² Interview with J.L.B., retired labour inspector, January 1995.

³ Recent studies concerning different periods and economic sectors show how employers used different methods depending on the gender of the workers they could hire. Jordan's, E. study of the exclusion of women from industries in nineteenth-century Britain is very revealing: “The Exclusion of Women from Industry in Nineteenth-Century Britain”, Comparative Studies in Society and History, 31 (1989), pp. 273–296CrossRefGoogle Scholar. Actually, employers imposed part-time working only when they could hire women. See Walby, S., Patriarchy at Work (Cambridge, 1986)Google Scholar and Beechey, V. and Perkins, T., A Matter of Hours. Women, Part-time work and the Labour Market (Cambridge, 1987)Google Scholar.

⁴ Bumette, J., “Testing for Occupational Crowding in Eighteenth Century British Agriculture”, Explorations in Economic History, 33 (1996), pp. 319–345Google Scholar, examining occupational crowding in eighteenth-century British agriculture, demonstrates that men and women were interchangeable in the unskilled workforce. “If the agricultural labour market was not characterized by occupational crowding, this tells us something about where discrimination was located. Gender discrimination originated, not in the competitive market, but in other institutions such as the family, the law, and distributional coalitions” (p. 341). Nevertheless, even if men and women were seen as different labour forces because of differences in productivity rates, it is necessary to study the interaction between the market and the family in order to understand market and family gender differences.

⁵ For a reappraisal, see Creighton, C., “The Rise of the Male Breadwinner Family: A Reappraisal”, Comparative Studies in Society and History, 38 (1996), pp. 310–337CrossRefGoogle Scholar.

⁶ Exogenous factors means factors outside the family economy and outside the family decision-making processes. Although capitalism, as a particular social and economic structure in society, and patriarchy, as a system of values structures and relations, are also family exogenous factors, this paper will name as exogenous factors other factors. These reveal a higher degree of heterogeneity in space and time, such as the regional economy, local labour market, employer's choices, laws, institutions, customs and culture.

⁷ The studies by Folbre, N., “Patriarchy in Colonial New England”, The Review of Radical Political Economics, 12 (1980), pp. 4–75CrossRefGoogle Scholar, and Sarasúa, C., “The Rise of the Wage Worker. Peasant Families and the Organization of Work in Modern Spain” (Ph.D., European University Institute, 1995)Google Scholar, are especially interesting on pre-industrial family economies. The convenience of using alternative sources is clear in the results obtained by Horrell, S. and Humphries, J., “Women's Labour Force Participation and the Transition to the Male-Breadwinner Family, 1790–1865”, Economic History Review, XLVIII (1995), pp. 89–117CrossRefGoogle Scholar, and the contribution by the same authors in this volume.

⁸ Creighton has recently voiced the following criticism: “while the increasing diversity of approaches has highlighted a wider range of factors relevant to the problem, we are in many ways further than ever from achieving their integration into a coherent whole. Each account has focused on a narrow range of determinants and has attributed excessive importance to their consequences” (Creighton, “The Rise of the Male Breadwinner Family”, p. 333).

⁹ This analysis was conducted jointly with Eloisa Baena. The tobacco workers database based on factory records and her research on the pottery industry were cross-referenced with municipal censuses of 1900 and 1924.

¹⁰ See especially Horrell and Humphries, “Women's Labour Force Participation' about women's labour force participation and Jordan, “The Exclusion of Women” about the exclusion of women from industry in nineteenth-century Britain. Barrett, M., “Reply to Brenner and Ramas”, New Left Review, 146 (1984), pp. 123–128Google Scholar, and Creighton, “The Rise of the Male Breadwinner Family”, criticize the underestimation of alternative arrangements for female biological determinism. Norris, J., “‘Well Fitted for Females’. Women in the Macclesfield Silk Industry”, in Jowitt, A. and Mclvor, A.J. (eds), Employers and Labour in the English Textile Industries, 1850–1939 (London, 1988), pp. 187–202Google Scholar, demonstrates that the employment of married women in the manufacturing areas in which they were most heavily concentrated actually rose in the second half of the nineteenth century.

¹¹ See the contribution by Sara Horrell and Jane Humphries in this volume.

¹² “If the term ‘industrial time’ designates the new time schedules and work disciplines imposed by the industrial system, ‘family time’ refers to the internal and external timing of family behaviour at different stages of individual and family development, particularly to the timing of major demographic events […]”: Hareven, T. (ed.), Family and Kin in Urban Communities, 1700–1930 (New York, 1977), p. 189Google Scholar.

¹³ For a critical view of the treatment of women's work in economic theory see the compilations made by Amsden, Alice H., The Economics of Women and Work (London, 1980)Google Scholar and recently by Humphries, Jane (ed.), Gender and Economics (Basingstoke, 1995)Google Scholar.

