3. The first nixtamal power grinding mills and ‘Tortilla Factories’ in Mexico City

The first nixtamal power grinding mills began to be established in Mexico City during the first decade of the twentieth century, and they spread increasingly during the following decades. By 1913, there were 72 nixtamal power grinding mills operating in Mexico City which employed 200 women and 85 men.Footnote ¹⁹ Eleven years later, in 1924, there were 147 mills that employed 555 workers.Footnote ²⁰ By 1930, there were already 291 mills operated by 862 workers: 451 men, 405 women, and 6 children.Footnote ²¹ By this year, the mills of Mexico City produced 20 per cent of the masa produced mechanically in the whole country, although the city only represented 7.5 per cent of the total population.Footnote ²² In contrast to mills established in rural environments that processed nixtamal brought by the public to be milled into masa for a fee, mills in Mexico City acquired sacks of corn and lime that they transformed first into nixtamal and then into masa for sale. They carried out phases 2 and 3 of the tortilla production process described in the previous section.

The establishment of nixtamal mills formed part of a more extensive process that took place in Mexico City at the turn of the twentieth century by which, as John Lear remarks, a few companies organised around large, semi-mechanised workshops increasingly dominated many crafts. This generated a high concentration of ownership and a deskilling of the labour force. In addition to corn grinding, this took place in shoemaking, baking, metalworking and printing. Overall, the effect was a drop in the relative importance of manufacturing employment in Mexico City, declining from 37 per cent in 1895 to 33 per cent in 1910.Footnote ²³

Women made up 35 per cent of the paid labour force of Mexico City in 1910, according to the population census, far above the national average of 12 per cent.Footnote ²⁴ The diffusion of the nixtamal mills displaced many of them. According to the censuses, there were 1,182 molenderas (female grinders) in Mexico City in 1895, 2,836 in 1900 and only 349 in 1910. From the 1921 census on, the occupation of molendera disappeared altogether.Footnote ²⁵ The censuses also report that in Mexico City, there were 3,763 tortilla makers (tortilleras) in 1895, 2,604 in 1900 and 1,813 in 1910.Footnote ²⁶ However, these numbers refer solely to women exclusively devoted to these tasks for commercial purposes. Before milling was mechanised, most corn was ground, and tortillas made at the household level and were thus not reported in the census as an occupation.Footnote ²⁷ Nonetheless, it is possible to estimate the number of women that must have been required to make the tortillas consumed daily in Mexico City using the calculations made by Miguel María de Azcárate in the 1830s. According to him, every person ate eight tortillas daily, which required 1.5 pounds of masa. He considered that the most efficient and hard-working woman could grind 24 pounds of corn daily.Footnote ²⁸ Since in 1900 Mexico City had 541,516 inhabitants,Footnote ²⁹ using Azcárate's assumptions, 4,332,128 tortillas would have been required to feed its population daily, which production would have required the work of 33,845 women.

Teresa Rendón and Carlos Salas observe that the vast majority of women working in the food manufacturing industry in Mexico between 1895 and 1910 were corn-grinders (molenderas). According to them, the number of women working in the food industry in the country increased between 1895 to 1910 from 247,020 to 330,791, but then declined drastically to 37,919 in 1921, and to 16,650 in 1930, which they consider could have been a result of the expansion of nixtamal mills. Rendón and Salas also suggest that the decrease in the number of women working in domestic services from 184,255 in 1910 to 131,970 in 1930 could have been related to a technological change in masa production.Footnote ³⁰ Nixtamal mills thus had a significant negative impact on women's labour participation in these occupations.

Since pre-Hispanic times, the production of masa had been an exclusive female job; once production was industrialised, it was dominated by male labour. Only a few women were employed, segregated into a limited set of activities that did not require the use of machinery. The mechanisation of the process broke the long-established social stigma against male participation in masa and tortilla production, as happened in other types of work, such as cloth-making, that were considered female labour before mechanisation.Footnote ³¹ While in 1913 women represented 70 per cent of nixtamal mills' labour force, by 1921 they were only 44 per cent. By 1930, this figure increased slightly to 47 per cent, but fell to 40 per cent in 1935.Footnote ³²

