufacturers' costs and so contribute to determine the lower limit of prices of mill products, were prices indicating, not so much a depreciation of the currency, as inability of manu- facturers to obtain raw materials on any considerable scale. As to the trading in cotton that was done, to the extent that both the raw cotton and dry goods markets were highly organized, there would be a tendency for their prices to respond quickly to changed conditions of the currency. At the same time, to the extent that in either of these markets there was dealing in futures and that in the conduct of cloth manufacturing there prevailed the custom of production of goods on orders for delivery at more or less remote dates in the future, this quick adjustment of prices to changed currency conditions would be stabilized. Moreover, the use in textile manufacturing of a large amount of elaborate and expensive equipment, involving heavy fixed costs, would further retard the adjustment of cloth prices to a depreciation of the currency. There is found, however, as was to be expected, little correlation between wage and price movements in the cotton manufacturing industry and the green- back prices of gold. In as much, also, as the prices of cotton textiles both at wholesale and at retail declined more rapidly after the termination of the Civil War than did the retail prices of commodities in general, we must look more particularly into the cotton manufacturing industry itself to secure an explanation which even approximates giving satisfaction, regarding the com- petitive forces at work affecting wages. ^ I. The downward movement of the prices of cotton goods from 1863 to 1871 may be considered to be chiefly the evidence of transition from abnormal to more normal conditions in the production and marketing of cotton and in the manufacture of cotton cloth. The more gradual downward tendency of prices from 1873 ^° 1879 is in general agreement with the downward movement of prices in general. Note that cotton goods, both at wholesale and at retail declined more than did commodities in general. FULL TLME EARN1N(;S, 1860-1891 51 In a sense the rapid upward movement of the curve of weighted average relative wages from 1862 to 1866 is misleading. The quotations upon which this curve is plotted arc not at all indicative of prosperous conditions and large employment of labor; but are rather quite similar to the nominal quotations of commodity prices at times when producers are not able to fill orders. The shutting off of the source of supplies of raw cotton at the outbreak of the Civil War caused serious stoppage of mills throughout the North. It happened that no one of the mills for which data are given in the Aldrich Report was closed at the dates for which pay rolls were examined. But the number of employees enrolled steadily decreased from 1,207 'r> January, i860, to 741 in July, 1863 ; andnotuntil January, 1866, do the records show a total number employed as great as the force in January, i860. In the W^eeks Report there are several instances to be noted in which there was complete stoppage of operation.' It happened that in view of a prospective crop shortage in 1861, with no reference to the coming rebellion, some manufacturers had laid in large supplies of raw materials and by careful husbanding of this stock were enabled to keep in operation much longer than their competitors. But the shortage began to be felt very keenly in 1862 and was at its maximum in 1863. Toward the end of the war some cotton was bought through the Government under regulations imposed by the Treasury Depart- ment. Fite tells- us that during the first year of the war the mills were for the most part operating on two-thirds time and during the second year on one fourth to one-half time. Mr. Samuel Batchelder, a mill manager, says'^ that in 1 863 about one- third of the spindles of the Northern states were kept in operation. Mr. Charles Cowley of Lowell stated^ in 1868 that "nine of the great corporations of Lowell, under a mistaken belief that they could not run their mills to a profit during the War, 1. Weeks Report, pp. 331, 348, 350, 352, 356, 358, 361, 366. 2. Fite, Emerson D., Soiiiil and /niittsfrial Conditions in the Xorth during the Ci7'il ir,ir, p 86. 3. Batchelder, Samuel, Introduction and Early Progress of the Cotton Afanu- factur,; (1863), pp. 97, 98. 4. Cowley, Charles, History of Lowell, (revised edition, 1868 \ p. 60. 52 WAOES OF COTTON MANUFACTURES unanimously, in cold blood, dismissed ten thousand operatives, penniless, into the streets !" The New York Herald in its issue of January 13, 1862 (p. 3) published a long article on the con- dition of the textile industries. In the discussion of the cotton manufacturing industry, under the heading, "What has become of the operatives?" there is the following. From extended inquiry we find that the contraction of manufacturing has released about seventy thousand operatives from the mills. About one-lhird of these are males who have mostly walked out of the factory into the camp, and are doing good service in endeavoring to wrest cotton from the grasp of the rebels. The female factory hands of New England have, in many instances, returned to their rural homes until the return of better times, and others have employed themselves in making up clothing for the army. But so far there has been nothing like suffering among the operatives, consequent upon the disarrangement of manufacturing When, at the conclusion of hostilities, supplies of raw materials became more easily available and resumption of mill operations commenced on a large scale the markets for labor and for goods were quite altered from the conditions of i860. Mr. David A. Wells in a report to the Secretary of the Treasury dated Decem- ber, 1866 stated' that throughout the northern states manufac- turing industries had suffered a loss of desirable employees since comparatively few of the men who entered the army from the factories returned to their former occupations. Mr. Cowley also refers^ to the stoppage of work by the Lowell mills as a great mistake in policy, since, when the mills reopened, there was difficulty in securing competent operatives. He especially criticises the mill owners for having made no adequate provision for the employment of overseers and more skilled workers many of whom in the years of war joined the enterpriser class. But more particularly was there a scarcity of women and girl operatives. It has already been stated that the reopening of the mills after the war failed to attract many of the former 1. Senate Executive Documents, 39th Cong., 2d Sess., Doc. 2, p. 21. 2. Cowley, op. cit., p. 61. FULL TLME EARNINGS, 1860-1891 53 female operatives of American stock. The situation in the market for women cotton mill workers may best be described in the words of Commissioner Wells. The opening up of many new employments to women, coupled with an increased prosperity of the agricultural classes, has also produced in many sections of the country, an unusual scarcity of female operatives. This is particularly the case in the manufacturing districts of New England, and has not been remedied by an advance in wages. The average rate of wages paid to adult female operatives in New England cotton mills is reported to be one dollar per day, while in cases of the more skilled operatives, earnings of from twenty to thirty, and even forty dollars per month, exclusive of board, are reported. As an illustration of the independence of labor over capital in this department, it may be stated that, during the summer of 1866, the product of the cotton mills of New England was variously reduced from five to twenty-five per cent., through the inability to, obtain female operatives, even with the inducement of the highest rates of wages ever paid in this branch of manufacture. In one instance specifically reported to the commissioner, (viz; the Amoskeag Manufacturing Company, New Hamp- shire,) at least twenty-five per cent, of the machinery stood still for a period of three months in 1S66, for the sole reason of an inability to procure operatives. ^ Another result growing out of this competitive demand for labor is, that the labor itself becomes unstable in its character ; to meet which a not uncommon practice has prevailed in New England of offering ten per cent, in addition to the ordinary rates of wages, conditioned on the continuance of the engage- ment for a certain definite period. - The phenomenon above referred to, that is, the relatively greatei scarcity of female as compared with male operatives in the years immediately following the conclusion of hostilities, is reflected in the diagram depicting the wage movements of the sex groups of employees. These curves of weighted medians of relative wages rise abruptly during the war years, attaining in 1. Senate Executive Documents, 39tli Congress, 2il Session, Doc. 2, p. 22. 2. Ibid., p. 23. 54 WAGES IN COTTON MANUFACTURES 1867 levels considerably favoring the women. From 1867 on until the end of the period here under study these superior gains, although subject to greater fluctuations than those of the men, were in general maintained. In 1880 the relative wages of the women, as compared with those of i860 by the use of the weighted medians, showed a gain of 65 per cent., while the gain of the men during the whole period was about 41 percent. Contemporary accounts give interesting information concern- ing the downward course of wages after the panic of 1873. In November of that year the Commercial and Financial Chronicle noted^ editorially the stoppage of the cotton mills, which, it said, was not due to any fundamental unsoundness in the cotton manufacturing industry itself, but to the panic, with its resulting stoppage of sales and difificulty of collections. But the next year witnessed a depression more peculiar to the cotton industry itself. On September 23, 1874 treasurers and managers of mills met in Boston to consider plans for the relief of the depressed market for cloth. It was voted at this meeting that- "it is imperatively required that the production of the cotton mills of New England should be reduced by at least one- third, until the proper relation between the cost of production and the market value of goods shall be re-established, and the relation between supply and demand adjusted. At a second meeting a working schedule of four days a week was agreed upon.'