1. Introduction

Given its relevance in generating employment, the concept of a living wage in the motor vehicle manufacturing industry emerges as a critical area of interest. According to the International ^{Organization of Motor Vehicle Manufacturers (OICA) (2022)}, Mexico has experienced a substantial increase in the generation of employment in this industry compared to the North American region, rising from 22% in 2001 to 32% in 2021. Understanding the living wage is key to improving working conditions under the new trade agreement and to counteract the precariousness of wages in recent years.

Specialized literature consistently identifies Mexico as a country with significantly lower labor costs than other economies. For example, a ^{Stanford study (2010)} ranks it as having the lowest labor costs among 18 countries analyzed. However, as ^{Hernández (2018)} argues, this comparative advantage has experienced inevitable tensions since 2008, when the global motor vehicle crisis led firms to seek cost-cutting strategies, including lower wages.

More recent research (such as that by ^{Rodríguez-Abreu and Sánchez-Peña, 2017}) has demonstrated the wage gap in the motor vehicle manufacturing industry. In 2016, a Mexican worker earned an average of USD$3.14 per hour, while their counterparts in the United States and Canada earned an average of USD$28.60 and USD$26.34, respectively. These findings highlight the need for a deeper analysis of wage dynamics in the sector and their impact on the working conditions of the workforce.

Meanwhile, ^{Covarrubias (2019)} argues that wages have experienced remarkable stagnation over the decades. According to his calculations, the daily wage has barely increased from USD$1.9 in 1994 to USD$2.3 in 2018. Similarly, ^{Moreno-Bridet al.(2021)} emphasize that wage levels in the automotive sector remain unsatisfactory, with a figure equivalent to less than twice the current minimum wage. These collective findings highlight the persistent problem of remuneration in the motor vehicle industry and the need to address it urgently.

In the current context of international discussions on wage regulations and union participation, wage levels are becoming unsustainable. Specifically, within the framework of the Treaty between Mexico, the United States and Canada (T-MEC), the persistence of low wages is referred to as an illegal practice (social dumping). According to the ^{Ministry of Economy (SE) (2020)}, this international treaty establishes a minimum wage of USD$16 per hour based on regional content requirements in order to benefit from preferential tariffs under the agreement. In addition, the T-MEC introduces provisions aimed at ensuring the full exercise of trade union rights, promoting trade union freedom and collective bargaining. Thus, these provisions aim to strengthen labor rights in Mexico, especially in an industry that shows limited wage evolution despite being crucial from an economic point of view (^{Carrillo and García-Jiménez, 2019}).

In the context of the labor dynamics of an industry of such importance for Mexico, there is an urgent need to address and conceptualize the living wage. This concept seeks to quantify the socio-economic reproductive capacity of workers in order to determine the amount necessary to meet their basic needs and guarantee a decent standard of living.

This article focuses on exploring the living wage in the motor vehicle manufacturing industry, specifically in the BMW plant in San Luis Potosí, Mexico, using the methodology proposed by ^{Anker and Anker (2017)}.¹This methodology assesses the socioeconomic reproductive capacity of workers at the first wage level (according to their collective contract). By comparing the living wage with the contractual wage, the minimum wage, and the income associated with extreme urban poverty and urban poverty, the necessary wage adjustment is determined in order to achieve a better level of economic welfare for the sector's working population.

The study is divided into four sections: it begins with an overview of the living wage concept and its application in the Mexican motor vehicle industry. Next, the methodology used is detailed, specifying the values and indicators considered. The results obtained are then presented and discussed. Finally, conclusions and suggestions for wage policy are given.

2. Living wage in the mexican motor vehicle industry

The concept of a living wage has been the subject of various definitions and methodologies, focusing on determining the adequate cost of living for workers and their families. According to ^{Anker (2006)}, it involves identifying a wage threshold that allows for a decent standard of living. Its calculation is based on the income of a standard workweek, avoiding dependency on overtime and servitude-like situations (^{Guziet al., 2022}). Historically, this concept has been widely debated in industrialized countries during the last century (^{Anker, 2011}; ^{Figart, 2004}; ^{Stabile, 2008}). In Anglo-Saxon literature, it is known as the living wage, while in Spanish, it is known as “salario digno” (^{Anker and Anker, 2017}).

^{Glickman (1997)} defines it as an income that ensures self-respect, family dignity, and civic participation. ^{Richardset al.(2008)} argue that it should provide economic security and prevent poverty. Meanwhile, ^{King and Waldegrave (2012)} define it as essential to cover basic needs. ^{Glasmeier (2022)} sees it as a tool to meet family needs and promote financial independence.

The ^{Global Living Wage Coalition (GLWC) (2023)} defines the living wage as compensation for a standard workweek that ensures a decent standard of living for the worker and their family. ^{Anker and Anker's (2017)} proposal is widely accepted and includes food, housing, essential family expenses and a margin for emergencies, resulting in a basic but decent standard of living.