The masculinisation of masa production must have been partly a result of prejudice about the capacity of women to work with machinery, and also a result of the muscular strength such work required, given the heavy loads that needed to be carried once the masa was produced on a large scale, as the description of the process by a medical inspector of the Department of Labor during the early 1920s suggests.Footnote ³³ According to his account, to prepare nixtamal, water was poured into large buckets with a capacity of more than 1,000 liters each. A constant current of vapour heated the water to a temperature of 68 degrees centigrade. Then the workers poured into it 750 kilos of clean corn grain, together with 3 kilos of lime. The mixture had to remain in maceration for six hours at a temperature that should not reach 100 degrees centigrade. Then it was taken out, allowed to cool down and taken to another device with a sloping plane, where it was washed until the corn became entirely white and stripped of its outer shell. Finally, it was taken to the mill where it was ground into masa that was then subdivided for sale. The mill visited produced around 1,000 kilos of masa daily.

The establishment consisted of two areas, one where the corn was prepared to be ground, and another, where the mill was located and sales were carried out. The inspector considered the conditions in the first room could be damaging to the health of the workers since they experienced abrupt temperature changes. According to this account, this room had little light or air. In this room, ‘the men who worked those rude chores’ were barefoot, so they frequently experienced rheumatic problems. The other room, where the mills were located, was better lit and ventilated, and it was free of dust. It was in this room where most of the women worked, weighting and selling the masa.Footnote ³⁴ Thus, labour was segregated among the sexes, not only according to occupations but also in the workspace.

The information gathered on 86 nixtamal mills by employees from the Department of Labor in 1924 enables us to understand better the way nixtamal mills operated. These surveys were part of the project of the Workers' Census that also included other sectors.Footnote ³⁵ The mills for which information is available in this source represented over half of the total mills established by 1924 according to a report of the Department of National Statistics.Footnote ³⁶

According to these surveys, nixtamal mills were a relatively small business; they had on average a capital of 2,578 pesos (around 1,300 US dollars of the time), and they hired on average one salaried employee and eight workers (see Table 1).Footnote ³⁷ For equipment, they usually had a boiler of between 5 and 20 horsepower, large tubs where they boiled the corn, one or two mills to grind the mixture run by one or two engines of between 10 and 25 horsepower that operated with diesel or electricity. Most of them (80) were owned or managed by men, while only 7 mills were owned or managed by women. The mills run by women had relatively less capital (see Table 2).

Table 1. Nixtamal mills and tortillerías in Mexico City, 1924

Source: Archivo General de la Nación, México, Fondo del Departamento del Trabajo, Sección Estadística, Cajas 776 exp. 1 y 2 y caja 777, exp. 1 y 2, ‘1924, Distrito Federal, Censo Obrero, correspondencia y cuestionarios mensuales relativos a los molinos de nixtamal’.

Table 2. Share of establishment, employees and workers by the gender of owners or managers

Note: These numbers are smaller than those in Table 1 since they are those for which marital status was reported.

Source: See Table 1.

These mills hired 84 salaried employees, of which 67 were men and 17 were women (20 per cent) (see Table 1). The average daily wage of male salaried employees in this sample was 5.25 pesos (see Table 3), whereas that of women was 2.56 pesos, that is less than half the average wage of men. Mills also employed 654 workers, of which 339 were men (52 per cent), 304 were women (46 per cent) and 11 were children (2 per cent) (see Table 1). While the average daily wage for male workers was 2.43 pesos, women workers earned, on average, 1.73 pesos (see Table 3). Around 45 per cent of the workers of nixtamal mills belonged to the Millers Union (Sindicato de Molineros) that formed part of the largest confederation of the time, the Mexican Regional Workers' Confederation (Confederación Obrera Regional Mexicana, CROM) (see Table 1). Wages in these mills were around 4 per cent higher than in non-unionised mills (see Table 4). Nixtamal mills operated on average seven days a week and eight hours per day, although a few of them operated 12 hours per day.

Table 3. Average wages of employees and workers in mills and tortillerías, 1924

Source: See Table 1.

Table 4. Average wages of employees and workers in unionised establishments, 1924

Source: See Table 1.