¹⁴ Although analyses concerning cigarreras' breadwinning patterns in other Spanish factories have not yet been done, the labour conditions were similar, as shown in Soto, C. Candela, “Trabajo y organizatión en la industria del tabaco: las cigarreras madrilenas, 1890–1920”, Sociología del trabajo, 20 (1993–1994), pp. 91–115Google Scholar; Luque, E. Baena, Las cigarreras sevillanas. Un mito en declive. 1887–1923 (Malaga, 1993)Google Scholar; Chápuli, C. Valdés, La Fábrica de Tabacos de Alicante (Alicante, 1989)Google Scholar; Alvarez, L. Alonso, “De la manufacture a la industria: La Real Fábrica de Tabacos de la Coruña (1804–1857)”, Historia Económica, 3 (1984), pp. 13–34CrossRefGoogle Scholar; and Radcliff, P., “Elite Women Workers and Collective Action: The Cigarette Makers of Gijón, 1890–1930”, Journal of Social History, 27 (1993/1994), pp. 85–108CrossRefGoogle Scholar. The model of the female cigar-maker extends beyond the Spanish border, mainly to France and Italy, where the tobacco industries were state-owned and production was carried out by skilled female cigar-makers. For France, see ZylberbergHocquard, M., “Les ouvrières d'Etat (Tabacs-Alluments) dans les dernières années du XlXème siècle”, Le mouvement social, 105 (1978), pp. 87–107CrossRefGoogle Scholar, and Retière, J.N., “Une Entreprise d'Etat Séculaire: Les Tabacs I'example de la ‘Manu’ de Nantes (1857–1914)”, Entreprise et Histoire, 6 (1995), pp. 109–127CrossRefGoogle Scholar. For Italy, see Nava, P., La fabbrica dell'emancipazione. Operaie della Manifattura Tabacchi di Modena: storie di vita e di lavoro (Rome, 1986)Google Scholar.

¹⁵ On the history of the Spanish tobacco industry, see Vidal, J. Pérez, Espana en la Historia del Tabaco (Madrid, 1959)Google Scholar; de Torres, J. García, El tabaco: consideraciones sobre el pasado, presente y porvenir de esta renta (Madrid, 1875)Google Scholar; Delgado, E., La Renta de tabacos (Madrid, 1892)Google Scholar; and Gordillo, J.M. Rodríguez, “La Real Fábrica de Tabacos de Sevilla”, in idem, Sevilla y el Tabaco (Seville, 1984), pp. 68–75Google Scholar, and idem, “Sobre la industria sevillana del tabaco a fines del siglo XVII”, Cuadernos de Historia, VII (1977), pp. 533–552. For the importance of the tobacco monopoly to the state, see Comín, F., and Martín-Aceña, P. (eds), Empresa Pública e industrializatión en Espana (Madrid, 1990)Google Scholar; Alvarez, L. Alonso, “La modernización de la industria del tabaco en España, 1800–1935”, Working paper, Programa de Historia Económica. Fundación Empresa Pública (Madrid, 1993)Google Scholar; and idem, “Estrategias empresariales de los monopolios españoles: de la gestión pública a la gestión privada en el estanco del tabaco, 1887–1936”, in Comín, F. (ed.), La empresa en la historia de Espana (Madrid, 1996), pp. 383–398Google Scholar. For tobacco consumption, see Castaneda, J., El consumo de tabaco en Espana y sus factores (Madrid, 1936)Google Scholar.

¹⁶ The percentage that the tobacco monopoly surrendered to ordinary treasury incomes was 13 per cent in 1850, 12 per cent in 1900, 7 per cent in 1935 and 2 per cent in 1970 (Comín, F., “El sector público”, in Carreras, A. (ed.), Estadísticas históricas de Espana, siglos XIX y XX (Madrid, 1989), pp. 399–460Google Scholar, esp. p. 404. Under its 1887 contract, the CAT had to pay an annual fee of 90 million Pts. Several changes in the contract established a proportional fee to be paid to the Treasury. During the CAT period, thisr varied from 90 to 95 per cent of tobacco industry incomes.

¹⁷ See Alonso Alvarez, “La modernización de la industria del tabaco”.

¹⁸ The organization was the same in all workshops: between six and ten cigarreras worked around a table called a rancho, directed by an ama de rancho who was charged with the distribution of tobacco, tasks and wages, and the maintenance of order and cleanliness on the rancho. A number of ranchos made up a taller, a workshop, directed by the maestra, who was the major authority in the workshop, having sole responsibility for the supervision of production and order. She also had to report the functioning and performance of the Workshop to the head of the factory. The oldest maestras became porteras, who sat at the entrance of each workshop to register all comings and goings and to inspect all workers leaving the factory to prevent the theft of tobacco.

¹⁹ Boot, H.M., “How Skilled were Lancashire Cotton Factory Workers in 1833?”, Economic History Review, XLVIII (1995), pp. 283–303Google Scholar, in his estimation of Lancashire cotton factory workers' lifetime earning profiles, subtracted for female workers 7.5 years for childbearing and six months lost through sickness, leaving a working life of 23 years working compared with 35 years for males. For the cigarreras, it is necessary to subtract from 48 years, the 18 per cent average absenteeism during this period; the new average Working life in terms of earning is then 39 years.

²⁰ See Table lb for an explanation on how they were constructed.

²¹ In the tobacco factory at Seville there were four different types of workshops. The base-workshop was the desvenado where the “stemmers” stripped away the midrib of the leaf, leaving the tobacco for the picaduras (cut tobacco) and cigarette workshops. It was an unskilled task and for this reason the cigarreras who worked in this workshop were the old ones who had lost their ability to make cigars or cigarettes. The other workshops were the picaduras, cigarettes and cigars. The cigar workshop was the place where the most skilled cigarreras were employed, because cigar-making was a very complicated and specialized task.