In addition, some terms are closely related to the concept of a living wage and facilitate its calculation and understanding. One of the most important is the poverty line, which ^{Anker (2006)} defines as the income required for a household to cover both a nutritious diet and other non-food needs. However, its value varies according to the national context in which it is applied.

Other relevant terms include minimum wage and subsistence wage. While the former is a legal requirement, the latter is a minimum income to cover basic needs. In contrast, the living wage, while desirable, is not mandatory and its adoption is voluntary (^{Guziet al., 2022}).

These concepts seek to establish a wage threshold based on national and international criteria (^{Mateeret al., 2020}). Practical examples include municipal living wage regulations in the United States, where the poverty line is used as a reference for calculating living wages (^{ACORN, 2003}).

The novelty and relevance of the living wage concept and calculation methodology have led to its application in various international contexts. In Vancouver, Canada, for example, a comprehensive measurement was conducted that included aspects such as food, clothing, housing, transportation, childcare, medical services, education, emergency funds, and other essential expenses (^{Richardset al., 2008}).

In Asia, the 2009Asian Floorcampaign for garment workers included elements such as the daily cost of food, health expenses, housing, clothing, childcare, transportation, fuel and education (^{Merk, 2009}). In New Zealand, a study detailed expenses for a family (two adults and two children) considering adequate housing, food, clothing, education and social participation (^{King and Waldegrave, 2012}). In the United States, it was based on family size, composition, and geographic location (^{Glasmeier, 2022}).

In the Mexican context, specifically in relation to the motor vehicle manufacturing industry, the estimation of the living wage has been addressed in previous research, with the study by ^{García-Jiménezet al.(2021)} being one of the pioneers in this sector. Said research adopted the methodology proposed by ^{Anker and Anker (2017)}, adapting it to regions with a significant presence of motor vehicle manufacturing companies, such as Aguascalientes, the State of Mexico, Puebla, San Luis Potosí and Sonora. Variables such as the cost of a model diet, housing, health, education, transportation and emergency funds were taken into consideration to determine the living wage. These values were combined and compared with the workers' contractual wages, revealing significant discrepancies between the contractual wage and the living wage.

One of the most important findings was that wage gaps were smaller in companies located in regions with a longer history in the motor vehicle manufacturing industry. In contrast, companies in areas with less time in the sector showed more significant differences. It is important to note that the detailed results of this research will not be discussed in this section as they will be addressed in section four, which serves as a point of comparison due to the use of a similar methodology.

3. Methodology of the living wage

This article will explore the concept of a living wage, defined by ^{Anker and Anker (2017, p. 270)} as:

the monthly remuneration that a worker receives for eight hours of work, which must be sufficient to guarantee a decent and dignified standard of living for their family, including food, housing, education, medical care, transportation, clothing and provisions for eventualities, among other essential needs.

The methodology used is based on standards specific to the Mexican state of San Luis Potosí, with complementary national and international indicators. Visits were made to grocery stores to determine the cost of essential products, as well as interviews² with operators and their families in July 2022 to understand their eating habits and other relevant aspects, ensuring a contextualized approach to regional standards.

The average household size in the state was first determined using data from ^{INEGI's Population and Housing Census (2020)} in order to calculate the living wage. The Anker methodology uses two approaches: the weighted average household size and the adjusted fertility rate. These values were then averaged in order to obtain an adjusted estimate of household size for the state in question.

Households with two to seven members were used to calculate the weighted average size. The resulting weighted average household size was 3.69. The adjusted fertility rate is obtained by applying the following formula:

The Adjusted Fertility Rate (TFA) is equal to the State Fertility Rate (TFR) multiplied by (1 - under-5 mortality rate in the federal state (TMR)) + 2

The reference mortality rate is the number of infants who died before the age of 5, while the fertility rate represents the number of live births per 1,000 women. Both measures were taken from the ^{National Population Council (CONAPO) (2020)}. The average household size based on the adjusted fertility rate was 3.84.

Following the above procedure, the reference household size for the calculation of the living wage (3.74) was obtained as the average between the weighted average household size (3.69) and the adjusted fertility rate (3.84). This value allows us to combine both measures and obtain a household size adjusted to the demographic characteristics of San Luis Potosí, Mexico (see Table 1).