In terms of marital status, around half of the male salaried employees and one-third of the male workers were married (see Table 5). In contrast, only 15 per cent of female salaried employees and 11 per cent of the female workers were married. The surveys also report that there were 23 sons or daughters of male salaried employees, and three of female salaried employees, as well as 93 sons or daughters of male workers and 39 of female workers (see Table 5). The separate registration of sons or daughters suggests that they were not hired by the mills but were helping their parents without a wage. These figures indicate that salaried employees were more likely to have daughters or sons working with them than were workers and that this took place more often among male than among female salaried employees and workers (see Table 5). This suggests that many mills operated as family businesses, and that male employees and workers had more leverage to bring their children along to work than women.

Table 5. Marital status of employees and workers of mills and tortillerías

Source: see Table 1.

In spite of the relatively small capital nixtamal mills required, and the basic technology they involved, nixtamal mills had a high concentration of ownership. By 1915, the Compañía Mexicana Molinera de Nixtamal (CMMN) owned by the Spaniards Moisés and Benjamín Solana, owned 100 of the 130 total corn mills in the city, and a few years later, a Labor Department inspector observed that the company could set the price of masa in the city at will.Footnote ³⁸ In 1924, 119 of the 147 mills operating in Mexico City (81 per cent) belonged to this company.Footnote ³⁹ Of the 88 mills visited by the Department of Labor, 39 (51 per cent) were owned by the CMMN. Thus, the percentage of firms that belonged to this company is under-represented in our sample. In the CMMN mills, around 50 per cent of the workers were unionised, and they represented 87 per cent of the unionised workers in all mills. Since the mills that belonged to this company were more highly unionised, but they are under-represented in our sample, the actual percentage of unionised workers must have been higher. The reasons behind the high concentration of mills in this company remain to be explored.

4. The first Tortillerías

Before the production of masa was mechanised, tortilla shops – tortillerías – that carried out the whole production process manually were established in urban environments such as Mexico City. Early tortillerías hired several women to make the nixtamal, grind it and make tortillas, as the account by an English traveller during the late nineteenth century cited earlier indicates.

The appearance of tortillerías in urban environments reflects the fact that there were increasingly fewer homes where a woman would spend several hours carrying out this chore, either because they had other jobs or because they had enough income to prefer using their time on other tasks, whilst not being rich enough to hire servants to make tortillas. Migration and urban expansion also made it more challenging to eat at home or to rely on a woman to bring warm tortillas together with the rest of the lunch to family members on the job, as had been customary in rural environments.

During the first decades of the twentieth century, when nixtamal mills began to spread in Mexico City, tortillerías no longer made the masa but bought it from the mills and specialised in the last part of the tortilla making process, that is from phase 4 to phase 6 of the process described in the beginning of this paper.Footnote ⁴¹ The spread of nixtamal mills in Mexico City coincided with the Mexican Revolution, which started in November 1910. Although the end date is not precise, violence began to abate by 1917. Although internal migration to Mexico City had been considerable since the late nineteenth century, and had distinguished itself from that of other areas of the country in that the sex ratio favoured women, it increased as a consequence of the Revolution.Footnote ⁴² Several villages surrounding Mexico City were affected by the war, especially in the state of Morelos – the stronghold of Emiliano Zapata – which faced severe episodes of warfare and repression.Footnote ⁴³ Many people from these villages fled to Mexico City to find refuge from the war. By 1922, it was calculated that around 100,000 people had gone to Mexico City during the revolutionary war.Footnote ⁴⁴ Thus, while the Mexican population decreased by 5.4 per cent between 1910 and 1921, the population of Mexico City increased by 25 per cent and that of Morelos decreased by 42.4 per cent. If we consider the more extended period between 1910 and 1930, for which census data are more reliable, Mexico's population increased by 9.2 per cent, that of Mexico City by 70.6 per cent and the population of Morelos decreased by 26.5 per cent.Footnote ⁴⁵ More women emigrated than men, so that between 1910 and 1930, the female population in Mexico City increased by 76.7 per cent and that of Morelos decreased by 27.1 per cent, while figures for men were 63.8 per cent and 25.9 per cent, respectively.Footnote ⁴⁶ As a result, in 1930, men represented less than 45 per cent of Mexico City's population.Footnote ⁴⁷

