Family, demography and labour relations

Keywords:

• Labour relations
• demographic transition
• family systems
• female labour
• household strategies

1. Introduction

Over the past decades, demographic, economic and labour historians have given ample attention to the relationships between demographic patterns, family systems and structural economic change. In this literature, family relations are usually analysed as being at the receiving end of economic developments. For instance, early historians of proto-industrialization theorized that the increasing need for family labour in the seventeenth- and eighteenth-century home industries stimulated men and women to marry earlier, in order to reproduce labour power for the home industry (Kriedte, Medick, & Schlumbohm, 1981; Mendels, 1972). Even if many empirical studies have refuted this prediction (Saito, 2013), the debate remained centred around the question of what the demographic effects were of these economic developments (Ogilvie & Cerman, 1996, p. 228). In subsequent periods, the process of industrialization, and its accompanying developments such as the rise of (male) wage labour and the decline of the family firm (farm), led to different patterns of cohabitation, most notably the decline of multigenerational families (see e.g. Ruggles, 2015).

What is less explored is the impact of changes in family systems and demography on changing labour relations and gendered work patterns. Although recently a debate about such a causality has arisen with regard to the impact of the European marriage pattern on wage labour relations, gender, and long-run economic growth (Carmichael, de Pleijt, van Zanden, & De Moor, 2016; Dennison & Ogilvie, 2014, 2016; de Moor & van Zanden, 2010), this debate has predominantly focused on the pre-industrial period, and on Europe. Therefore, in order to explore the effects of family formation and demographic change on labour relations in a global framework, connecting pre-industrial and industrial societies, a workshop was organized at the International Institute of Social History (Amsterdam, the Netherlands) in December 2014. The workshop aimed to connect ongoing research in the fields of demography, family history and women’s work to the national distributions of self-employed, wage earners, household kin producers, etc. that have been charted worldwide in the Global Collaboratory on the History of Labour Relations.

The Global Collaboratory on the History of Labour Relations is a long-standing effort uniting labour historians across the globe in an attempt to trace and explain historical shifts in labour relations over a timeframe roughly coinciding with the rise and subsequent development of capitalism (1500–2000).¹ Labour relations are understood as the full range of vertical and horizontal social relations under which work is performed, starting from a basic subdivision in society between people who are not expected, or are unable, to work (the young, the elderly, and the infirm) as well as the unemployed, and people who do work, whether part-time or full-time, outside the home or at home, in self-employment, or as wage earners, in slavery, or as employers (Hofmeester, Lucassen, Lucassen, Stapel, & Zijdeman, 2016). The first phase of this project (2007–2012) consisted of data-mining. Data on labour relations were brought together on the basis of a shared taxonomy (see the Appendix A to this introduction for the definitions of the various labour relations) in five benchmark years (1500, 1650, 1800, 1900, 2000) for a wide range of countries across Europe, Asia, Africa, and the Americas.²

The second phase of the project sets out in search of explanations for shifts in labour relations as well as for the possible patterns observed therein. Causes and consequences of shifts in labour relations are explored by looking in depth at possible explanatory factors in a series of dedicated workshops in order to subsequently combine the insights this produces into proper multi-causal explanations. The central question that interests us here is under what conditions shifts take place, how these affected the way work was valued and compensated, and what the consequences were for the level of inequality in society (Lucassen, 2016).

The third workshop in this series examined the role of changes in family and demography in effecting shifts in labour relations. We invited experts on the USA, Latin America, Africa, Europe and Japan to reflect on demographic trends in relation to labour relations and the engagement of women in gainful employment.³ Some of these papers were already committed to publishing in other journals (e.g. de Haas, 2016; Ruggles, 2015) and others were in too early a stage of the research to be included here. Therefore, the global coverage that was presented during the workshop could unfortunately not be attained for this special section. Nevertheless, the current selection offers an interesting collection of four papers that, as we will elaborate on below, give insightful information on how the interplay between demographic and family formation factors influenced labour relations in industrializing Europe and Asia.

2. A classification of labour relations

In the set-up of the second phase of the Collaboratory project, family and demography form one cluster of the explanatory factors that were selected to explore their influence on changing labour relations.⁴ What could be the connections between changes in family formation and demography on the one hand and shifts in labour relations on the other? In order to understand what we mean by labour relations, the set-up of the Collaboratory project should first be explained. The Collaboratory begins with a comprehensive definition of work as provided by the sociologists Charles and Chris Tilly: ‘Work includes any human effort adding use value to goods and services’ (Tilly & Tilly, 1998, p. 22). To develop a new all-encompassing classification of labour relations, one necessary for long-term global comparisons, we start from the assumption that labour relations define for whom, or with whom, one works, and under which rules.