²² See footnote 65.

²³ The strong influence exercised by local elites and the government in their effort to avoid unpopular measures in the factory suggests that the factory itself was used as a stabilizing factor in the labour market. Posadas, C. Arenas, Sevilla y el Estado. Una perspectiva local de la formación del capitalismo en Espana (Seville, 1993), pp. 298–305Google Scholar, maintains that the factory did not have a positive spillover effect on the Seville economy except in the labour market. He articulated a theory which partly attributes the survival of some other local industries to the fact that the factory was able to guarantee a secure wage for more than six thousand working women and their families, thus allowing the local bourgeoisie to fix low wages in their own industries and ensure their continued survival. In support of this theory, he has noted that the first newspapers to alert cigarreras to the potential consequences of the introduction of machinery into the factory, and to the attendant risk of job losses, were those of the bourgeoisie. This author failed in not ascribing importance to the tobacco manufacture in the local economy other than the labour market. During the nineteenth and the twentieth centuries, this huge factory had positive spillover effects on other city industries, even if the tobacco (input) and some complementary inputs together with the machinery came from other places. But even if he exaggerates the importance of the factory to the survival of other local industries because of cigarreras' wages, he does show the importance that everything concerned with the factory had in Seville. Therefore, this factory is an ideal place in which to observe how political clientelism worked.

²⁴ See in this article, the section on mothers and work and L. Gálvez-Muñoz, “Management, Labour and Gender: The Use of Time in the Tobacco Factory of Seville”, in O. Hufton and C. Sarasúa (eds), Gender and the Use of Time (forthcoming).

²⁵ Goldin, C., Understanding the Gender Gap (New York, 1990), p. 158Google Scholar, explains that changes in participation in the labour force among white married women in the period from 1890 to 1980 are in a very important way related to the reduction of hours of work, to part-time employment and, more recently, to childcare arrangements: “Indeed, the increased participation of women over the long run resulted more from a change in the nature of jobs, such as the decrease in hours of work […] than from shifts in social norms and attitudes”.

²⁶ The Decreto del seguro de maternidad was implemented by the Republican Regime, 26 May 1931. Years before this, cigarreras had already won the right to sick pay for maternity leave. The state paid 90 Pts for six weeks. In the tobacco factory, breastfeeding had always been done in the workshops. A rule of one hour of breastfeeding per day began to be applied wherever possible, particularly among employees in the mechanized workshops where it was forbidden to breastfeed babies inside the workshop.

²⁷ Following Becker, G., Human Capital (New York, 1964)Google Scholar, the training the cigarreras received was completely specific because it did not endow the worker with any skills that might be utilized in other enterprises, especially in this case because of the monopoly. Among the male tobacco labour force in Spain, only a few workers such as escogedores, who were in charge of the selection of tobacco, had specific skills.

²⁸ Article 6 of the Rules of 1927 provided for all cigar-makers older than 60 to be given the opportunity to retire. Nevertheless, there were still many over-60s who entered the Faenas auxiliares workshop after 1927, illustrating the persistence of a work culture of a lifetime in the factory, as well as family economic needs. The earning differences between wages in the desvenados workshops, where all the older cigarreras were employed and the Faenas auxiliares pay were not very high. In any case, the ratio was always variable because the first was piece-work and the second, the Faenas auxiliares allowance, was a fixed amount.

²⁹ By contrast, the tabaqueros entered the factory after having worked in other jobs. Thus their work culture was more identified with the culture of the union rather than with the culture of the factory. Arenas Posadas, Sevilla y el Estado (1892–1923), p. 620 and Baena Luque, Las cigarreras sevillanas, pp. 123–124, have criticized the lack of solidarity among cigarreras in Seville with the rest of the working class prior to the 1920s. These authors have seen this attitude in a negative light, without taking into account that their behaviour diverged from that touted in the patriarchal message of the unions (see Rose, S.O., “Gender Antagonism and Class Conflict: Exclusionary Strategies of Male Trade Unionists in Nineteenth-Century Britain”, Social History, 13 (1988), pp. 191–208CrossRefGoogle Scholar), which exercised a strong influence on the development of the working class (see Scott, J., Gender and the Politics of History (New York, 1988)Google Scholar). During World War I, at roughly the same time as the success of the major unions in Seville and the increase of tabaqueros in the factory, NICOT, the local section of the national cigarreras' and tabaqueros' union was founded. This was also the time of the entry of new cigarreras into the mechanized workshops, where the noise of the machinery replaced the work in groups chatting around a table. The end of this continuous chat was paralleled by a decrease in skills and the obligatory wearing of a uniform in the mechanized workshops, illustrating how the cigarreras' work culture was weakened by mechanization.

³⁰ AFTS Libro de Personal 1891 (sample size: 5,832) and Personnel dossier 1888–1945 (sample size: 1,655). In the second, because it covers a whole period and not a specific year, the widows are considered together with the married women.

³¹ Sara Horrell and Jane Humphries demonstrate this generational relief in the participation rates of married women in nineteenth-century Britain. See their contribution in this volume.