Many of these immigrants were women who had to find a means of survival in Mexico City. An account from a woman from Tepoztlán, Morelos, who was interviewed between 1947 and 1948 by the anthropologist Oscar Lewis when she was 40 years old, suggests that several women must have found a means of subsistence in making and selling tortillas, atole (a drink made with nixtamal dough), or tacos (a rolled tortilla filled with other food). She recalled,

This testimony provides a qualitative complement to the quantitative data that will be analysed below, data that show that women established a large number of tortillerías during this period, which in turn gave many jobs to other women. Several women found in tortilla production a way out of their difficult situation. Opening manual tortillerías and working in them can be considered a survival strategy since it allowed many women to support their households in a critical context when, as a consequence of the Mexican revolution and technological change, many had become heads of their families and found few alternative jobs. As Michael Redclift has acknowledged, surviving usually means accommodating to structural changes rather than resisting them, since resisting for too long carries the risk of not surviving.Footnote ⁴⁹ However, strategies which enable groups of people to survive their incorporation into urban, industrialised societies often imply cultural forms of resistance, through the construction of ethnic and gender identities and social relationships, which reduce the impact of total assimilation or exclusion. Manual tortillerías did not only enable many women and their households to survive physically but also helped them to remain involved in a job that had historically been theirs, for at least a while longer.

Whereas women had been largely excluded from masa production as a result of mechanisation, the shops that made and sold the finished product, the tortillerías, remained their territory. The data collected by the employees of the Department of Labor in 1924 who visited 526 tortillerías allow us to take a closer look at how tortillerías operated. Most of them, 520, produced tortillas manually, while only six were mechanised. Only ten tortillerías were integrated into a mill. The 510 tortillerías that produced tortillas manually and were not part of a mill were tiny endeavours, with an average capital of only 113 pesos, or around 4 per cent of the average capital of mills (see Table 1). Their equipment was very rudimentary, generally a charcoal stove (brasero), some metates, a scale and a table. Only around half of them hired a salaried employee, and they generally employed around five workers, mostly women, to make the tortillas called ‘tortilleras’ or ‘palmeadoras’. Some also hired a woman to weigh the tortillas and sell them to the public, a ‘pesadora’ or ‘despachadora’. Most of the workforce in the manual tortillerías was female. In 1924, they hired 264 women as salaried employees (91 per cent), and 2,281 as workers (98 per cent), while only 27 men were hired as salaried employees (9 per cent) and 34 as workers (1.5 per cent) (see Table 1).

Most women working in manual tortillerías, either as salaried employees or as workers, were single. Of those who reported their marital status, only 29 per cent of the female salaried employees and 16 per cent of the female workers in tortillerías were married (Table 5). Although they were still a minority, these shares were larger than in nixtamal mills. Four female salaried employees (2 per cent) were widows. In contrast, most male salaried employees (82 per cent) and workers (72 per cent) in tortillerías were married. Figures show that there were 67 sons or daughters of salaried employees and 428 of workers. As in mills, salaried employees were more likely to have sons or daughters working with them than workers. However, it was more likely that both groups had sons or daughters working with them than in mills (see Table 5). This suggests that more tortillerías operated as family businesses than mills. Moreover, the fact that more workers brought their sons and daughters to help them may have been a result of their lower income. A smaller share of their children are likely to have gone to school, fewer workers could afford to have someone to take care of their children at home and more of them must have required their children's help to support the family. This was facilitated by the fact that most jobs in tortillerías were paid by the piece.

Although the 1917 Constitution had stipulated that the work-shift should be limited to eight hours and the working week to six days, this was not complied with by many tortillerías. They worked on average 8.14 hours a day; six worked for 9 hours, three for 10 hours and two for 14 hours daily. Most of them worked seven days a week, with only 127 tortillerías reporting that they had a day of rest. According to Susie Porter, employers often sought exemption from Sunday rest laws by arguing that women's work was more akin to domestic duties and that women's lack of morality disqualified them from the protection of the law. Mr. Agrove, a tortillería owner who had been repeatedly found in violation of the Sunday Law, argued in his defence that since the tortilla was an item of primary consumption, as its producer, he should be exempted from the Sunday Law. As an additional excuse, he argued that his female workers were unreliable and of questionable morality, and rarely completed a continuous week of work.Footnote ⁵⁰