To start with the first part of this assumption: people almost never work alone. Nor do they work only for themselves; they can work for the family or the household or groups of households we coined as communities. Next to that, people can work for the polity: the state, a feudal lord or a religious institution. Finally, in a society based on production for the market people can either produce directly for the market, or indirectly via non-market institutions. All this brings us to the following taxonomy:

In order to classify the total population (column 1: ‘Point of departure’) according to this taxonomy, we applied the following logic. We stress that this taxonomy should primarily be considered a tool to characterize individuals (column 4: ‘Labour relations (individuals)’). The scheme should therefore be read from right to left and enables us to shed some light on the character of that society within a given place and period.

First, the taxonomy distinguishes between those who are able to work and those unable to work (the category non-working in our taxonomy). We start with the total population, first deciding who is not able to work and then assuming that all other people are working. This forces one to be aware of what work is (not only gainful work) and to cover the entire population, thus explicitly also taking working women and children into account. Next, column 2 (‘Type of exchange’) distinguishes between the three types of exchange in organizing the exchange of goods and services, including work. These types of exchange are linked up with the three levels of analysis listed in column 3 (‘Goal of production’), which reflect the target of production: the household and/or community, the polity, or the market. The principles under which this exchange takes place are reciprocity (work done for other members of the same household or a group of households that form a community), tribute giving (work based on obligations vis-à-vis the polity), and market exchange in which labour is ‘commodified’ (i.e. where the worker – or, in the case of unfree labour the owner of the worker – sells their means of production or the products of their work).

The second aspect of our definition of labour relations: under what rules do people work, is partially captured in column 3 as it gives the context of rules; they are expressed in more detail in column 4 where the labour relations of individuals are indicated, including the type of relation and including indentured, servile, slave and wage labour. (For the various definitions of labour relations see the Appendix) An important element of the Collaboratory approach is that it acknowledges that individuals more often than not combined different types of work and as a consequence often also different types of labour relations. Following from this, in the datasets made by Collaboratory participants working individuals are often attributed with more than one labour relation (up to a maximum of three), preferably also with an indication of the amount of time spent per year in each specific labour relation.

3. The impact of family and demography on labour relations: results from the workshop

Since our definition of the non-working population is partially based on age, the most simple connection between demography and labour relations is that a very young or old population will lead to more people in labour relation (labrel) 1. A second, more subtle connection is based on the demographic transition. One might suggest that population growth in the first stage of the demographic transition increased pressure on the land, which could lead to a shift from reciprocal to commodified labour, i.e. from self-subsistence agriculture to wage labour. Thirdly, the age and intensity of marriage could affect the level of female wage labour participation. For Western Europe and North America Steven Ruggles has shown that neither hypothesis was correct: in Europe commodified labour relations increased long before the demographic transition and were stimulated by an exogenous demand for labour, not by surplus population. For North America he showed that timing and intensity of marriage did not have a large impact on female wage labour participation. For most of the period when wage labour for women was growing, age at marriage was declining and marriage rates were increasing (Ruggles, 2014, 2015, 2016).

Ruggles also referred to the Second Demographic Transition that would have taken place in the 1960s, leading to changes in sexual and reproductive behaviour in Western Europe and North America. This would also have changed the attitudes towards married women working for wages; however, in 1970 only 19.5% of married women agreed that married women should work outside the home, while 41% of married women were already doing so. Ruggles reversed both hypotheses and argued that changes in labour relations, that is a shift to commodified labour, led to demographic changes and not vice versa. Ruggles’ research was confined to Western Europe and North America, but he states that in ‘some poor countries in the twentieth century, population growth may have sometimes encouraged the growth of wage labour’ (Ruggles 2014, p. 16). When we specify this a bit more and we look at other parts of the world and shift our focus from the macro to the more micro level, we have to take into account that people could combine labour relations. Many people worked and still work part-time in self-subsistence agriculture and part-time in (proto) industry, combining reciprocal and commodified labour relations. This way population growth could indeed lead to an increase in wage labour, for both women and men.