³² Between 1922 and 1925, around 400 apprentices were admitted to the workshops following a draw among the cigarreras' daughters and granddaughters who wanted to work in the factory. They were between 16 and 30 years old and literate.

³³ Burnette, J., “An Investigation of the Female-Male Wage Gap During the Industrial Revolution in Britain”, Economic History Review, 2 (1997), pp. 257–281CrossRefGoogle Scholar, has recently demonstrated that in most cases during the Industrial Revolution women were paid market wages and the size of the earning gap should be explained by measurement errors and productivity differences. However, the present article suggests that differences in productivity rates should be explained separately from biological differences, for differences in consumption of calories, human capital and apprenticeship.

³⁴ For Britain see Higgs, E., “Women, Occupations and Work in the Nineteenth Century Censuses”, History Workshop Journal, 23 (1987), pp. 59–80CrossRefGoogle Scholar, and for Spanish censuses, see Gálvez-Munoz, L., “Une approche aux sources statistiques. Les recencements à Seville du 1900, 1910 et 1920”, paper given at the European University Institute, mimeo (1993)Google Scholar.

³⁵ In the industrial municipal census of 1906, 25 per cent of the industrial population of Seville were women. In the Padrón de Retiro Obrero of 1921, 6,560, that is 37 per cent, of the industrial workers were female. For Arenas Posadas, “Sevilla y el Estado (1892–1923)”, p. 409, they constituted the “true proletariat of Seville”.

³⁶ Horrell and Humphries, “Women's Labour Force Participation”, call attention to the timing in the long run. It seems that the industrialization in Britain first increased female opportunities, only to reduce them later.

³⁷ In opposition to the view that unions played the main role in the exclusion of women from the labour market, Jordan, “The Exclusion of Women”, invokes the gender ideology of the period, based on the gendered division of labour established in the new industries. Employers undoubtedly encountered considerable opposition from male trade unionists when they tried to extend their female workforces, but it was the gender ideology of the period which created an androcentric blindness that prevented them from considering the possibility that some of the tasks within their industries could be performed just as well and far more cheaply by women. Burnette, J., “Testing for Occupational Crowding in Eighteenth Century British Agriculture”, Explorations in Economic History, 33 (1996), pp. 319–345CrossRefGoogle Scholar and idem, “An Investigation of the Female-Male Wage Gap”, refers to differences in productivity among male and female workers.

³⁸ Rose, , “Proto-Industry, Women's Work and the Household Economy”, as well as Tilly, L. and Scott, J., Women, Work and Family (New York, 1978)Google Scholar, underline the importance of local traditions and labour market opportunities for family employment rather than new ideologies. The participation of working-class women in the nineteenth century is less the product of new ideas than of the effects of old ideas and pre-industrial values operating in new or changing contexts. Local traditions regarding family employment, rather than new ideologies, may have influenced the reorganization of some industries during the transition to industrial capitalism, resulting in the generation of particular types of employment opportunities for women. See Horrell and Humphries, “Women's Labour Force Participation” and their contribution to this volume; Jordan, “The Exclusion of Women”; and Sarasúa, “The Rise of the Wage Worker”, for the Spanish case, have demonstrated the key role of the local labour market in explaining female participation rates.

³⁹ In the first quarter of the twentieth century the population of Seville increased by 100,000 people – from 140,000 to 250,000. In 1900, more than 40 per cent of immigrants came from the surrounding countryside. The proportion including the rest of Andalusia was 85 per cent, and 95 per cent for 1924. See Arenas Posadas, Industria y clases trabajadoras, P. 185.

⁴⁰ For the social and economic conditions of latifundio in Andalusia, see Bemal, A.M., Economía e Historia de los latifundios (Madrid, 1988)Google Scholar.

⁴¹ The single exception, according to census figures, occurred in 1920. And the reason seems to be more related with quantification problems than with real changes in the local economy.

⁴² See Arenas Posadas, Industria y clases trabajadoras, and E. Baena Luque, “Las trabajadoras sevillanas 1900–1936”, in Arenas, Industria y clases trabajadoras, pp. 225–245.

⁴³ The tobacco factory itself could be taken as an example of the way in which this labour market worked. The factory had established a system of substitutions whereby every morning a limited pool of male workers were admitted into the factory or not according to the number of absences. As a result, tabaqueros endured the pressure of the labour market every morning at the entrance to the factory. If they were late, other workers might take their jobs and their wages for the day. This policy was also common in other industrial contexts: “In a number of cotton weaving sheds in Lancashire, temporary weavers would come to the mills in the morning and would be given the looms of any permanent weaver who was late, even though the weavers were employed under piece rates. In one mill, if a permanent weaver was five minutes late, she lost her looms for the day”: Clark, G., “Factory Discipline”, Journal of Economic History, 54 (1994), pp. 128–257CrossRefGoogle Scholar, esp. p. 132.

⁴⁴ Arenas Posadas, “Sevilla y el Estado (1892–1923)”, studying a sample of the municipal censuses of 1900 and 1924 and the padrón del retiro obrero of 1921, calculates unemployment as between 20 and 40 per cent. The variation was related to the demand for agro-export customs. For him, these figures are horrific because the only way to survive was public or private charity. Although these figures are certainly horrific, kinship was still a fundamental institution. In fact, it was the family who maintained this inflated labour market.