In the second step of the methodology, we proceeded to estimate the number of permanent workers in the reference household size for San Luis Potosí (3.74). To do so, we first calculated the probability that a person between the ages of 20 and 59 is a full-time worker using the formula presented in the methodology of ^{Anker and Anker (2017)}:

Labor force participation rate for workers between 20 and 59 years old multiplied by (1 - the labor force unemployment rate for persons between 20 and 59 years old) multiplied by (1 - 0.5 multiplied by the partial employment rate for workers between 20 and 59 years old)

The labor force participation rate for persons between 20 and 59 years old is obtained by dividing the economically active population (PEA) by the total population of persons in that age group. Meanwhile, the unemployment rate for the same group is calculated by dividing the total number of unemployed persons by the PEA. The part-time employment rate for individuals between 20 and 59 years old is calculated by dividing the part-time working population by the total working population who work between 1 and 45 hours per week. For this study, part-time workers are defined as those working between 1 and 34 hours per week, while full-time workers work between 35 and 45 hours per week. The information for these indicators was taken from the ^{National Survey on Employment and Occupation (ENOE, 2022)}, corresponding to the second quarter of the same year.³

The probability that a person between 20 and 59 years old is a full-time worker is 0.59. Following ^{Anker's (2017)} methodology, this value is increased by 1 to obtain the number of people working in the reference household size, which results in 1.59.

In the third methodological phase, we proceeded to calculate the cost of a dietary model that was both nutritious and inexpensive. The foods in the basic food basket were selected based on the recommendations of the following sources:1)the list of foods indicated by the ^{National Council for the Evaluation of Social Development Policy (CONEVAL) (2022)} for urban communities, and2)the local dietary habits of the population of San Luis Potosí, obtained from interviews with the families of BMW workers.

The dietary model proposed by the ^{CONEVAL (2022)} for urban communities was used as a reference point in the first stage. Interviews were then conducted with workers and their families in order to develop a dietary model adapted to the eating habits of San Luis Potosi. In addition, the ^{U.S. Department of Agriculture (USDA) (2022)} database was used to determine the amount of calories, proteins, fats, and carbohydrates present in each food item included in both the dietary model adapted to local eating habits and the diet recommended by the ^{CONEVAL (2022)}. This information was used to adjust the amount of grams consumed per day and to calculate the calories needed for a family.

The calculation of the number of calories required for the reference family size (3.74 members) is based on several fundamental parameters of the Anker methodology:

Based on the above values and applying Anker's methodology, it was determined that each person in the reference household requires a total of 2,282 calories per day (see Table 2). However, since the dietary model created based on data from the ^{CONEVAL (2022)} and local eating habits provide a lower level of calories (2,130 vs. 2,282), the number of grams consumed was adjusted to obtain the calories required by each person in the reference family size according to their level of physical activity and their protein, fat and carbohydrate needs (see Table 3).

The cost was estimated once the dietary model (basic food basket) was established. For this purpose, visits were made to places where BMW employees do their shopping, mainly supermarkets and convenience stores (seven in total), in order to investigate the prices of the foods selected for the model. Based on this information, average prices were calculated and those whose coefficient of variation was less than 1 were used.⁴ The estimated costs associated with the basic food basket, adjusted to the average family size in San Luis Potosi (3.74 members) and the grams consumed per day adjusted to the calories required (2,282) according to the proportions of proteins, fats and carbohydrates necessary for physical activity, are detailed in Table 4.

In the fourth phase of the methodology, we proceeded to estimate the costs associated with housing in San Luis Potosi. Various regulations and legal guidelines were considered. The cost of adequate housing was defined based on the technical specifications of social housing provided by the ^{National Fund for Affordable Housing (FONHAPO) (2014)}. For this article, the characteristics of social housing are equivalent to our designation of "decent housing". The price of social interest housing was studied by means of interviews with workers living in them in order to estimate the cost. However, it was observed that most of the dwellings visited did not comply with the parameters established by the ^{FONHAPO (2014)}. For this reason, it was decided to use the cost of housing as the one reported in the average rental cost in San Luis Potosí, according to the National Household Income and Expenditure Survey (ENIGH) (^{INEGI, 2018}), which is set at MXN$2,467.

In the fifth stage of the methodology, we proceeded to calculate the costs related to health, education and transportation, i.e., those not associated with food or housing (No Food No House, NFNH). For this purpose, the ENIGH (^{INEGI, 2018}) was used and adjustments were made based on direct observations during fieldwork. The data corresponding to the third income quintile was extracted due to the fact that the operators' wages exceeded the levels of the first and second quintiles, consistent with an average wage in the Metropolitan Zone of San Luis Potosí. The procedure was as follows:

As a result, the proportion of expenditures unrelated to NFNH was determined to be 61.20%. Dividing this total by the adjusted food share yielded an adjusted NFNH/F ratio of 2.02. By multiplying this amount by the food model expenditure (MXN$3,953.77), a preliminary non-NFNH expenditure figure of MXN$8,019.67 was obtained (see Table 5).