There was practically no unionisation in tortilla shops. Only five of the tortilla workers were unionised (0.2 per cent), including the tortillerías that worked together with a mill, as well as the mechanised tortillerías. They all worked in tortillerías that belonged to the Compañía Molinera Mexicana de Nixtamal (see Table 1). During the following years, women working in tortillerías and mills would carry out a critical struggle to become unionised and form their own unions. In Mexico City, this process lagged behind those of other cities such as Veracruz, where, by 1919, women had already organised a Union of Female Nixtamal Mill Workers (Sindicato de Obreras Molineras de Nixtamal),Footnote ⁵¹ or Guadalajara, where, in 1926, a Union of Male and Female Workers of Nixtamal Mills (Unión de Trabajadores y Trabajadoras en Molinos de Nixtamal) was formed. In 1930, women in this union decided to split away, creating the Union of Female Workers of Nixtamal Mills (Unión de Trabajadoras en Molinos de Nixtmal).Footnote ⁵² The strength of the CROM in Mexico City was likely behind this delay since studies for other sectors have shown that it marginalised women.Footnote ⁵³ A better understanding of this issue would require further research that goes beyond the scope of this article.

Most of the unmechanised tortillerías, 75 per cent, were owned or managed by women. Many must have been family businesses since one-third of them indicated that their owner, one family member or both, worked in them. Female-run manual tortillerías hired almost exclusively women and hired a larger percentage of female labour than those run by men. As with nixtamal mills, tortillerías owned or managed by women had a lower capital investment than those run by men (see Table 2). Wages in tortillerías, for salaried employees and workers, were around one-third of those earned in nixtamal mills. Moreover, salaries for female salaried employees and workers in tortillerías were about 40 per cent of those of men (see Table 3). Tortillerías were less concentrated than the mills. The CMMN owned 18 manual tortillerías that worked as separate endeavours, and three that were integrated with mills.

Only six mechanised tortillerías were operating in Mexico City by 1924, one of which belonged to the Compañía Molinera Mexicana de Nixtamal. Mechanised tortillerías were larger businesses; they had on average a capital of 2,126 pesos, very close to that of nixtamal mills. One of them indicated that it had two machines to shape and cook the tortillas that were moved by two engines of one-quarter horsepower each. Around two-thirds of these tortillerías hired one salaried employee, and on average, they hired between two and three workers. Instead of recruiting women to make the tortillas, they hired a machine operator (maquinista). One of them also had a delivery man (repartidor) and a helper. In total, they hired two salaried employees and 18 workers. Their workers' wages were similar to those of nixtamal mills, well above those of manual tortillerías (Table 3). In mechanised mills, all male and half of the female salaried employees were single, while all the workers were single. None had children or daughters with them (see Table 5).

Mechanised tortillerías were owned or managed mostly by men (67 per cent), and although their percentage of female salaried employees (50 per cent) and female workers (44 per cent) was higher than that in mills, it was lower than that in unmechanised tortillerías (see Table 1). This indicates that as more tortillerías became mechanised, the share of female owners or managers, employees and workers would diminish in the business of making tortillas.

5. The gender wage gap in tortilla production

As Susie Porter has shown, although women's participation in the Mexico City industrial workforce shifted between 1879 and 1930 from concentration in a small number of female-dominated occupations to a wide range of work in mixed-sex factories, significant gender wage gaps prevailed. Women's wages were lower than those paid to men, whether in large factories or small workshops. This was partly a result of the fact that women were generally employed in low-wage jobs while men worked in higher-paying positions. In the textile industry, for example, men were employed as weavers with an average wage of 1 peso per day in 1896, while women were employed in carding and were paid 50 centavos, or half as much. In the domestic production of shawls (rebozos) in the 1920s, men were hired as weavers and paid between 1.50 and 2 pesos a day, while women were employed as spinners and paid between 40 and 80 centavos a day, or as embroiders and paid between 75 centavos and 1.25 pesos a day. However, even when employed in the same job category, women frequently earned half of what men earned, as was the case among mattress makers.Footnote ⁵⁴ Similarly, in agricultural production women earned as wage labourers half or less of what men earned for the same work, as salaries from Oaxaca in 1907 show.Footnote ⁵⁵

Male and female wages in mills and tortillerías in Mexico City by 1924 show that a similar situation prevailed (see Table 3). Women received lower wages than men as a result of the different jobs they undertook. Occupations were defined in terms of gender, even by the name given to them in the workers' census. Thus, for example, tortilleras or palmeadoras had the female gender in the occupation's title, while peón or nixtamalero had the male gender in the title. Although there were some exceptions, there were very few occupations that were carried out by both genders. While 19 per cent of workers in ‘female occupations’ laboured in jobs with a higher pay such as encargada or cajera, 51 per cent of men worked in such occupations, such as nixtamalero or maquinista (see Table 6 and Tables A1–A3).