^ At a similar meeting held at Providence on September 25th it was unanimously agreed to reduce production one-third for the following three months. At this meeting, according to the Chronicle,* "facts were submitted showing that the prices of most of the cotton fabrics [were] below the actual cost of production." 1. Commercial and Financial Chronicle, November i, 1873, pp. 583, 584. 2. Ibid., September 26, 1874, p. 333. 3. Ibid., October 3, 1874, p. 355. 4. Ibid., September 26, 1874, p. 333. The Chronicle of January i, 1876 com- ments (p. 4) on the difficulty of carrying out systematic and combined curtailment. Some additional references in the same publication to curtailment are April 3, 1867, p. 455; June 8, 1867, p. 757; June 22, 1867, p. 789; November 16, 1867, p. 615; October 29, 1870, p. 554. FULL TIME EARNINGS, 1860-1891 55 The extent and seriousness of the depression which this policy of curtaihnent was intended to reh'eve is shown by the fact that despite the remedial measures attempted the market was very slow in reviving. Although there was some difficulty in securing uniformity of curtailment — some mills, particularly those in Lowell, failing to get into line in the carrying out of the plan — the systematic curtailment probably amounted to 25 per cent, of normal production. This was supplemented by enforced idleness because of shortage of water power and because of strikes. The reduction of operation caused by all of these factors was estimated to amount to 331/S per cent, which was what the original plan contemplated. The market, however, did not quickly respond and would not take the small amounts of goods offered.^ Not until the summer of 1876 were there indications of a recovery of the print cloth market'^, which had been most affected by the depression. In 1880, then, we have the following situation. Gold in terms of Greenbacks was at par. Raw cotton was 12 per cent, higher in price than in i860, although in 1879 it had been at the same price as in i860. Cotton goods were selling at wholesale for 10 per cent, less than in i860, and at retail for 8 per cent. less. Commodities in general retailed in the East at prices averaging about 15 per cent, more than in i860, while wages in the cotton industry, so far as the Aldrich data are typical, averaged about 53 per cent, higher than immediately before the outbreak of the war, while the cost of living was about 25 per cent higher. We have seen that the premium on gold had little direct explanatory connection with the war prices of cotton and of cloth and could therefore have had little direct effect upon the earnings of the comparatively few operatives employed during the war. A temporary scarcity of operatives, especially of women and girls, apparently operated to secure for the workers a higher wage standard when full manufacturing activity was resumed ; while the dull years of 1873 to 1877 ^^ reflected in lower prices and curtailment of production brought gradually decreasing 1. Commercial and Financial Chronicle, February 13, 1S75, p. 166. 2. Ibid., August 5, 1S76, p. 124. 56 WAGES IN COTTON MANUFACTURES earnings. The fi<^iires for 1879 and 1880 show slightly improving conditions with respect to both prices and wages. In order to trace at all accurately the course of full time earnings in the New England cotton manufacturing industry from 1880 to 1890, it is necessary to have recourse again to the data of the Aldrich Report, since these constitute the most reliable and comprehensive body of information available. Yet it is not possible to rely upon the final report of this committee as formally presented, for the reasons which caused Professor Mitchell to go to the original Aldrich Report exhibits and apply to the information therein contained more discriminating statistical methods than were applied in the compilation of the Aldrich Report itself. As Professor Mitchell shows, ^ in this report either there was no allowance made for differences in the sizes of the groups of employees represented by the several wage quotations in the original exhibits or there was used a faulty method of weighting (on the basis of the census returns for occupations). Also, an arbitrary and unwarranted dis- tinction was made between the establishments reported as manufacturing cotton goods and the one establishment reported as manufacturing ginghams. Therefore there is presented here a continuation for the period 1 880-1 89 1 of the work done by Professor Mitchell for the green- back period. In the application of Professor Mitchell's methods, however, the number of series of quotations has been consider- ably reduced. Professor Mitchell constructed 135 series of relative wages, but they were not all representative of distinctly textile operative occupations. Beltmen, blacksmiths, boilermen, carpenters, machinists, masons, oilers, watchmen, scrubbers, sweepers and yard hands, although employed by textile mills, are not in a narrow sense textile operatives. The wages which these persons receive have not necessarily a close relationship to the wages of spinners, weavers, carders, slashers and pickers, but are rather determined by conditions in the local labor market in these several occupations. Moreover, overseers in I. Gold, Prices, and Wages under the Greenback Standard, pp. 169- 1 72, and History of the Greenbacks, pp. 280-282. FULL TIME EARNINGS, 1860-1891 57 the mill departments, while textile workers, are not operatives in the sense in which this word is commonly accepted. They are rather technical experts and skilled managers. The series of quotations in the Aldrich Report, then, for these groups were excluded from consideration, with the result that there remained 74 occupational series. Of these 27 represented females, 46 represented males, 29 represented males who in i860 received wages of less than $1.00 per day, and 17 represented males who in i860 received $i.oo or more per day. In January 1880, 1,419 persons were represented as compared with 1,774 in the series used by Professor Mitchell; in July, 1891, 1,883 were represented. To these 74 occupational series, then Professor Mitchell's methods were applied retaining as a base (100) the wages reported as being paid in January, i860, in order to secure com- parability with the results of the study of the greenback period. That Professor Mitchell's inclusion of the non-textileoccupational groups did not destroy the representative character of his results so far as concerns the New England cotton manufacturing industry, and that it is possible to treat the present computations of relative full time earnings for the period 1880-1891 as com- parable with Professor Mitchell's series, is evidenced by the figures of Table X, showing the results of the application of the same statistical methods to the two bodies of data for the year 1880.^ The discrepancies in results are so small as to be negligible for the purposes of this study. I. See Mitchell, Gold, Prices, and Wages under the Greenback Standard, pp. iiS, 120,210. By "weighted averages" is meant an average computed by counting each relative wage in the series once for each employee listed at that wage. Similarly a "weighted meiiian" is a median found by listing each relative wage once for each employee listed as receiving that wage. Hy the "simple median" method each relative wage is listed only once regardless of the number of employees represented. 58 WAGES IN COTTON MANUFACTURES TABLE X Relative Full Time Earnings in the Massachussetts Cotton Industry in the year 1880 Base (100) = quotations for January, i860 A. Results of the use of 135 occupational series of quotations. Weighted Averages. Weighted medians. Simple medians. • All employees. All employees. Females. Males. All employees. January July Average 152 149 152 150.5 166 163 164.5 141 140 140.5 135 137 136 B. Results of the use of 74 occupational series of quotations. January 152 149 164 141 136 July 157 152 163 140 137 Average 154-5 150-5 163-5 140.5 136.5 The results of the statistical analysis of the raw materials of the Aldrich Report for the years 1 880-1 891 are presented in Table XI and Chart V, showing respectively series and curves (based upon the series) of weighted average relative wages. The base used by Professor Mitchell for the greenback period is here retained, so that the curve of weighted average relative wages of all employees for the later period may be regarded as a continuation of the corresponding curve plotted by Professor Mitchell. 1 1 . There is, of course, a failure of the series for the two periods to connect per- fectly, when weighted averages of relative wages are used, as in the body of this study. (See Table X). In constructing a series to summarize the movement of earnings for the whole period since i860, the writer has used for the year 1880 a compromise figure, 154. See Chapter \I. FULL TIME EARNINGS, i860- 1 89 1 59 TABLE XI The Movement of Full Time Earnings in the Massachusetts Cotton Industry, 1880-1891 Base (100) = quotations for January, i860 .5-SS •5-2 s u J= jj ~ «* _ « T u '^ rt c« c a. n ^ ;: ^ =-6 f y UTJ >• E a ■r. t. j: u. « !i u S " X S iSZ^ ■5-2* »■ ^00 ID w « — < 2« w S-i* 1880 January 152 170 139 143 129 July '57 171 141 '44 '34 Average 155 171 140 '44 132 1881 January 147 '55 138 140 '33 July 151 167 132 '36 '3' Average 149 161 '35 138 132 1882 January 158 171 141 143 134 July 156 162 148 154 '35 Average 157 167 '45 '49 '35 1883 Januar)- 160 173 145 '49 '36 July 156 171 136 '37 '33 Average 158 172 141 '43 '35 1884 January 154 '65 '38 141 '3' July 156 167 141 145 129 Average 155 166 140 143 '3° 1885 January 150 160 136 '39 128 July 149 158 136 140 126 Average 150 '59 '36 140 127 1886 January 149 157 136 140 126 July 157 168 142 148 126 Average 153 163 139 144 126 1887 lanuary 158 163 150 '55 '34 July 162 170 '5' 156 '35 Average 160 167 151 156 '35 1888 January 162 169 150 '55 135 July 165 172 '55 162 '36 Average 164 171 153 159 136 1889 January 166 177 148 152 '38 July 172 182 '57 162 139 Average 169 180 '53 '57 139 1890 January 168 178 154 158 141 July 173 182 158 '63 '43 Average 171 180 156 161 142 I89I lanuar)' 171 180 158 '63 '43 July '73 '83 kS 164 142 Average 172 182 158 164 '43 In averaging the two relative wages for each year .5 or greater has been considered as i. in order to secure comparability with the work covering the earlier period. 