⁴⁵ However, it may be more interesting to study fertility rates than marital status to explain the length of the cigarreras' working lives. Neither the municipal census of 1924, referring to a specific year, nor the personnel dossier are reliable sources of this information. The only source which gives information about the number of children of cigarreras is oral history, but these data are not susceptible to computation because they are approximate. It seems that the fertility was high according to the rates at the time, and it was encourged by the facilities of the factory.

⁴⁶ For an explanation of the construction of Table lb, see the section on the factory and life pattern, pp. 91–100. Only cases where the husband's profession is known have been used for this regression.

⁴⁷ Pessimistic and optimistic estimates are considered in relation to the number of days worked per week. For the construction of both scenarios, see Table 3.

⁴⁸ AFT Personnel dossiers, 1921. Angeles Castano Reina.

⁴⁹ In Valdés, Palacio, La hermana San Sulpicio (Madrid, 1887)Google Scholar the confidant of the main character is a cigarrera and her husband is shown in the story as an alcoholic and an idler.

⁵⁰ Interview with J.L.B., January 1995.

⁵¹ For Arenas Posadas, “Sevilla y el Estado (1892–1923)”, pp. 431–432, the estimates have only a superficial statistical value. He defines this proportion as the arithmetical value of a dual reality in the labour market showing a duality in industry, employing workers with different characteristics.

⁵² In the municipal poverty census of 1904, families with less than 3 Pts a day were considered poor. There were 16,158 families and 64,632 persons in this category, 60 per cent of the population. Measurement changes reduced this percentage, but it still remained very significant: ibid., p. 418.

⁵³ Interview with F.V.G., July 1996.

⁵⁴ Archivo Municipal de Sevilla, Sección estadística.

⁵⁵ The idea of de-skilling with industrialization has been substituted by that of heterogeneity with regard to changes in the workforce. First, with a redefinition of the concept of skill in the pre-industrial period, Berg, M., The Age of Manufactures. Industry, Innovation and Work in Britain, 1700–1820 (London, 1985)Google Scholar has demonstrated that, with the division of work, skill was sometimes identified with ability or speed. In the majority of cases, the de-skilling process was actually a re-skilling process and in only a few cases was the de-skilling process accompanied by a loss in wages. See Lazonick, W., “Industrial Relations and Technical Change: The Case of the Self-Acting Mule”, Cambridge Journal of Economics, III (1979), pp. 231–262Google Scholar, and Robertson, P.L. and Alston, L.J., “Technological Choice and the Organization of Work in Capitalist Firms”, Economic History Review, XLV (1992), pp. 330–349CrossRefGoogle Scholar. That was not the trend in the female skilled labour force, which suffered a pure process of degradation.

⁵⁶ Arenas Posadas, Industria y closes trabajadoras, p. 219, maintains that the importance of complementary wages to the overall family income increased during the first quarter of the twentieth century. During this period the contribution of the main wage fell from 78 to 73 per cent of total family income, which seems to confirm the degradation of work. Nevertheless, Arenas considered the diversification of family income and the degradation of skilled workers a negative phenomenon. But in many cases this degradation made it possible for women to have a job and to make a contribution to the family economy, and subsequently changed power relations within the family unit.

⁵⁷ The idea of re-skilling with industrialization is represented by the adoption of new labour roles, mostly related to supervision in the workshops by the old skilled workers. But the consequences of industrialization in a female labour force appear to have been a purely de-skilling process. Because of patriarchy, women were not considered capable of supervising other workers.

⁵⁸ It is necessary to analyse the cigarreras' family economy in the post-Spanish Civil War period.

⁵⁹ Photographs and literary descriptions reveal a beautiful way of dressing. Cigarreras were highly concerned about the way they dressed.

⁶⁰ Jordan, “The Exclusion of Women”, p. 276, says that in nineteenth-century Britain women continued to be employed in industries where the transition to industrialism was gradual, but were excluded from new fields and from industries that underwent radical reorganization.

⁶¹ Contract of Lease of the Tobacco Monopoly for the Production and Sale of Tobacco in the Peninsula, Balearic Islands, Ceuta and the rest site in Northern Africa, 16 June 1887. AHFT, Expedientes generales, Leg. 628.

⁶² The CAT already knew that it had to be very prudent in introducing machines to the workforce. For example, the Luddite-inspired revolt of 1885 in the factory at Seville was still a very recent experience. The reason for this insurrection was the rumour of the arrival of cigarette machines. The actual arrival took place more than thirty years later.

⁶³ Independent of the lease of the right to a state monopoly by a private company, the fiscal monopoly involves a distortion in the allocation of resources relative to the competitive model. In fact, when the management of a monopoly is run by a private company, the subjection of the company to private law does not mean that the company develops a private activity; the main characteristic of the company activity is derived from the existence of the monopoly, which determines its dominant position in the market.

⁶⁴ See Joyce, P., Work, Society and Politics (London, 1982)Google Scholar, and Burawoy, M., The Politics of Production (London, 1985)Google Scholar.