During the workshop Maria Camou presented a paper on shifting labour relations in Latin America in the twentieth century, showing that the situation in various Latin American countries (Argentina, Brazil, Uruguay and Bolivia) differed enormously. From 1900 to 2000, while Latin American countries underwent the demographic transition, in Argentina female wage labour participation increased most, followed by Uruguay (Camou, 2014). However, apart from the demographic transition, other factors had an impact on this shift from reciprocal to commodified labour for women, such as the modernization of industries. Also, the link between lower fertility and increasing wage labour by women was often more indirect: if women had fewer children, there were more resources for female education, which could lead to more waged job opportunities for women. In twentieth-century Latin America, as Enriqueta Camps-Cura showed in her presentation, the increase in levels of education of women led to the decline of child labour, leading to a shift in labour relations for this group from self-employment in the informal sector to non-working.

It would be interesting to see how and whether in other non-Western countries the demographic transition was linked to an increase in wage labour, especially by women. For twentieth-century Uganda, Michiel de Haas has shown that a substantial demographic increase did not lead to pressure on the available land and consequently to an increase of wage labour. Rather, smallholders who produced partially for subsistence and partially for the market switched for part of their production from food to cash crops such as coffee and cotton. Working the farmland together with other household members, combining farming tasks with household manufacturing, small-scale trading and in some cases part-time (migrant) wage labour, the smallholders as portrayed in the village surveys De Haas studied did not make a massive shift to wage labour, which was often an urban activity. The village surveys do nevertheless show a large regional variation: in areas where cash crop production was not an option an increase in wage labour migration occurred, although men preferred to work as sharecroppers on other people’s land rather than enter the wage labour market in the city (de Haas, 2016).

Whereas Ruggles and Camou both looked at the macro level, the four papers in this special section study female labour, especially in industry, from a micro-level perspective, focusing not (only) on census data but also on other types of sources that contain data on individual lives and local industries and crafts. All contributions show that women’s share in commodified labour was much larger than the censuses reveal, even if it is hard to quantify exact levels of female labour force participation. This is precisely, as particularly the Dutch and the Japanese cases show, because the household was pivotal for married women’s activities for the market, whether actually performed outside or inside the household. Scrutinizing sources that, as opposed to the censuses, gives a much more dynamic representation of work activities. By doing so, the four papers show how varieties of labour relations were combined, by individuals at the same time, or over their life course, but also how households and families could combine labour relations by allocating particular tasks to particular members of the household. In early twentieth-century Catalonia, for instance, the widespread occurrence of the stem family allowed for a particular allocation of labour, in which husbands often were farmers or worked in a particular craft or industry, wives performed wage labour in the emerging textile factories, and grandmothers cared for the children in the household (Borderías and Ferrer, this issue).

All the papers in this section also show the importance of the interplay of family formation and institutions in different societies. By institutions we mean the rules that humans create to structure the functioning of a society, and that can be both formal, in the sense of legislation and inheritance rights, and informal, in the sense of traditions, values and norms. Thus, inheritance laws and practices influenced the ways in which people decided to live together, which in turn influenced the participation of the different household members in the labour market. More directly, child labour and women’s work were affected by labour legislation, although Conchi Villar shows that households easily circumvented Franco’s restrictive legislation prohibiting married women’s labour in twentieth-century Spain. On the other hand, religious and social norms about the appropriate role for married women influenced their position in the household and in the labour market, and possibly also affected the age at which couples decided to get married. Also, Franco’s policy of family subsidies (paid for the second child under 14) prolonged the period during which women were not working for wages. In order to give more of such concrete examples, we will now briefly summarize the four articles in this special section.

Mary Louise Nagata takes a long-term approach and explores the relationship between marriage systems, family firms and labour relations in Kyoto, focusing on the seventeenth to nineteenth centuries, but with an eye for earlier developments. She highlights the particular importance of the family firm, and how different patterns of family formation and inheritance determined the continuity of the firms and the household members’ activities in them. Nagata signals various life cycle shifts in labour relations for both men and women, who often entered service in other families’ firms at a young age working (for wages) and living in-house with their employers, which could entail a degree of unfreedom. After marriage, most men worked in their own family firm as self-employed workers or employers, whereas women worked as kin-producers in these family firms, even if we do encounter some women acting as employers. The stem family was important for the continuation of the firm, and traditionally, grooms were taken in by the family of the wife, leading to uxorilocal or duolocal residence patterns. Moreover, boys and girls were both entitled to inheritance. This all led to a relatively high degree of economic leeway for Japanese women, who participated in the daily activities and at times also in the management of the firms. Over time, virilocal marriages became the norm, making it harder for women to act as employers, but older patterns also still existed in the nineteenth century. Moreover, both husband and wife kept their own property rights. These particular residence and inheritance practices, Nagata argues, have led to the remarkable resilience of the Japanese family firm up until the present.