00 WAGES IN CO'ITON MANUFACTURES CHART V The Movement of Full Time Earnings in the Massachusetts Cotton Industry, 1880-1891 Base (100) = quotations for January, i86o . All employees. .F'emales. - X — Males. Males receiving in i860 less than $1 per day. — ••Males receiving in i860 $1 or more per day. ^ 1 1 1 I 1 1 I I 1 1 1 1 - -240 240- -220 220- -200 200- -180 .♦.•.^•- leo- - \ ."• "*-N "■^:^ — - -i60 V- -\^^^:^/ •- 00 <-l It 1 1 t 1 1 1 1 1 1 ao- For the employees treated as a single group, the upward movement of wages which began in 1878 and continued in 1880 was of short duration. The decline to the low relative wage of 149 in 1881 was likewise temporary, wages rising in 1883 to a point higher than in any preceding year since 1874, FULL TIME EARNINGS, 1860-1891 61 the first year of depression following the panic of 1873. The panic year of 1884 witnessed a slight decline and the following year a greater one ; so that wages approached again the low level of 1 88 1. The years following 1885 showed steadily increasing earnings, the relative wage of 1891 being 172. Examining the curves of wages for the separate groups of workers, we find that, as in the greenback period, the relative wages of females (using the January, i860 quotations as the base of 100) were higher than those of males. Moreover, the men who in i860 were paid less than $1 per day received throughout this period greater relative wages than those who in i860 received $i or more per day. At the same time it should be noted that while during the short period following the greenback era the relative wages of the several groups were at the levels just indicated, using i860 as the base year, the net gains of the male employees, and especially of those males who were the poorest paid in i860, were far greater than those of the women. Measured in terms of the relative wage "points," all the employees advanced between 1880 and 1891 17 points, the females, 11 points, the males 18 points, the males who in i860 received less than $1 per day 20 points, and the better paid males i i points. In other words, those employees who in the greenback period made the least gains in wages began during the period 1880-1891 to catch up with their fellow workers. 62 WAGES IN COTTON MANUFACTURES CHAPTER IV THE MOVEMENT OF FULL TIME EARNINGS IN THE COTTON MANUFACTURING INDUSTRY OF MASSACHUSETTS, MAINE AND RHODE ISLAND, 189O-1916 The most complete and detailed body of information avail- able for the study of the movement of earnings in the New England cotton manufacturing industry for the years since and including 1890 is that published by the United States Bureau of Labor and its successor, the Bureau of Labor Statistics. This Bureau, in its 19th Annual Report (1904), commenced a series of tables showing wages and hours of labor in sixty or more manufacturing and mechanical industries. The informa- tion in this report covered the period 1 890-1903. Bulletins 59, 65, 71 and TJ, appearing in 1905, 1906, 1907 and 1908 respectively, continued the se^ries of wages and hours quota- tions for the several industries. Bulletins 128, 150 and 190 (1913, 1914 and 1916 respectively) continued the series for the textile industries. Bulletin 239 (191 8) extended the quo- tations for the cotton manufacturing industry.^ The materials relating to cotton manufacturing contained in these serial reports were collected by the agents of the Bureau from the payrolls of establishments in different states. Sum- maries were usually either for all the establishments repre- sented or in some cases by geographical divisions. Obviously such a grouping of data could not meet the requirements of the present study, and therefore recourse was had to the original exhibits or general tables of the reports in order to make a classification of data suited to the purpose in hand. From these scattered tables it was found possible to construct series of quotations for the period 1 890-1916 for four of the I. With the exception that no information was secured for 1915. FULL TIME EARNINGS, 189Q-1916 63 chief occupations in the cotton mills of three New England states, Massachusetts, Maine and Rhode Island. These occu- pations were those of card strippers or carding machine tenders, loom fixers, frame spinners and weavers. In the last named group separate series for the sexes were available. The members of the first two groups were males; those of the third group females. While, of course, the members of these few occupational groups form only a part of the force of operatives in any mill, it is to be remembered that the)' are the workers performing the tasks connected with the major or more funda- mental processes.^ And since it has been found that changes in rates of wages have affected all mill departments about equally, there is reason to suppose that the five complete occu- pational series available represent fairly accurate!)- the changing wage conditions in the industry. In the original tables published by the Bureau there were usually^ presented for each year of each scries three figures which are of interest here: (i) the average earnings per hour of the operatives whose wages were reported; (2) the nominal full time hours of labor per week ; and (3) the nominal full time earnings per week. The third item is the product of the first two. The first item was found by ascertaining for each operative included in the investigation the total earnings for the payroll period and the number of hours worked and then dividing the former by the latter. Then the hourly earnings of all the operatives in the occupational group in all the mills were added and the sum divided by the number of items. Piece and time workers were all massed together, if it happened 1 . For a particularly detailed description of the tasks of the several occupational groups in the cotton manufacturing industry, see Bulletin 239 of the Bureau of Labor Statistics, pp. 140-205. The occupation of mule spinning is the chief one of importance for which a com- plete series of (juotations is not available. This occupation, while important in so far as it is connected with a fundamental technical process and also since it belongs to a group of workers who may be considered to be among the most skilful of all textile operatives, comprises a membership which has in recent years declined relatively to the numbers of all textile employees. 2. The practice of the Bureau was not uniform. Sometimes the third item named in the text was omitted. 64 WAGES IN COTTON MANUFACTURES that in any group workers were employed under both systems. Thus earnings were reduced to a standard unit of calculation, the hour of labor. Regarding the second group of items, full time hours of labor per week, in all but two or three cases the full time hours reported in the publications of the Bureau were the maximum hours allowed by state laws for women and minors. The excep- tional cases appeared very early in the series, at a time when it was the custom of the Bureau of Labor to make allowance for short time operation if such a condition prevailed over ten months of the year. The first step, therefore, in the revision of the official figures consisted in the elimination of such unusual figures computed by a method no longer in use, A second step of revision consisted of a refinement which makes allowance for the exact dates of effectiveness of state laws reducing the maximum hours of labor for women and minors. For example, the fifty-eight hour law of Massachu- setts went into effect on July i, 1892. This meant that the operatives worked for six months in 1892 under the sixty-hour system and for six months under the fifty-eight-hour system. The average full time operation of the mills in Massachusetts in 1892 was therefore fifty-nine hours per week, and to ascer- tain the nominal or full time earnings for a week in 1892, assuming the hourly quotations to be correct or representative of conditions, these hourly earnings should be multiplied by fifty-nine. In 1910 and 1912 in Massachusetts the reductions of the working week became effective on January first. In Maine, likewise, the fifty-eight hour law went into effect on Januar}- i, 1910.^ In Rhode Island a similar law took effect on April 4, 1902.'-^ The operatives in that state in 1902 therefore worked for one-fourth of the year under a full time system of sixty hours per week and for three-fourths of a year under the fifty eight hour system. The average full time week for the year was fifty-eight and one-half hours. The Rhode Island law of 1. Maine, Bureau of Industrial and Labor Statistics, 1909, p. 476. 2. Rhode Island, Public Laws, 1902, p. 73. (Ch. 994). FULL TIME EARNINGS, 189O-1916 65 April 7, 1909 providing for the fifty-six hour week took effect on the first Monday in January, 1910.' The fifty-four hour law of the same state was efifective July i, 19 13.'