Table 6. Employment and wages of occupations in mills and tortillerías by gender, 1924

Notes: Male and female occupations were defined as such by the main gender.

The occupations included were those that comprised five or more workers.

Occupations considered as high pay for women were those who paid more than 1 peso per day, and for men, those who paid more than 2 pesos per day.

Source: See Table 1.

However, the wage gap was also a consequence of the way occupations were distributed among mills and tortillerías. The problem was that the share of workers in ‘male occupations’ was more abundant in the businesses with higher capital and more mechanisation, such as nixtamal mills and mechanised tortillerías, that paid higher wages, and the share of workers in ‘female occupations’ was concentrated in manual tortillerías, that paid lower wages. Moreover, most ‘female occupations’ were piece-rate jobs. According to Susie Porter, the prevalence of piece-rate work for women contributed to their low wages because after the passage of minimum wage laws, employers manipulated piece-rate wages to avoid compliance.Footnote ⁵⁶ Even in similar positions, such as salaried employees in mills, men earned wages twice that of women.

6. Conclusions

The mechanisation of the production of tortillas shows some similarities with the early development of the textile industry. As has been described in several studies, when the textile industry began its development, the spinning part of the process was the first to be mechanised, allowing many weavers to continue carrying out the manual process, with the benefit of acquiring the yarn more cheaply from the spinning mills. Very soon, however, as the weaving part of the process became mechanised as well, it caused the laying-off of many of the former weavers.Footnote ⁵⁷ In the mechanisation of tortilla production, a parallel process took place. Initially, the milling process was mechanised but not the process of shaping and cooking the tortillas. Many women continued to work in this last part of the process, and many owned or managed tortillerías. However, as mechanisation reached the tortilla-making part of the process, these businesses became increasingly controlled and operated by men.

The early development of the tortilla industry is in line with what Ester Boserup regarded as the main consequences of early industrialisation for women.Footnote ⁵⁸ Industrialisation wiped out the home industries in which most women worked, and women's participation in industrial employment declined, remaining largely on the margins in low skilled and poorly paid work. Nonetheless, it is clear that when, as a consequence of mechanisation, the production of tortillas that had historically been a woman's endeavour, became increasingly masculinised, women did not remain passive. They found ways to continue participating in this production.

The mechanisation of milling took place during the same years that the revolutionary war forced many women to become the heads of their households, and in many cases to flee violence in their hometowns and find refuge in cities. Through this process, the population of Mexico City increased as well as the share of women in it. The testimony gathered by Lewis together with the quantitative data analysed attest to the significant number of women who found a way to support themselves and their families as owners of or workers in tortilla shops. It is remarkable that so many entrepreneurial women established tortillerías, providing jobs to many more women. This can be considered a means of survival in a context where a technological change was marginalising women. It was also a means to cope with the great difficulties that had caused the revolutionary war.

The fact that there were so many women who owned or managed manual, non-integrated, tortillerías but so few who owned or managed nixtamal mills or mechanised tortillerías tells that they faced severe difficulties in raising the much higher amounts of capital that were required to establish these last two types of businesses. This could have been a result of larger restrictions on women borrowing money through formal or informal capital markets. It could have also resulted from the fact that they were less wealthy, either as a result of being less educated and able to obtain only lower-paying jobs, as a consequence of heritage traditions that placed them at a disadvantage, or because these entrepreneurial women generally belonged to less well-off families than the men who established these types of businesses.Footnote ⁵⁹ In any event, it suggests that the low participation of women as owners or managers of mills and mechanised tortillerías was not due to lack of entrepreneurship but as a result of lack of capital and education.