Corinne Boter shows how norms and values about married women’s work and the ideal of domesticity were embraced by working families in Dutch industrializing regions around 1890 and how – at the same time – many women did perform gainful work, more than the censuses reveal and also more than previous scholars accounted for. Women typically combined various types of labour relations, which often were not noted down as gainful employment, such as homework for the industry (commodified labour), the cultivation of land and keeping livestock (reciprocal labour or self-employed commodified work, depending on the goal of the output) or working for the family firm (also self-employment). Households often combined these various types of work and labour relations. Norms and values concerning unmarried women’s and children’s work were less pronounced, and these groups were well represented in the statistics of factory workers. In fact, local data show that the demographic transition that was in full swing in the Netherlands at the end of the nineteenth century provided the factories with ample labour by young adolescents. Particularly among the age groups of 14–18 years old, very high participation rates (almost 70%) prevailed (Boter, this issue, van Nederveen Meerkerk, 2009).

Cristina Borderías and Llorenç Ferrer-Alos analyse the position of the female members of stem families in rural industrial areas in Catalonia from 1900 to 1936. In these areas the stem family system, based on the sole-heir system, prevailed. The eldest son inherited the family patrimony, stayed in the family home and had to take care of his parents. As a consequence, stem families consisting of grandparents, parents and children lived together, sometimes with the siblings of one of the generations if they stayed single. In the eighteenth century some 36% of all families were stem families in these areas; in the nineteenth century this was still 27%. Many of the women in these families specialized in proto-industrial textile production, whereas the men concentrated on agriculture. When mechanized textile industry moved to these areas – driven by demand for cheap labour – the women from these families, including the married women, started to work in the factories, shifting from self-employment for the market to wage labour for the market, whereas men stayed in the partly reciprocal, partly commodified agricultural sector. It was not the number of children that determined whether women worked or not, but the number of other available household members, as there were always other women around to take care of the children. Borderías and Ferrer-Alos present their work as an amendment to the Hajnal thesis: although women married late and there were many singles, the European marriage pattern with its new, nuclear families did not develop everywhere. Instead, the stem family in Catalonia remained vital and constituted an adaptive strategy to the demands for wage labour, helping the industry to develop. Their conclusion is therefore that the stem family and its female wage labour supply stimulated, instead of hampered, industrial development in this region.

Finally, Conchi Villar investigates female wage workers in Barcelona from 1930 to 1950, looking at family or household strategies, but also at the demand side of the industry that was booming in this period. At the same time Francoist formal policies discouraged married women’s work outside the home. She concludes that the ages of women, the number of children they had and more generally the composition of the household and family strategies for the distribution of productive and reproductive reciprocal work, determined whether married women worked for wages in the factories or not. Again, extended and multifamily households played a role, as they combined various types of work and labour relations including forms of labour that are usually hidden from the census such as providing paid lodgings.

4. Issues for further research

How do differences in the timing, the nature, and the speed of demographic change affect economic developments and labour market distributions? More particularly, are the global differences in the timing of the demographic transitions connected to the shifts in labour relations occurring in different regions and different periods? The papers and articles discussed above suggest that if we look at labour relations on a more micro level, using sources other than censuses, we see that many individuals, men and women, combined different labour relations, at the same time or during their life cycle. Moreover families, family firms and households combined various types of labour relations. This makes the shift in labour relations both on a micro (individual) level but also on a more macro (societal) level less abrupt than we might think. Still, we have seen examples where the demographic transition had a clear impact on women’s waged work, just like adaptations in family systems. Further research, in various parts of the world, based on sources that contain microdata on individuals, households and families can help us understand the connection between changes in family, demography and labour relations even better. These could also be used as a proxy to guestimate female labour force participation and women’s labour relations for areas where sources are really lacking.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1. The Global Collaboratory on the History of Labour Relations is based at the International Institute of Social History, Amsterdam, and financed by the Gerda Henkel Stiftung (Düsseldorf), the Netherlands Organisation for Scientific Research (NWO), and the International Institute of Social History. See https://collab.iisg.nl/web/LabourRelations/, accessed 20 May 2016.

2. For the data, please refer to https://collab.iisg.nl/web/labourrelations/results, as retrieved on 20 May 2016. For Africa a sixth cross-section for 1950 has been included in the data-mining programme.

3. For the full programme see: https://collab.iisg.nl/c/document_library/get_file?p_l_id=273223&folderId=338337&name=DLFE-202101.pdf.

4. The other factors are: the role of the state, economic institutions, geographic and social mobility.

⁵ These minimum and maximum ages are very much culturally determined. The age brackets chosen will always be indicated in the database and explained in the methodological paper.