- Tables XII, XIII, and XIV and Charts VI, VII, and VIII exhibit the results of the work of piecing together the disjointed parts of series in the Bureau of Labor data and of making the slight modifications just described with respect to the calcula- tion of full time hours and of full time earnings. The quota- tions of hourly earnings as given in the original tables have been unchanged except that for certain years where the quota- tion is marked by an asterisk quotations published in different bulletins covering different groups of establishments have been averaged. Such marked quotations do not include all the cases in which there has been an overlapping of reports, for cases frequently occurred in which different bulletins covering the same year gave identical quotations. The relative wages given are the relative full time weekly earnings. The base for each series of relative full time earnings has been constructed by taking a simple average of the nominal full time earnings for the ten-year period 1 890-1 899. Obvi- ously a series of relative full time weekly earnings for a series of years may serve as a series of relative full time earnings for the whole year, provided that the conditions of employment at the time the agents of the Bureau studied the payrolls were typical of conditions for the year, i, e., with respect to piece rates, quality of stock used, number of machines tended by each operative, and so forth. Each one of the charts, which refer to the conditions in Massachusetts, Maine and Rhode Island, shows that during the last decade of the last century relative full time earnings for the groups of operatives covered fluctuated back and forth above and below the base line, being slightly above the base in 1893, slightly below in 1894 and 1895, very slightly above in 1896 and 1897, and below in 1899, The year 1900 witnessed the beginning of a marked upward movement which, although 1. Rhode Island, Public Laws, 1909, pp. 41, 42. (Ch. 384). 2. Ibid., 1913, pp. 6, 7. (Ch. 912). 66 WAGES IN COTTON MANUFACTURES checked in Massachusetts (in 1901 and 1902) and even reversed in Maine (in 1902 and 1903) and in Rhode Island (in 1901 and 1904) culminated in 1907. The panic of 1907 and the years of depression immediately following caused a marked decline of earnings and not until 191 2 did earnings generally begin again to rise. Although the published reports of the Bureau of Labor Statistics do not bring the wage quota- tions up to date, they are sufficiently complete to indicate some of the results of the known advances of piece rates under the war conditions of 1916. Only two groups of operatives studied, male and female weavers in Maine, fail to show exclusively increasing earnings since 1913, and these exceptions are such as to be of negligible importance, the female weavers appear- ing to have received in 19 14 only 4 cents per week less than in 1913, and the male weavers only 12 cents less. TABLE XII The Movement of Earnings in the Massachusetts Cotton Industry, 1890-1916 Card Strippers -Male. Frame Spinners — Female. . Nominal fFull Time) ui Nomina (Full Time) e5 V E— " " 1; == Weekly Earnings. ~ c Weekly Earnings. > ° = 0^ o'5 c'E ^u.x 4< ^ Absolute Relative. s« Absolute. Relative. » (Base =$6,187) M (Base z= $5,408) 1890 60 $.0999 $5-99 96.8 $.0911 $5-47 lOI.I 189I 60 •1035 6.21 100.4 .0894 5-36 99.1 1892 59** • 1057 6.24 100.9 .0912 5-38 99-5 1893 58 .1089 6.31 102.0 .0992 5-75 106.3 1894 58 .1065 6.18 99-9 .0892 5-17 95.6 189s 58 .1064 6.17 99-7 .0912 5.29 97.8 1896 58 .1086 6.30 101.8 .0967 5.61 103.7 1897 58 .1083 6.28 101.5 .0960 5-57 103.0 1898 58 .1074 6.23 100.7 .0916 5-31 98.2 1899 58 .1027 5.96 96.3 .0892 5-17 95-6 1900 58 .1183 6.86 1 10.9 .1041 6.04 III. 7 I9OI 58 .1186 6.88 III. 2 .1025 5-95 IIO.O 1902 58 .1186 6.88 III. 2 .1028 5.96 no. 2 1903 58 .1206 6.99 113-0 .1055* 6.12 1 13.2 1904 58 .1194* 6.93 1 1 2.0 .1028* 5-96 no. 2 1905 58 .1196 6.94 112. 2 •1055 6.12 113. 2 1906 58 .1286* 7.46 120.6 .1221 7.08 130.9 1907 58 .1441* 8.36 I35-I .1307* 8.07 149.2 1908 58 •1325 7.69 124.3 .1322 7.67 141. 8 1909 58 .1262 7-32 118.3 .1242 7.20 I33-I I9IO 56 .1283* 7.19 116. 2 .1307* 7-32 135-3 I9II 56 .1287* 7.21 116. 5 .1314* 7-36 136.9 I9I2 54 • 1437 7.76 125-4 .1449* 7.82 144.6 I9I3 54 .1489 8.04 130.0 .1490 8.05 148.9 I9I4 54 .151 8.15 131-7 .150 8.10 149.8 I915 No data I916 54 .188 10.15 164. 1 .182 9-83 181.8 FULL TIME EARNINGS, 189O-1916 67 1 .IS. H.5 _ c few .5 " 1* ^ II OQ M — r<-)m>- l^t^ror^rl— r^O 0^■- iriO^^O r>.>0 N O "^"^ r^n (S r^i^ro- Q O\rno6 (soc'vo - t^O^ tnoo i^vD r*! 4- r^ 1^ d < - 'i-vO r<^ 0^ O\0O r^vO vOOO fON vnO 0\0 rO>-i — t^OO O 00 00 00000000 t^i^t>.t^t>. t>.oo 0000 c^a^c^o - c^c^c^o^o^o 4A "^ "^ "^ "^ r« ro w-i 0\ ro\0 OOOO'^f^t^N^ CT^^ r^ - ro tJ-oo »^ 1^00 iriT)-vO iriNvO r~.r-»>^ r^OO O^roOOO 'tr'iO 'I-'-' N roo^ — vO f*lfOrOTfr<1P< rommN 'T'*- i/^>0 irivO i^ O^OO vO ^O vO t^OO 00 (A o e V b 1 in V > IT . .§& — c ■3 « It >> « II vo ■* Tf t>vO0 N u-i O "^ C^VO N N Ttoo 00 — NvO — O^OO^OO C>O^OOCNO^OOOC\— — •- — — — Nr-00 rOTl-t^i-iOO t^ « H4 1-1 \o OSOO N " N O^OnO^ON"- N mOOO Tl-t^^N O 0\ O q t^ t>. t^ t^\0 VO t^ t~ t^\0 r~« t^ t^OO 00 00 0^ O^ O\00 00 00 O^00 0^ Xijno}{ r^ iri t>. u-i\o I^M rn ^ i/^rr— O -000O\O — Tj-u-it^ r^oo r^ 0000 •-' "- OOO vrirO00 1 u § 'v . .5 M Is z 2 If Tj-N TfO N^OOO r^Osr^N u-ir<^»- ONt^ O^c<0 "^ N r^ O N O ^^ 1/1 OOO^OC^C^OOC^C^-►--"---^^<^ro^^^^^l^o^'^rO 3 < r<1 O ONOO 0^\0 lOvO « vO 0^ t^OO r~.vO N •-I d'l-f^i-' OOOvO O 00 vB O 'I- a^00 00 vO m t^ t^OO N r^ tn r!tTfT}■ tA XpnoH C\OvO roi-^Tj-N ►-00 OOO ro r^OO JSvOvO Tj-O^C^^^'u-lN00 00 00 r^ O^00 t^OO OOOOr^OOO-"-NU^'1-(SNrO lilvO vo „„►,►-«„ — »-««{ScsNr)NNNNNf^NNNNN iaj sjnoH (auiixiin.-l) JCUIUIOtij O O ONOO OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOVOVO ■ £ i C 3 " 2 St jc R J, > .2? £ >. E « < H 68 WAGES IN COTTON MANUFACTURES c o X -c 4^ M pa < t3 On M c I •" o bX) 00 H "m E 1 lA V c c 'S. fi V E .E « :2ii n r^OO f~»OvO O NvO INITIO O^C^fO— ro— C^r<1"- — 'tt^ "-.vO OOOO00CT^OCT^O^O^OOOO^^P^Tl■\O •>:r •^ u^ u-, t-^ciO 00 < NO 00 ON •sSa lOH iO"^00000 ■*00 r^O\ I^vO fn Tf t^ Tl- 1-" 00 On TfOO ""i On O ro r^OvO On"^0 r^j^r^o i^^-oc mro ■-< (^rON/-i-^fOr<-)<^'ttu^w) 00 O "1 O "^ r^ Tt TfOO t^ O N r^NO TtOO O0NN00»/^N""N'c}- On O 00 <^ N On ^J-OO i- >0 N lO lONO Ot^ — 00N-"O"-'NOr^ON ■^i-iri-^u^inTi-mLovorj-u^uiin u^no no r>.t^t~.t^i~»r^r^i^t-^ On •s3ni lOH 4^ ^ ^ O <^ On fO lOOO NO rOf<^ONt<0 OnpOI^OnO 'S-^tt^NOOO N O ro ro O 00 I^ N O t^NO I^nO P)NTl-OfOONOOON—N"-roi^ 00 00 00 00 00 00 On OnOO t^OO ONONONO"--r^cl"-NNroror') <4 NO (3 CUIOI A. OH 3N n OOOOOOOOOOOOOOOOOOOOOOOOOOOOOO^ nOnOnOnOnOnOnOnOnonOnOnOnOnonOnOnOnOnOno i^i^Lriiriiri 00 •JE3^ ^ O •- N ro •* "1*0 t^OO ON O '-" N ro ■* u^nO t^OO On O -" N t^ ^ w^nO OnOnOnOnOnO^OnOnOnOnOOOOOOOOOO'-'"-'-''-'""-'- 00000000000000000000 OnOnCnOnOnOnOnOnOnCnOnOnCnOnOnOnOn FULL TIME EARNINGS, 189O-I916 69 F If 00 u . • 1 II NvO ror-'^r^O Or^^^- I^QO On O OO N ") O O^vO ^ <^00 o\ II Tj-oo t>.«tOCiO r^rJOO r^'^O "^O "^n -too •* N t^ M ^o 00 >0 00 l3 .S V u 3 "c rot^i^io — OO ro — OO OnO O^"- ^<^r^o^0^f<^"^^^l0^^ — "l-N P0>0 wl"-00vO r<10^ ir^— roO»0 >- -^Tl-inTl-N ONr^'1-"^'* On 1^ r^ t^ t^OO r^ r^OO OO t^ t^OO On OnOO 00OnO'-O0O\O""« ro >-"00—00"OONCNO"^O"^NONt^'-'00NNNmOt^O N r>.vO lO O 00 OO r^OO O u^^O O r^sO 00 TfM '^— — OOnO OnI^ N C) N r«imp< ror<1N N r-i»-ii-ii-iOO>- — O — Nt^NNNi-iNrONNNNrnr'lTl- NO N fO On O "i-OO 0'*-ONONmO»^'-"^*OON OnvO 00 t"* r^oo "OO o on'^'-oo n "i-Tru^N Tfr^o ►» t>.0 ioonOoo 00 00 Onoo 00 I^OO on t^OO 000000n0<^"-0"- — rOTl-u-> »l-l•-lM>-l<-■l-.■-.l-.l-l(s^4M^4^4>-HN^4^tc4N^4^l^<^) vonOnOnOnOnOnOnOnonOnOnOnOnOnOnonOnOnO" U-l 10 li-> U1 U^ Q VO O ►- N ro Tt u-ivo r^oo On Q " N m "^ »rno r^oo on O •- N <^ •^ "^nO OnCTnOnOnOnOnOnOnOnOnOOOOOOOOOO'-''-''-''-" — — 00000000000000000000 OnOnOnOnOnOnOnOnOnOnOnOnOnOnOnOnOn 70 WAGES IN COTTON MANUFACTURES 3 a a o o U C > o "H J=l ^ XI p^ Ss w < \o XI I ■^ o c J3 _»3 e 1 V c c ■5, CO u . 10 « II ^ t^ vnoo r^t^r^t^Tj-r^OOOM^ O -^u-)" 0^0^•-00 in r^ r)- o> 00Ci00n0CTv0'-'0--'"^OvO "i •^'j-OOOsO "-"^ r^ <^ f^ rri m rn ■rt- r'l m r^n Tt -^ u~> vrt Tf lo^O t^ t^vO t^ t^^O t~~00 XphoH 0»0'*'0\N (*>t^««00 "-I r^vo 000 ro ro I~N 1^00 t~< ro ^^00 0^ ro N "^ 00 On t-« r^oo p<1>0 N rovO iriOO Orl-'^OvON'- O <^>-' lou-iu-ivo U-) iri^o \0 iJ-ivO r^ t^ On On 00 OnO f^N ►- rorON ^ir> OOOOOOOOOOOOOOOO-""'-'-"---" (A a 13 1 0) a D. "2 a U — c = rt few — >. li 1^ iJll N CJ q i-.u^N ON'!-'**-' ONt^ t^^O ►- "-VO ONrnri^O '^N O rn Ni-;Mroes"0\6"d'^ <~nod (Air^^ «S «' N rJod t^I-^irir^ OOO0ON0ON0ONO00">-'O'-'>-'r'lfSM"-"-iNNM NO NO N u-i Ti-oo •-'00>-''-'rOfO'-'NNTi-f<^rororn'«trfii-i 00 ON 0000000000 OodoOOOOOOOOOOOOONONONO"^'*- •1- ++ a "^ •JE3A " N r'l ■* u-ivo r^oo On Q " N fO ■* "^nO r«.00 ON O •- N ro rj- wivO OnOnOnOnOnOnOnOnOnOnOOOOOOOOOO'-'""-'"""-"" 00000000000000000000 OnOnOnOnOnOnOnOnOnOnOnOnOnOnOnOnOn FULL IIME EARNINGS, 189O-I916 71 V .2 is h c u. ■3i c c u z Si!" r^'*-0 OOO 000 CTv^OO'O irsKO O. 't'^rO'^vOOO ri — OO r) CT^OO 00 CT\O0 00 00 00 r^ t>. a>00 on on 00 to ON""0000000 OnOO O mvO N ONr<^ONO •-' "- ONr^rot^-l-N OnnO ly^ PI >- "1 — ^orOTfLnr^TtrO'~n-' rO"-iTi- u-ivo -t -t^o OnCO 00 00 00 On O^ O H he ONt^N NvOOO OnNvO onN OOO ■*■- N t^N N rooo <^vO ro Tf >-2 ea O O t^OO ONOO to 1-1 N o I'NNO OnnO r^»- OnO NnO rot^ r^OO ■"^ OOOOOOOOONO'-'-'fONi-i'-'NTl-rorofOrOTr^'^ "^•-■mOi-iOOOOromN"ONOf^P»ONO'-''-' 'I-nO N f^ vOOO ONroONO ri'-"NO0000 u^OO ^ fOvO NO ■* itnvo vO t^ ionO \0 r^ r^ t^OO NO t^ t^ r^NO i^OnOn*-" ^ O O ^ Lncoror^ ^vO nO t^ |BUUUOf-i N ro Tt \j-,\o t^OO On O ►- N ro ^ "^nO t^OO On O ■- N rn t1- u-inO OnOnOnO^OnOnOnOnOnOnOOOOOOOOOO"-'-''-'"- — ""- OOOOOOiOOOOOOOOOOOOO OnOnOnOnQnOnOnOnOnOnOnOnOnOnOnCnOn 72 wa(;es in cotton manufactures CHART VI The Movement of Full Time Earnings in the Massachusetts Cotton Industry, 1890-19 1 6 Base (100) = averages of quotations for 1 890-1 899 -Card strippers, male Frame spinners, female . Loom fixers, male — x — x Weavers, female ... — ... — Weavers, male 1 1 1 1 r I T I r 1 1 i 1 1 I- r III! 1 1 II 11 1 - - -240 240- - - -Z20 220- - - -ZOO 200- - - -ISO -160 / / 1»0- 160- -140 140- -120 J6c-^ 7 ^\ Safeii^^r^ 120- - >ki «!