⁶³ The boss of the factory wrote the following to the director in Madrid in May 1922: “I could observe, with the biggest and most intimate protests, that all requests for personnel moves, leave of absence, promotions and the like are made by demanding recommendations from referees who are most of the time not related to the factory and unknown to me […]. I am aware that all these workers are observing an old and well-established custom. How-ever, I beg them to desist from these practices. No more policies based on influence, but rather on good behaviour.” The political route was resurrected during the Civil War and post-war period because the factory employed ex-soldiers and collaborators of Franco.

⁶⁶ Hareven, T., Family Time and Industrial Time (Cambridge, 1982), p. 200Google Scholar, describes a similar case. In the Amoskeag Company, even though young workers did not actually enter factory work before the age of 16, children were socialized to the work experience in the mill at an early age. Industrial labour became part of their life even before they actually worked.

⁶⁷ Prus, M., “Mechanization and the Gender-Based Division of Labour in the U.S. Cigar Industry”, Cambridge Journal of Economics, 14 (1990), pp. 63–79CrossRefGoogle Scholar, maintains that the beginning of mechanization in the American tobacco industry was due to the interest of the employers in controlling production, thus depriving cigar-makers of the importance of their skills and eliminating the apprenticeships which had been a powerful weapon of these skilled workers.

⁶⁸ To reserve positions in the factories for the workers' relatives was very usual from the beginning of industrialization, in part because of the control exercised over the workers through this practice, and in part because of the role played by the family in the apprenticeship and recruiting system. Joyce, Work, Society and Politics, gives the example of Lancashire-shire factories' bosses who obliged workers to bring their wives to the factory. Another example is the Amoskeag Company in Manchester, New England. Hareven, Family Time and Industrial Time, shows first that for the company the workers' families constituted an elastic reserve workforce adaptable to changing circumstances; second, that the factory saved on the apprenticeships because the workers were instructed by their relatives; third, that the workers developed a sociability in the factory to avoid conflicts; and finally, that the factory hired workers with a particular work culture learnt from childhood, and who therefore saw a job in the factory as the only possibility. Furthermore, the families profited from these practices because they found jobs for new family members coming into the city, and knew that the family would have sufficient employees in the factory to allow them to find lifetime employment in some capacity. The kinship system and these “labour rights' produced a comfortable atmosphere for the workers. For the economic origins of paternalism, see Huberman, M., “The Economic Origins of Paternalism: Lancashire Cotton Spinning in the First Half of the Nineteenth Century', Social History, 12 (1987), pp. 177–192CrossRefGoogle Scholar, and for some objections of his theory see Rose, M., Taylor, P. and Winstanley, M.J., “The Economic Origins of Paternalism: Some Objections”, Social History, 14 (1989), pp. 89–103CrossRefGoogle Scholar.

⁶⁹ The boss in Seville was not in favour of recruiting cigarreras' daughters, but company management and the state preferred this option to a social uproar: “Apprentices for the manual workshop? If you take account of my modest opinion, do not oblige me to hire apprentices for manual tasks; here and now, in this factory, it is better to talk about machines, mechanized workshops, apprentices for new tasks and cigarreras ‘of the future’ with other manners and education; do not talk about the cigarreras' daughters trained by their mothers or grandmothers in the infamous tasks which ultimately manipulated their families, and introduced them to the customs and vices of older cigarreras': AFTS, Leg. 636. Exp. 12, Seville, 10 September 1909.

⁷⁰ The factory began to demand cigar-makers who knew how to read and write, so the cigarreras began to send their daughters to school: “[…] I said before that the apprentices should not be older than 20, and that all of them should know reading and writing. Since then, you see all the daughters of the cigarreras learning at the schools and colleges of Seville, as a result of the mothers' objective of seeing their daughters being recruited on the same terms as the first 24 apprentices used to implement smoothly and peacefully, the reforms needed to transform the factory from manual tasks to mechanical ones without excessive protests”: AFTS, Leg. 636. Exp. 12, Seville, 10 September 1909. So literacy became a means by which to broaden professional opportunities. This accords with the thesis of Nuñez, C.E., La fuente de la riqueza. Education y desarrollo económico en la Espana Contemporánea (Madrid, 1993)Google Scholar on differences in literacy by region and by sex in Spain, which shows that the literacy rates depended on the modification in labour opportunities that the parents decided for their children in response to the changes in the labour market.

⁷¹ Under CAT management, it was possible in theory for cigarreras to bring only their very young babies with them to the workshops. This led to a loss in the transmission of skills, since in the former period older children — those over two years of age — were also in the factory caring for the little ones and learning the job.

⁷² This part of the article is mostly based on Gálvez-Muñoz, “Management, Labour and Gender”.

⁷³ For Brenner, J. and Ramblas, M., “Rethinking Women's Oppression”, New Left Review, 144 (1984), pp. 37–71Google Scholar, childcare and work outside the home were impossible to combine under the conditions of capitalist production.

⁷⁴ The Baluarte of 28 January 1896, quoted in Baena Luque, Las cigarreras sevillanas, p. 76.

⁷⁵ The Real Decreto of 23 April 1919 introduced the eight-hour working day. The CAT introduced the eight-hour working day from October, 1919: “[…] According to this law that […] has harmonized the uses and conditions of each town with the familial necessities of thousands of working mothers, wives and daughters […] the daily work time has been made more human in terms of average working time per day since this does not exceed 8 hours […] according to the opinion of maestras and cigar makers thus reforming the internal rules of the factory […]”: AFTS Correspondencia, Leg. 19, 23 September 1919.

⁷⁶ Cigarreras opposed to the eight-hour working day went on strike in December 1919 and January 1920. The law imposed a fixed working time and cigarreras considered they lost money because they used to arrive later in the morning. They asked for an increase in piece-rates of 50 per cent. The conflict ended when the company increased' the piece-rate to 25 per cent, even though it felt 15 per cent would have been enough: AFTS, Correspondencia, 22 January 1920.

⁷⁷ In January 1922, Seville suffered a very severe flu epidemic which affected a lot of factory workers. During this time, the factory was running at a loss, and the directors of the company issued the following rule: “To balance this deficit […], some measures will be established in order to improve punctuality in the factory: first, admission to the Factory ends at 11, which increases the time entrance band by one hour with respect to the official one. Despite the inconvenience of this rule to the cigar-maker, it will increase cigarette production by at least 10 per cent […]”: AFTS, Correspondencia, Leg. 21, 25 January 1922.

⁷⁸ Toleration of non-attendance among cigarreras in order to take care of other members of the family was reported in the 1927 rules, article 8. They received the fixed part of their wages for an maximum of eight days. This article did not apply to tabaqueros.

⁷⁹ The pure piece-work system ended in 1917, when workers started to receive a fixed portion of their daily wage.

⁸⁰ Among the cigarreras, there was a high degree of solidarity and mutual support. One cigarrera (C.M.A., interviewed in March 1994) said that she was more productive than her sister and used to help her to reach the quota every day. Cooper, P.A., Once a Cigar Maker: Men, Women and Work Culture in American Cigar Factories, 1910–1919 (Urbana, 1987)Google Scholar, on work culture between men and women in the American tobacco industry, has also encountered this solidarity among women who worked in groups: “Women created a bank so that they could share tobacco with each other out of the foreman's sight and help everyone meet quota, which reflected a collective, rather than individual, approach to shop-floor discipline” (p. 321).

⁸¹ This double use of time was sanctioned in the 1927 rules, when the company knew that the end of the manual workshops was only a question of few years. Article 40 of the 1927 rules established that: “The official number of working hours at the factory will be established as 8, which can be adapted to any special personal and town circumstances. In the manual workshops, considering that their members are mostly aged, the time of entrance to the workshops can be more flexible but, given that the exit time is fixed, they will work less than the 8 hours when they arrive later.”

⁸² Because cigarreras' absences were a real problem for management, as reflected in company records, it is possible to use the term absenteeism when referring to the cigarreras' absences.

⁸³ Through the study of the financial aspects of the companies, it is possible to know whether or not companies profited from non-attendance. See Barmby, T., Orme, C.D. and Treble, J., “Worker Absenteeism: An Analysis Using Microdata”, Discussion paper series, Centre for Economic Policy Research (London, 1990)Google Scholar, and idem, “Worker Absence Histories: A Panel Data Study”, Working paper, School of Accounting Banking and Economics (University of Wales, Bangor, 1994), who have demonstrated the potential importance of considering the financial aspects of the firm and attendance control in explaining persistent absenteeism.

⁸⁴ Nevertheless, part-time jobs continued to be performed by women. In fact, studies of household labour beginning in the 1910s and continuing through to the 1970s show that the amount of time a full-time housewife devoted to her housework remained virtually unchanged over a 50-year period, despite dramatic changes in household technology (see Folbre, N. and Wagman, B., “Counting Housework: New Estimates of Real Product in the United States. 1800–1860”, The Journal of Economic History, 53 (1993), pp. 275–288)CrossRefGoogle Scholar. Schor, J., The Overworked American. The Unexpected Decline of Leisure (New York, 1991), p. 8Google Scholar, explains that over time, new responsibilities were created, such as the need to pay more attention to the education of children and to personal care.

⁸⁵ The 1835 rules said that: “[…] workers should carry with them chairs, scissors […]: Obligaciones y Facultades del Superintendente 1835. AHFT (Archivo histórico fábrico de Sevilla), Expedientes generates, Leg. 628.

⁸⁶ Perrot, M., “De la Manufacture à l'Usine en Miettes”, Le mouvement social, 125 (1983), pp. 3–12CrossRefGoogle Scholar, esp. p. 5. The piece-work system did not demand a continuous control as did work in the textile factories because piece-work labourers enjoyed a flexible use of time. The disciplinary crisis at the beginning of the twentieth century was in part a consequence of the application of a “textile” discipline to other industries.