«4n iMisrsr^ - -100 — ^*-^iffi •f^cf-. .ir *^J^ ™"*5^ - lOO- - 80 80- - - - 60 60- - - -40 40- - - -20 1 tH 1 1 1 i 1 1 1 1 1 1 1 1 1 i-JL.l_X. S 2 S 2 -L-l L. 1 1 1 i . 20- FULL TIME EARNINGS, 189O-I916 CHART VII 73 The Movement of Full Time Earnings in the Maine Cotton Industry, 1 890-19 1 6 Base (100) = averages of rjuotations for 1890-1899 . Card strippers, male Frame spinners, female .Loom fixers, male — x — x — Weavers, female . ... Weavers, male - 1 1 1 1 1 1 1 1 1 1 1 1 1 1 I 1 1 1 1 1 1 1 1 1 1 1 1 - - -240 V ' 240- - - -220 / 220- - / - -200 / / 200- -180 -160 -MO >-l20 -100 — ^•■Jt^'T^^ C^ / / 1 1 /" ' x\ ...^2^^^^'' .••'' **■•-— •■' 4 :«o- 160- 140- 120- — 100— ^.iJiyjX^-^T:-;^^ -•^^ - 80 60- - 60 60- - 40 \ 40- - 20 r> 03 CO . i 1 1 1 1 t 1 <« 0. ~ 0^ 2 <% iiiTiiiiTiiiiTiii 1 1 •-• I 20- 74 WAGES IN COTTON MANUFACTURES CHART VIII The Movement of Full Time Earnings in the Rhode Island Cotton Industry, 1890-1916 Base (100) = averages of quotations for 1890-1899 -Card strippers, male Frame spinners, female .Loom fixers, male — x — x — Weavers, female . Weavers, male 1 I ] I I I I I I I I I I I I I I I I I I I I I I I I / / / -300 -280 ' ^ / \ / -360 / / / ■«40 / -220 -«00 -180 -160 -140 .120 -100 80 -60 -40 - 20 o A o •^ o» thai^^^a l^ '•^m,^ J_J. riTli lirilliVllllliilIlT If we examine separately the charts for Massachusetts, Maine and Rhode Island, it is apparent that in each of these states the movements of full time earnings in all of the occupational groups have been quite similar. However, the chart represent- FULL TIME EARNINGS, 1890-1916 75 ing the conditions in the Massachusetts section of the industry- exhibits by far the greatest amount of correlation between the different occupational series. From 1891 to 1907 this corre- lation is very marked, while after the latter year the only curve which diverges considerably from what may be called the normal course is that for female frame spinners. This diver- gence seems to be due to some change which occurred about 1907, placing the female frame spinners in that year at a level considerably higher relatively, than the other workers. The relative heights of the occupational series established in 1907 were maintained approximately through the remaining years of the period, the general direction of movement for frame spinners corresponding quite closely with the movements of earnings of the operative groups. While it is true that the charts for Maine and Rhode Island show a greater scattering of the occupational earnings curves, and that there is apparent in these cases a much less amount of correlation between the occupational series of full time earnings quotations, it is nevertheless true that there is to be observed in each of these charts a common tendency of the wage movements of the occupational groups. In each of these cases, as in the Massachusetts case, it is the female frame spinners who appear to have departed from the path of com- mon progress. On the whole the evidence presented by these charts seems to add to the plausibility of the theory that not only have changes of rates of wages at Fall River been shared equally by all the occupational groups in that city but such standardization has to a considerable degree come to pervade the whole industry, being most highly developed in Massachu- setts, and much less developed in Maine and Rhode Island. Aside from giving some indication as to the direction of changes of full time earnings and as to the standardization of wage movements as between occupational groups, the materials of Tables XII, XIII, and XIV throw some light on the rela- tionship between legislation reducing the hours of labor and the earnings of the operatives. First, let us examine in some detail the effects of legislation in Massachusetts reducing the 76 WAGES IN COTTON MANUFACTURES hours of labor per week for women and minors from sixty to fifty-eight, from fifty-eight to fifty-six, and from fifty-six to fifty-four. In 1 89 1 card strippers at an average hourly earning of $.1035, working sixty hours per week, received $6.21 for the week's work. Working an average week of fifty-nine hours the following year at an hourly earning of $.1057 the weekly pay envelope showed a gain of $.03, while the next year, the first in which the fifty-eight hour week prevailed exclusively, with an hourly earning of $.1089 the weekly earnings reached $6.31. But declining hourly earnings in 1894 and 1895 reduced the weekly amounts below the level of 1891 and not until 1900 did the amount of the weekly earnings rise to the level of 1891 not to fall back again during the period of our study. Similarly the year 1900 marks the permanent establishment of weekly earnings at or above the level of weekly earnings prevailing in 1 891 for each of the other occupational groups studied. Of course such a long period of readjustment reflects the uncer- tain trade conditions and changing piece rates of the decade, and it is probable that this long period of fluctuating weekly earnings is not representative of the normal effects of legisla- tion decreasing the hours of labor. Again taking the Massachusetts card strippers for an example, note the effects of the fifty-eight and fifty-six hour laws which became operative January i, 1910 and January i, 1912, respectively. In 1909 at an hourly earning of $.1262, working fifty-eight hours, these men earned $7.32 per week. In both 1910 and 191 1 the hourly earnings were greater than in 1909, but not sufficiently so to compensate for the reduction of hours to fifty-six, so that in these years the weekly earnings were less than in 1909. When the fifty-four hour law went into force labor agitation throughout Massachusetts, of which the strike at Lawrence was the most violent and determining, caused advances of wages which brought increases of hourly earnings sufficient to raise the weekly earnings above the level of 1909. What happened in the case of the card strippers happened also in the cases of the other workers with the excep- FULL TIME EARNINGS, 189O-1916 77 tion of the female frame spinners, whose peculiar position since 1907 we have already noted. Their hourly earnings in 1910 and 191 1 were so considerably greater than in 1909 as to counteract the influence of the restrictive legislation. A similar result followed the restriction of the hours of labor in Maine in 1910. All that has just been stated to have hap- pened in Massachusetts with regard to the course of weekly earnings happened in Maine, except that the loom fixers and the male weavers recovered their earning powers under the new conditions by 1911. An examination along the same lines of the conditions of the operatives in Rhode Island shows that the legislation of 1902 caused no unfavorable movement of the curve of weekly earnings as compared with the level of 1901, with the exception that the male weavers appear from the table to have received in 1904 less per week than in 1901. But it is doubtful whether this item in the table shows a result of labor legislation. Rather it appears that the marked but temporary decrease of weekly earnings was due to a similar decrease of the hourly earnings. In explanation of the general rise of weekly earnings following reduction of hours is suggested the marked increase of the hourly earnings which began in 1902 and which is probably a reflection of the changes of rates which are known to have occurred in 1902, if we assume the Fall River conditions to be typical of the industry. This assumption is the more reason- able in view of the proximity of the Rhode Island mills to Fall River. This reduction of hours in Rhode Island occurred at a time most favorable to assuring beneficial results to the workers. The uncertain conditions of the panic of 1893, the depression of 1894 and 1895, the political campaign of 1896, and the war conditions of 1898 were at an end and business in general was on the road to prosperity. The Rhode Island legislation of 1909 which took effect in January, 1910, caused changes in the weekly earnings remark- ably similar to those caused by the contemporary reductions of hours in Massachusetts and Maine. The female frame spinners suffered no reduction of weekly earnings, the card 78 WAGES IN COTTON MANUFACTURES Strippers and loom fixers did not receive weekly earnings equal to those of 1909 until 1912, while the male weavers did not experience a restoration of earnings until 1914, the fifty- four hour law becoming effective in the meantime. The female weavers, according to the table, received a marked advance of weekly pay in 1910, but it is doubtful if the figures for this year are typical of conditions in the industry. The hourly earnings reported by overlapping bulletins of the Bureau of Labor were $.1950 and $.2382, of which the average, used in the present tables, is $.2166. Even the lower figure of the original exhibits would give a weekly earning of $10.92, an amount greater than that received in any other year preceding. If, then, we ignore the figures for 1910, the female weavers, like the male weavers, had to wait until 1914 for a restoration of their earnings to the level of 1909. The legislation of 1913 appears to have affected the weekly earnings of the different occupational groups in no uniform manner. In the preceding parts of this study of the movement of full time earnings the attempt has been made to ascertain the facts relating to the operative force as a whole or to certain sex or occupational groups. Wage movements have been represented graphically as line movements, the assumption being that changes of rates of wages have affected all mill employees in a substantially uniform measure. Such an analysis as the fore- going, in so far as it must be based upon data of a somewhat fragmentary nature, can be well supplemented by the study of the distribution of workers in different wage groups. The data available for such a study are unusually abundant and adequate. For the period since and including 1889 the Massachusetts Bureau of Statistics has collected from manufacturers through- out that state returns of a very complete and detailed nature, showing among other facts the wage classification at a certain time of each year of all their employees other than clerks and salaried officers. For the purpose of the present study these published summaries relating to the cotton manufacturing industry have been rearranged to secure uniformity in the groupings throughout the entire period, and all absolute figures FULL TIME EARNINGS, 189O-1916 79 have been converted into percentage figures, so that there might be made a comparison of the distribution of employees from year to year. The results of this work of revision and analysis are presented in Table XV and Chart IX.' I. For the years from 1889 to 1905 for the materials of this table, and for the years from 1886 to 1905 for the materials of Tables XVI, X\TI, and XVIII in Chapter V, double reports were prepared by the Massachusetts Bureau of Statistics. The nature of these double reports should be explained here once for all, but can be done best by reference to Table X\TI. If the reader will turn to this table he will obser\e that the Massachusetts Statistics of Manufactures dated 1894 contained data for 148 cotton manufacturing establishments which replied to the inquiries of the Bureau of Statistics for information regarding conditions in both 1893 and 1894. In 1895 nine more establishments than in the preceding year gave information concerning conditions in 1894 and 1S95. Thus there are two reports for 1894, the one covering 148 establishments, and the other 157 establishments. In no cases have there been found serious discrepancies between the reports covering any one year. In Table X\' for the years for which double reports exist averages of these figures are shown. The report on the Statistics of Manufactures for the year 1909 does not give any information regarding the distribution of employees in wage groups. 80 WAGES IN COTTON MANUFACTURES ^ 3 C C o o U «5 ^ U2 00 cs \:; , >> o J^ i) a. ii w 3 bO c o C-i ss; O r^— —0000 Tj-iON Ovir^O P» O O r^I^"^"^r^ t^vO t^ i^vo vo r^ r^ r^vo r~.oo oo oo omj^oo 0\ n n r^ t^ r^ t^ \j^\0 NO O O ~ M <~0 r^ ro OOOOOOOOOOOOOOOOOO"- "-"-►" — — - rn SIS — - ~ r^ N u^ O N vy-) ►. ro N On Tl- «»« — i-i«i-i»--N«i-i-i — ■-PJNi-i-iNr'iro N *■ vr^ 1^1 ^ — N N On O — O On N t^ On mror<1f^roP4 N N mM mTTTru-iu-iri- •^1-vO o t^ t^ r^ On O M N Tl- -" - — N §ls f'l rOOO « 00 On " 'J-VO 1^00 On "^ 00 00 0> TfvO 00 O f> OnOO O n >-iOO r>.irit~.r^«- moo tj- in ro t'l O f^ Tl- - O NO NO 00 "- ►- vO "^ vri m t^ Tt vrM^irir^t^iAirMriir^'^iril^tA.o O^ocj 6^ ei \6 -"T N N vri r^ r^oo "^ XI rOvnvr>ONONTt"-00 O 0>-" OvO t^r^Tj-Q l^P< '- rOOO O - 1- N " 00 N ■-• vn r^ uino t^ t>. ri r-. r^ \J^\D \r^\0 tJ-J^O 0>0>0 OnOO n n — N r<^ po r^ Tl- ►- "^ IT-' t^Tf- r^u-ir^N00\O00vO 'tN TtOO r^N u^tJ-On OnM r^O r<^in«- rtmmr^>- "IN OnN OnN On"" Tl-00 u^ u^oo O 00 OnnO tJ- ■* ■^ p^« r^OO OnOnOOn— — "OnO"'-'-' — N"-TtfOTr Tf tI-nO no no I^ On V =1s N ON'^mOOO r^r*■<*■ N t^ Tj- O r^OO " O NvO t^- "- l^mCvm 00 "^ "^ f!l- On « « rot~>00 O NO u^u^u-irfTTt^NONONO r^vriTfTtN M rorOOCC On O O t~»m Tf ■"i- « u V e NO OnvO t}- vri r}-NO On'^'S-OnN N n O ^^X N nO vO O NO NO ^nO On ^^ ^nO nO r^vO 1^00 t^ W^OO 00 t^NO PI 1- r >> a 3 U u 15 a, < V c 3 •-> 1— > 3 3 < V S V V c V .a u u B V 7, .0 £ 1889 99-5 99.8 98.5 100. 100. 1 lOO.O lOO.O 100. I 100. 1 100.5 100.6 100.6 1890 99-3 100. 1 99-3 100.3 100. 100. 0; 98.8 99.2 100. 100.7 100.9 100.9 I89I 100.5 100.5 100.4 100. 99-5 99-4 99-4 99.1 99.4 100.3 100.8 100.8 1892 98.9 100.3 100.4 100.7 100.5 99-8 99.4 99-3 99-6 99-8 IOI.2 100.8 1893 101.9 103. 1 102.9 104.3 104.4 103.9 103.6 89.0 81.0 98.7 loo.o 99.4 1894 105.2 105.0 104.2 103-7 103-5 103. 1 98.1 86.9 79-3 99-5 105.6 106.7 1895 100.4 100.4 100.8 100.7 100.6 100.4 98- 1 97-4 99-5 99-6 lOI.I lOI.I 1896 103.0 103.8 104.3 103.0 102.7 102.0 96.3 86.1 92.1 100.7 102.4 104.3 1897 103.4 100. lOI.O lOI.O 101.5 99-8 96.3 90-5 100.9 102.2 102.5 101.4 1898 98.9 93-1 94.4 98.4 99.8 102. 1 100.9 100.9 lOI.I 103-7 I03-3 104.0 1899 99.0 99.2 99.8 100.7 100.0 100. ol 99.8 99-0 100. 1 99.6 loi.o 102.4 1900 99.8 100.3 101.6 100.6 100.5 lOI.O 99.6 97-5 97-8 99-6 100.7 101.3 I9OI 101.9 101.8 101.9 95-4 98.0 98-3 97.0 98.6 99.0 100.4 104.2 103-5 1902 99.8 99-7 101.4 100. 101.2 99-6 99.01 99.0 98-7 100.6 IOI.7 100.7 1903 104.2 104.8 103.9 92.0 91.4 100.2 99-7 98.1 97-9 102.4 102.8 102.7 1904 108. 1 108. 1 107.8 106.9 105-5 102.8 99-2 84.1 84.2 94-3 97-9 101.5 1905 97-3 99.9 100.8 102. 1 101.8 101.4 101.2 101.2 101.8 96.1 97-7 99.4 1906 99.9 100. 1 100.5 100. 1 100. 99.8 99-5 98-7 99-1 100.5 101.4 100.4 1907 99.6 100. 100.5 100.2 100.2 TOO. 2 100. 99.0 100.2 100.8 100.4 98.4 1908 101.4 99-3 96.2 90-9 91.0 99-3 95-4 96.2 101.2 108.0 no. I 1 10.6 1909 97-9 98.4 99-7 99-7 99-9 90.6 100.3 lOO.O 101.2 100.7 100.7 101.7 I9IO 101.7 101.7 100.9 100.6 100.8 100.2 95.6 96.4 97-4 101.4 102.4 102.8 I9II 102.3 101.8 102.6 102.4 100.6 97-5 97.0 94.1 95.2 lOI.I 102.4 103. 1 1912 99-5 98.7 100.4 96.1 101.4 101.5 98-5 94-3 99-9 102.5 103-7 104.1 I9I3 101.8 101.8 102.4 101.6 99.1 97-1 96-7 94.9 99.9 lOI.I !OI.8 101.8 I9I4 104.4 103-7 103.4 103.2 101.7 100.4 98.2 95-9 96.4 97-2 97-9 97-4 I9I5 96.7 97.8 99-3 99-7 99.6 100. 1 99.6 99-0 99-6 102.2 103.1 103.1 I9I6 100. 1 100.2 100.6 100.2 99.9 99-6 98.6 98-4 99.0 100.2 101.5 101.6 Average for 28 lOI.O 100.8 lOI.I 100.2 100. 1 100.3 98.8 96.2 97.2 100.5 101.8 102.0 years 1. As in the case of Table XV the figures presented are average figures wherever two reports were published covering one year. The smallest base (100 per cent.) is that for 1889 when the average number of employees for the year as reported in 1889 for 13ti establishments was 6'>,400; as reported in 1890, for 1889, for 1.57 establishments was 68,700. The largest base was 114,900 employees representing 167 establishments in 1913. FLUCTUATIONS OK EMPLOYMENT 85 relative or percentage figures, the base (iQO) for each year being the average number of employees reported by all the establishments for that year. The results of this process of conversion are presented in detail in Table XVI. A graphic representation of the course of employment in the Massachu- setts cotton manufacturing industry from 1889 to 1916, using TABLE XVII Unemployment in the Massachusetts Cotton Industry, 1886-19 16 The Deviation from the Monthly Average Number of Employees (From the Massachusetts Statistics of Manufactures and Fable XVT of this monograph ) Per cent, which Per cent, which Per cent, which Year. lowest monthly highest roll difference between roll is of the is of the these is of the average. average. average. 1886 93-3 104.5 10.7 1887 96.1 103.6 7-5 1888 93-1 103.8 10.7 1889 98.5 100.6 2.1 1890 98-8 100.9 2.1 189I 99.1 100.8 1-7 1892 98.9 100.8 1-9 1893 81.0 104.4 23-4 1894 79-3 106.7 27.4 1895 97-4 lOI.I 3-7 189O 86.1 104.3 18.2 1897 90.5 103.4 12.9 1898 93-1 104.0 10.9 1899 99.0 102.4 3-4 1900 97-5 IOI.7 4.2 190I 95-4 104.2 8.8 1902 98.7 IOI.7 3-0 1903 91.4 104.8 '3-4 1904 84.1 108. 1 24.0 1905 96.1 IOI.8 5-7 1906 98.7 101.4 2.7 1907 98.4 100.8 2.4 1908 90.9 1 10.6 19.7 1909 97-9 101.7 3-8 I910 95.6 102.8 7-2 I9II 94.1 103. 1 9.0 1912 94-3 104. 1 9.8 I913 94.9 102.4 7-5 1914 95-9 104.4 8.5 I915 96.7 103. 1 6.4 1916 98.4 101.6 3-2 80 WAC.ES IN COTTON MANUFACTURES FLUCTUATIONS OF EMPLOYMENT H7 this method of a changing base, is given in the upper portion of the truncated Chart X. The fact which appears most clearly in the table and in the portion of Chart X referred to is that in the Massachusetts section of the industry there has been little seasonal swing of employment. The only approximation to such a swing is the slight tendency toward curtailment of production in July and August when the mills are waiting for the marketing of the new crop of raw cotton and are apt to proceed cautiously in their buying operations. We may, therefore, omit further con- sideration of this phase of the unemployment problem, and devote attention to the second phase, that of measuring the unemployment that has occurred irregularly and at times on a very considerable scale. There are two methods of making such a measurement. The one uses the statistical data just described and is in terms of of variations within each year of the numbers of operatives on the pay rolls of the reporting mills. The other is in terms of the operating time of the mills. Let us first give attention to the former. Referring again to Table XVI and to the upper portion of Chart X and also to Table XVII, it appears that in 1916 the total deviation of the operative force from the average of the monthly figures constituted 3.2 per cent, of this average. That is to say, in 1916 the number of cotton mill operatives who at any date covered by the monthly figures were unemployed constituted 3.2 per cent, of the average number of individuals employed in the industry that year. Applying the same method of computation to the available statistical information for the other years of the period 1886-19 16, we have the figures of the last column of Table XVII and on Chart X the solid line curve of the portion below the lines of truncation. On this curve the locations of the circles indicate for the several years the total deviation from the average of employment. To connect in meaning the upper and lower portions of Chart X, the perpendicular distance from any one of the circles on the solid line curve of the lower portion of the chart to the zero 88 WAGES IN COTTON MANUFACTURES line equals the perpendicular distance from the lowest point in the upper curve to the highest point in the same curve for the year in question. \]y this method of computing the fluctuations of conditions of unemployment it appears that in the period of thirty-one years covered by Table XYH the average