⁸⁷ For example, when comparing the Spanish tobacco industry with its American counterpart, in which the skilled cigar-makers were men, the influence of the labour conditions on the different use of time between women and men becomes more apparent. Cooper, Once a Cigar Maker, has shown how male tobacco workers (unionists) in the US, because of their level of skill, had a higher piece-rate and enjoyed greater freedom in their use of time than women using moulds and machinery (team work). Evans, W.D., “Effects of Mechanization in Cigar Manufacture', Monthly Labor Review, Report No. B–4 (1938), pp. 1–21Google Scholar, in his report on the American tobacco industry, writes for the Department of Labour: “[… ] hours of labour in factories making cigars by hand especially before the introduction of machines, were in general somewhat informal […] many cigar-factory employees do not work all the hours the factory is open and work afforded. The work is so largely individual in many factories that the coming and going of employees does not interfere materially with the work of others – this information is related to 1911–12 […] “ (pp. 18 19). The remuneration system is very important in explaining the absence behaviour of workers. Treble, J., “The Wages Book of Garesfield Butepit, County Durham: Data Description and Summary Statistics, June 1890 to June 1892” (Labour Economics Unit, Research Papers 90/2, University of Hull, 1990)Google Scholar, a study of the Durham mines workers, established the way in which workers were remunerated as an important determinant of absenteeism: as important as labour conditions in the different areas of work in the mines, such as humidity, depth, etc.

⁸⁸ Their house or their tasks were the terms under which women with no market occupation are recorded in twentieth-century Spanish censuses prior to the advent of democracy.

⁸⁹ Control of production in the workshop was no longer exercised by the cigarreras but by the mechanics. Comparing the models of the evolution of the sexual division of labour in the Spanish tobacco factories and those in the United States (see Prus, “Mechanization and the Gender-Based Division of Labour”), it is possible to observe that the homogenization of both models was only accomplished with complete mechanization. In the US, the skilled workers were men, and women joined the tobacco industry at the end of the nineteenth century, parallel to the application of the first semi-mechanical techniques. When factories became fully mechanized, the women took over the men's posts. As in the Spanish case, workers making cigars mechanically lost the importance in the production process possessed by the manual cigar-makers. In both tobacco industries, the new key job in the workshops was that of the mechanic, and in both this was the province of men. While in Spain control over production passed with the mechanization from women to men, in the US – even though there was a switch in duties – the transition was man to man. See Gálvez-Muñoz, “Management, Labour and Gender”.

⁹⁰ The high rent that the CAT had to pay to the state was due to the importance of fiscal monopolies, and especially the tobacco monopoly, to the Spanish Treasury. It has to be related to the insufficiency of tax resources and Spanish backwardness in refusing modem tax systems. About the fiscal monopolies, see F. Comín, “La empresa pública en la España contemporánea: formas históricas de organización y gestión”, in Comín, La empresa en la historia de Espana, pp. 349–367.

⁹¹ Esping-Andersen, G., “Decommodification and Work Absences in the Welfare State”, Working paper, IUE, no. 367 (Florence, 1988)Google Scholar, studying absenteeism in the Scandinavian countries, has demonstrated that the state promoted non-attendance more as an employer than as a legislator.

⁹² The concept of labour hoarding comprised the company strategy applied during the economic recessions, mostly among workers with specific skills. This strategy consists of maintaining more workers than production needs rather than firing them, because of the transaction costs that the company has to suffer, not only for the fired workers but more for the cost of the substitution of these workers on the labour market when the economic situation changes. This idea has its origins in the variability and uncertainty of modern economies, and in the concept of work as a quasi-fixed factor of production. This implies that managers have an incentive to consider workers as capital when they have taken on part of the training of the worker. The first person to sustain the idea of work as a quasi-fixed factor of production was Oi, W.Y., “Labor as a Quasi-Fixed Factor”, Journal of Political Economy, 70 (1962), pp. 538–555CrossRefGoogle Scholar. The Marxist concept of excess army could not be used in this model. This concept implies the existence of an excess labour population which worked as crowbar for capitalist accumulation: see Brayerman, H., Labor and Monopoly Capital (New York, 1974)Google Scholar. Nevertheless, this concept saw the worker as exploited, while the cigarreras profited in terms of time from this company strategy, which was in itself mostly an adaptation of the earlier situation.

⁹³ The fixed part was 0.25 Pts a day when it was established in 1917. It was raised to 0.5 Pts in 1918; to 1 Pts in 1919; to 2.5 Pts in 1930; and to 3 Pts in 1933.

⁹⁴ In an inspection of the workshops made on 18 May 1931, the administration made a list of cigarreras in the workshops and cigarreras taking any kind of leave. For example, in the mechanized cigarette workshops, it was calculated that the number of workers needed for production was 54, but there were 65 on the list. These 11 cigarreras were an excess of supply which smooth the functioning of the workshop when other cigar-makers did not attend. If the number in attendance exceeded 54 they were employed in other workshops: AFTS, Leg. 917.

⁹⁵ In 1919, the cigarreras' opposition to the eight-hour working day could be understood as an opposition to a general working-class demand. But on this occasion, the opposition is specifically to the cigarreras' and tabaqueros' union. The Seville section, NICOT, denounced the doctors because they refused to become members of the union.

⁹⁶ This law was promulgated on 2 July 1936. But the law came into force only when the Civil War ended and the management of all Spanish tobacco factories was reunified. That retirement was beneficial for the workers and for the company is shown by the fact that the company managers and the tobacco workers' union jointly petitioned the government for it. The company expressed itself in these terms: “This workforce is mainly aged and has imposed a constraint on technical development. We have compensated for this economically by intensive production and redistributing work spaces in order to reduce pressure on some workshops, as well as by better discipline and supervision […] Socially, we have observed moral responsibility towards those workers who have devoted all their lives to the factory”: quoted in Valdés Chápuli, La Fábrica de Tabacos de Alicante, p. 73.