deviation from the normal amount of employment per year has been 8.9 per cent. But since the distribution of this idleness has been so uneven in point of time, this average figure has little significance. Certain years, as 1886, 1888, 1893, 1894, 1896-1898, 1903, 1904 and 1908, stand out preeminently as years of unusual hardship, the deviation being in each of these years at least 10 per cent, and going well over 20 per cent, in 1893, 1894, and 1904 (the last named being a strike year). But it must be emphasized that the estimate just made is in terms of one dimension; that is, in terms of the numbers of persons on the pay rolls at different times. In this estimate there is no allowance made for a second dimension which concerns the duration of unemployment. For this phase of the problem again the best available infor- mation is that of the Massachusetts Statistics of Manufactures, which show the average operating time of all the establishments reporting each year. By converting the information contained in these reports, whether originally given in terms of days or of weeks, to percentage figures, considering the normal working year to consist of 51 weeks of 6 days, or 306 days, it is pos- sible to make a comparative analysis of the idle time of the mills in successive years. The results of this process of con- version are shown in Table XVIII and by the broken or dash curve of the lower portion of Chart X. It appears that the average amount of idle time per year for the period under study was slightly less than 4.5 per cent, of the normal working year. By this method of calculation, it seems, the hardships of unemployment are minimized. But it is quite significant that with respect to relative conditions prevailing in different years there is substantial agreement of the results of the two methods of unemployment measurement. FLUCTUATIONS OF EMPLOYMENT 89 The years in which the greatest idleness existed as shown by the first calculation are likewise the years of greatest idleness as shown by the second. Beyond this point in the analysis it is unsafe to proceed. There is not available the material which would make possible TABLE XVIII Unemployment in the Massachusetts Cotton Industry, 1 886-1916 The Operating Time of the Mills (From the Massachusetts Statistics of Manufactures) Time in Operation. Per cent, of Per Cent, of Year. (Days) Working Year in Operation. Working Year Idfe. 1886 301.56 98.55 1-45 1887 302.34 98.80 1.20 1888 302.64 98.90 1. 10 1889 300.13 98.09 I.91 1890 296.62 96.94 3.06 1891 308.49 100.80 —0.80 1892 304-83 99.62 0.38 1893 281.99 92.15 7-85 1894 272.91 89.19 10.81 1895 297.15 97.11 2.89 1896 279-73 91.41 8.59 1897 285.83 93-41 6.59 1898 293-05 95.76 4.24 1899 302.02 98.70 1.30 1900 299.06 97-73 2.27 1901 296.66 96.95 3-os 1902 302.66 98.91 1.09 1903 290.49 94-94 5.06 1904 280.08 91-53 8.47 1905 295.46 96.56 3-44 1906 303-56 99.20 0.80 1907 300.62 98.24 1.76 1908 269.62 88.11 11.89 1909 299.50 97.88 2.12 1910 280.80 91.76 8.24 I911 275.7 90.10 9.90 1912 284.9 93-'o 6.90 1913 289.S 94.61 5-39 1914 288.5 94.28 5-72 1915 282.8 92.42 7.58 1916 294-3 96.18 3-82 The figures for the years 1887-1906 are averages of double returns, punying Tables XV and XVI, Si-e the notes accom 90 WAGES IN COTTON MANUFACTURES in this case a more exact estimation of the extent of unemploy- ment and of its effects upon the earnings of operatives. For it must be remembered that the two phases of the problem indi- cated by the two methods of measurement used above are not mutually exclusive. There is no clear separation in the returns published by the Massachusetts Bureau of Statistics of the results of the different methods of production curtailment — short time operation of whole establishments with no discharge of operatives, and partial discharge of the operative force. It is possible that the greater proximity to each other of the curves in the lower portion of Chart X in the later years of the period than in the earlier years is indicative of a growing custom of systematic curtailment of production, when such curtailment is necessary, by part time operation with few dis- charges of employees. On this point there is avaible no con- clusive evidence. No attempt is made, therefore, in view of the nature of the information available, to compute a series of discounts to be applied to the figures showing relative full time earnings with a view to finding exact relative actual earnings. The signifi- cant facts which it is intended that the materials of this chapter shall emphasize are, first, that unemployment in the cotton manufacturing industry has been to no considerable extent regular and seasonal and thus possible of anticipation by prudent workers, and second, that when unemployment has come thus irregularly it has been, at times, on a scale so great as to affect very seriously the yearly earnings of the employees. CONCLUSIONS 91 CHAPTER VI CONCLUSIONS By way of conclusion there remains a three-fold task : first, to piece together the information relating to the movement of full time earnings in the different historical periods separately studied, thereby to construct one curve of wage movement for the entire period since i860; second, to compare this move- ment of earnings with changes in general prices, in order to form an opinion as to the changes of real as opposed to money earnings; and third, to compare changes of rates of wages with the changes of full time earnings, to see whether changes in manufacturing technique have exerted a favorable or unfav- orable influence upon operatives' earnings. First comes the task of constructing a single curve repre- sentative of the movement of full time earnings since i860. Because of the nature of the statistical materials available for study such a summary curve must relate only to the changes of conditions in the Massachusetts section of the cotton manu- facturing industry. So that the task in hand is to connect the Mitchell series of relative earnings in Masschusetts mills for the years 1 860-1 880, the series constructed in the present study from the original exhibits of the Aldrich Report for the years 1880-1891 (both of these presented and discussed in Chapter III), and some single series summarizing the five occupational scries of the Bureau of Labor Statistics for 1890- 1916 (presented and discussed in Chapter IV). The first connection is a simple one to make. As shown above in Chapter III, the application of the same statistical methods to the data of the 135 occupational series of the Aldrich Report used by Professor Mitchell and to the 74 occu- pational series studied for the years 1880-1891 gives nearly 92 WAGES IN COTTON MANUFACTURES identical results for the overlapping year 1880 — namely, weighted average relative wages of 153 and 155 respectively. The simplest and at the same time probably the most satisfac- tory method of connecting these two curves appears to be to call the quotation for the year 1880, 154, and to leave unchanged all other relatives of the two series. The element of possible error involved is small and is probably no greater than would enter into a more elaborate computation. As a preliminary step to the making of the second connec- tion, that of the series just completed from i860 to 1891 with the material covering the years 1 890-1916, it is necessary to reduce the occupational series constructed from the Bureau of Labor Statistics data for Massachusetts to one summary series. This has been done by taking a simple average for each year of the period of the five occupational quotations of relative full time earnings. The series composed of these average figures is, in view of the close correlation of the occupational series, serviceable as a summary of changes in the Massachusetts mills. It is presented in column (6) of Table XIX. The conversion of this series into a similar series continuing the one just completed to 1891, using i860 as the base(ioo), has been accomplished by the application of a formula of pro- portion — a method which, while not quite as accurate as would be an entire reworking of the quotations in the original returns made by the agents of the Bureau of Labor Statistics, is suffi- ciently accurate for summary purposes. The formula used is as follows : X (the relative wage for any year in the new series continu- ing the old series for 1 860-1 891 : the corresponding relative wage of the old series for 1890-19 16 =172 (the relative wage for 1891 of the old series for 1 860-1 891) : 100.4 (the relative wage for 1891 of the old series for 1890-1916). The figures of the new continuous series of relative full time earnings for the entire period 1 860-1916, using i860 as the base, are presented in columns (2) and (7) of Table XIX, and the curve based on these quotations is plotted as a solid line on Chart XI. CONCLUSIONS 93 TABLE XIX Wages in the Massachusetts Cotton Manufacturing Industry and General Prices, 1 860- 1 9 1 8 ^^ v<~.E U ,^ - *-<-.S M 3 C oc. « S 3 C •5c o",, *J ^^ y'5 Sst-K 10 R ^ be ^■^g.5 ( Relativ sale P Commo Gen — '^ B ^ >- ^l^-s 5E.S kP ^ ^ ^H <5P vol ec-cj ^ "- i860 lOI 100. 1890 100.8 172.7 80.1 134.8 I86I 97 97-7 1891 100.4 172.0 79.2 134.8 1862 98 115. 7 1892 100.4 172.0 75-3 •37-6 1863 108 159-4 1893 105.2 180.2 74-9 142.1 1864 127 225.2 1894 97-8 167.5 68.2 123. 1 1865 140 207.3 1895 96.1 164.6 66.4 J23-3 1866 161 184.6 1896 102.0 174-7 64.1 127.7 1867 167 165.3 1897 101.4 173-7 63.6 127.7 1868 165 161.4 1898 99-9 171. 2 66.2 II3-5 1869 169 157.2 1899 95-8 164.1 72.1 126.2 1870 166 1 40. 1 1900 110.5 189.3 78.4 140.5 I87I 177 130.0 1901 110.6 189.5 77.0 140.5 1872 •83 134-7 1902 111.3 190.7 80.1 151-5 1873 178 134.0 1903 114.7 196.5 80.6 152-9 1874 163 128.0 1904 114.2 195-7 80.1 131.6 1875 150 119.8 1905 116.8 200.1 82.2 124.4 1876 '45 iii.i 1906 126.0 215.9 86.9 137-6 1877 142 108.4 1907 140.0 239-8 91.9 163.8 1878 145 97-9 1908 133-2 228.2 87.1 151.8 IS79 144 93-9 1909 122.9 210.5 89.7 139-5 1880 154 104.6 I9I0 122.1 209.2 93-3 139-5 I88I 149 101.6 149-Ot 191 1 121.0 207.3 91.6 139-5 1882 157 103.2 149.0 I9I2 130.0 222.7 94-8 150.1 1883 158 97-6 149.0 I9I3 132.2 226.5 95-9 '53-4 1884 155 88.9 132.8 1914 133-9 229.4 94-9 '53-4 1885 150 82.1 119.1 1915 95-9 153-4 1886 153 80.5 126.9 1916 161.7 276.9 118.0 172.8 1887 160 81. 1 128.8 1917 167.8 208.3 1888 164 83.6 1 34- 1 1918 262.2 1889 169 80.6 134.8 • The relative full time earnings shown in column (fi) have as the base (100) the average of the years 1890-189'.'. t The relative lull time earnings shown in columns (2) and (7) have as the base (100) the quotations for January, 1860. J As explained in the body of the text the series of index numbers of anniinl average weaving rates presented in column (t>) of Table I h.is here been converted into a similar series having as the starting point for 18S1 the index point 149, which corresponds to the point 149 in the scries of relative lull time earnings. The base (10(1) of the series of relative wholesale prices of commodities in general is the quotations for the year 18f>0. u WAGES IN COTTON MANUFACTURES Ph c 3 ^ 'C' t; 1) c ,1^ c X ^ fc o o a rt ■t . H :/: fe o Pi D C ■frt o < O a u »2 i) 1) V a a in 0) i) ^ ei =^ > -0261 -5T6I' T r I I I n I I I I I I 1 1 r I I I I ^161- 506T- |006I' [S68T -S88I -0881 -0/.8T :S98I ^981 o o to M I I I I I I I I I I I I r o o o to o in TT 0261- o o in o to to o o o it) ■ I ■ t I I ■ I I ' I I I I I I I ■ I 1 I I I I I I I CONCLUSIONS 05 From inspection of Chart XI it appears that relative full time earnings dropped from the high point of 183 attained during the greenback period in the year 1872 to the point 142 in 1877, and that the latter relative wage has been the minimum point since the close of the Civil War. And the long continued upward movement since 1877 reached in 1916 a point approxi- mately 177 per cent above the base year i860. So much concerning the movement of full time money earn- ings, but what about real wages as measured by the purchasing power of the contents of pay envelopes over commodities in general? To help answer this question resort has been had to the index numbers of wholesale prices published from time to time by the United States Bureau of Labor Statistics. ^ These index numbers have been revised so as to make the relative price figures for i860 the base of 100 for a new series, which is presented in columns (3) and (8) of Table XIX, and has served as the basis for plotting the broken or dash curve on Chart XI. As is commonly known, the downward movement of whole- sale prices from the highest point attained during the Civil War did not conclude with the resumption of specie payments, but was merely interrupted slightly in 1880, 1881 and 1882. In agreement with the general movement, of prices in gold standard countries, the decline continued until in the United States the minimum point was reached in 1 897, the index number for that year (using i860 as the base of 100) being 6^.6. After 1897 the price movement was steadily upward until in 191 3, just before the outbreak of the Great War, the index number was 95.9 — not quite up to the level of the pre-war figure of i860. War conditions since 1914 have resulted in bringing the index number well above the base line of i860 to a point in 1917 (using annual average figures) about 68 per cent above the level of i860. Comparison of the curves of relative full time earnings and of relative wholesale prices of commodities in general plotted I. Especially to Riillelins 140 ( pji. 177, 17S) and 181 (pp. 264-266) and to the November, 1918, number of the Monthly Labor Review (_p. 88). 96 WAGES IN COTTON MANUFACTURES on Chart XI shows clearly that from 1867 to 1897, with the exception of the latter part of the greenback period not only was the purchasing power of the dollar increasing but the purchasing power of the contents of cotton mill employees' full time pay envelopes was increasing. And when, after 1897, the course of commodity prices became radically altered, the wage movement in the cotton manufacturing industry of Massa- chusetts kept very close pace with this changed movement. So that, making no allowance for fluctuations in the conditions of employment, it appears that the gain in real earnings secured between i860 and 19 16 is to be assigned to the years before rather than to those after 1897. Finally, it is worth while to note the relationship between changes of piece rates and those of relative full time earnings. In column (6) of Table I (Chapter 11) there appears a series of index numbers of annual averages of weaving rates at Fall River. The base (100) of this series in the weaving rate pre- vailing thioughout 1881. In order to secure comparability of the data concerning weaving rates with those concerning the movement of full time earnings, the figures of column (6) of Table I, just referred to, have been converted into a new series, still using the year 1881 as a starting point, but calling this point 149 (the relative full time earnings figure for 1881, when i860 is the base) instead of lOO, and computing for each year since 1881 a pew figure which has the same relationship to 149 as the corresponding figure of the old series had to 100. This new series is shown in columns (4) and (9) of Table XIX, and from it there has been plotted the dotted line curve of Chart XL With this curve added to the diagram there is made to appear in a striking way a phenomenon which may have passed unno- ticed in the reading of the preceding pages. Although in the years succeeding 188 1 until the time of the second sliding scale experiment the piece rate at Fall River fluctuated back and forth within a very narrow range of movement, the relative full' time earnings climbed upward quite steadily. Even the strong upward swing of the piece rate in 1907 was of brief duration, CONCLUSIONS 97 and not until the war conditions of 1916 were reached did the piece rate get very far outside the range of the preceding thirty-five years. In short, although there is a rather close correspondence between the minor fluctuations of the two curves, the general directions of their movements have been those of divergence. While it is not possible, because of the nature of the data studied, and especially because the piece rates are distinctly local and the data concerning full time earnings cover a wider field of investigation, and also because the series of relative full time earnings refers to both piece and time workers, to treat the evidence here presented as very conclusive, it appears not unreasonable to see in these figures and curves an indication of the beneficial effects to workers of improvements in technical processes ; to consider that the introduction of new machines and new methods has in a very concrete way resulted in industrial gains which have to an appreciable extent been shared by employees ; and at least to consider that the increase of real wages prior to 1897 and their maintenance since that date are not solely to be explained in terms of rates of wages or of hours of labor or of a combination of the two. 98 WAGES IN COTTON MANUFACTURES APPENDIX •THE TEXT OF THE SLIDING SCALE AGREEMENT OF MAY 14, 1907 1 It Is agreed by the Cotton Manufacturers' Association of Fall River and the Fall River Mule Spinners' Association, the Fall River Weavers' Progressive Association, the Fall River Loom Fixers' Association, the Card Room Protective Associa- tion of Fall River and the Slashers' Union of Fall River that wages in Fall River, so far as the mills represented by the Manufacturers' Association and the operatives who are members of the above organizations arc concerned, shall be determined in the following manner, which shall be binding upon the mills represented by the Manufacturers' Association and upon the members of the various operative associations until changed or terminated as hereafter provided. 1. That 21.78 cents per cut shall be the recognized stan- dard price for a margin of 95 points, based on the cost of 8 pounds of Middling Upland cotton and the average selling price of 45 yards of 28 in. 64 x 64 print cloth and 33.1 1 yards of 38^ in. 64 X 64 print cloth. Quotations from the "New York Journal of Commerce" shall be considered authoritative. 2. The standard of wages shall be fixed every six months, beginning the last Monday in May and November of each year, and no oftener, and shall be based on the average margin, as fixed above, for the previous six months. Prices for weaving shall be as follows : I. Quoted verbatim from The Member's Contribution Book of the Card and Picker Room Protective Association of Fall River and Vicinity, pp. 16-18. APPENDIX With a margin of With a margin of 1 15 points 23.96 85 points 20.69 1 10 points 23.42 80 points 19.66 105 points 22.87 75 points 18.68 95 points 21.78 72; y2 points 18.00 99 but there shall be no change in prices on either the ascending or descending scale unless the margin has reached a point named in the above schedule. Eighteen cents per cut shall be the minimum rate paid for weaving; 23.96 cents shall be the maximum. Wages in all departments other than weaving shall be adjusted on the price for weaving as above determined. 3. No change shall be made in this agreement and it shall remain in force unless the Cotton Manufacturers' Association on the one side or the members of the various operatives' organizations on the other side give notice of the proposed change at least three months prior to the last Monday of May and November in each year. ACCEPTED BY THE DEPARTMENT OF ECONOMICS AND SOCIAL INSTITUTIONS JUNE, 1916 / 5 8 4 6 2 This book is DUE on the last date stamped below I AUG 8 1942 l"or]ii I,-!.)- 1 ",,„ 7,:; 1 HD Howard - 4966 The movement G8U5 Qf wages in the cotton manufacturing In- dustry of New England since 1860. cans IDIHIHNMK.IUNAI llBKAHnA iiiiiiiiiiir UNIVERSITY of CALIFORNIA AT LOS ANGELES UBRARY Iii5> iitii^ (umntitif.