To determine the costs associated with health, education and transport, the following procedure was used based on the methodology of ^{Anker and Anker (2017)}:

To determine the total living wage, the costs associated with each expenditure category were combined for the reference family (3.74 members) and 1.59 full-time workers per family, following the methodology proposed by ^{Anker and Anker (2017)}:

When these components are added together, the total monthly cost of maintaining a basic but adequate standard of living for the reference family is MXN$22,030.83 (row 5, Table 6).

To determine the net monthly living wage per worker, the total monthly costs (MXN$22,030.83) were divided by the average number of workers per household (1.59), resulting in a net monthly living wage of MXN$13,855.87 (row 6, Table 6), which is the amount needed to cover the total living costs of the reference family.

Mandatory deductions and taxes associated with the wage must be taken into consideration to determine the gross monthly living wage. According to the proposed methodology:

Adding the net monthly living wage to the percentage corresponding to income tax and social security deductions gives the gross monthly living wage. This amount is MXN$16,165 (line 8, Table 6), which is the total wage before deductions and tax obligations.

To obtain an accurate representation of the living wage that a first-level operator should receive at BMW's San Luis Potosi plant, certain benefits and allowances provided by the company must be considered:

After considering these deductions, the operator's net and gross living wage is obtained, considering typical in-kind benefits and cash allowances provided by the company. These amounts are MXN$12,848 and MXN$15,157, respectively (rows 10 and 11, Table 6).

4. Results and discussion

The analysis reveals a significant discrepancy between the contractual wage and the living wage for a family of 3.74 members. According to BMW's collective agreement, the contractual wage for the first level is set at MXN$306 per day (MXN$9,307 per month, calculated by multiplying 306 by 365/12), while the living wage is MXN$13,856, a difference of 49%. This demonstrates the need to re-evaluate the wage structures at the BMW San Luis Potosi plant to ensure a living wage.

When taxes and social security contributions are considered, the gap between the contractual wage and the living wage widens. The wage received is MXN$9,966, while the living wage, which includes these factors, is MXN$16,165. This 62% difference illustrates the challenge for operators and their families to achieve a decent standard of living according to the criteria established in this analysis (see Table 7).

When compared with other indicators (see Table 8), the contractual wage is 47% higher than the minimum wage in Mexico, MXN$9,966.17 versus MXN$5,258.13 (172.87 multiplied by 365/12). With respect to the urban extreme poverty wage threshold, the contractual wage is 48% higher at MXN$9,966.17 compared with MXN$5,139.35. Nevertheless, compared to the urban poverty wage, workers' wages are 4% lower (MXN$9,966.17 compared to MXN$10,327.32).5

Meanwhile, the discrepancy between the urban extreme poverty line wage (MXN$5,139.35) and the urban poverty line wage (MXN$10,327.32) with respect to the living wage (MXN$16,165) estimated in this exercise is 214% and 56%, respectively. These differences may be associated with the fact that, in the exercise conducted in this article, it is clear that the nutritional content of the dietary model considered by CONEVAL is below the macronutrients necessary for the physical activity of workers in BMW, as established by NOM-037-SSA-2012 (^{Diario Oficial de la Federación(2012, July 13)}).

Another possible source of discrepancy lies in the prices used by CONEVAL to establish the poverty lines since they are based on the INPC published by ^{INEGI (2022)}. However, these prices do not necessarily reflect the real cost of goods and services purchased by workers in San Luis Potosi. Local market conditions and the consumption habits of the population in the region may differ from the national averages used in the index, which could lead to discrepancies in the estimation of poverty lines.

These results highlight the urgent need to adjust the nutritional content of the basic food basket, as well as the prices of food and the set of goods and services considered in the CONEVAL methodology.

Although the contractual wage does not represent a living wage, it exceeds the official minimum wage and urban extreme poverty (see Figure 1). However, a wage increase is still needed to overcome urban poverty.

Source: prepared by the authors based on the adaptation of the methodology of^{Anker and Anker (2017)}.

Figure 1 Comparison of living wage vs. paid wage at BMW San Luis Potosi plant, 2022 

When comparing the contractual wage with the living wage of the BMW plant in San Luis Potosí with other Original Equipment Manufacturers (OEM) in Mexico, according to ^{García-Jiménezet al.(2021)}, it can be seen that the company in question occupies an outstanding position (4th place). The Volkswagen-Puebla plant leads the ranking, exceeding the living wage by +9%, followed by Chrysler/Fiat-Toluca, with a difference of -9% below the living wage. In third place is Nissan A1-Aguascalientes with a difference of -46.5%. BMW San Luis Potosí is in fourth place with a difference of -49%. Below BMW are companies such as Nissan A2-Aguascalientes with a difference of -69.1%, Ford-Hermosillo with -70%, and General Motors-San Luis Potosí, which has the largest difference with -90%. This data suggests that BMW San Luis Potosí has similar wage differentials to companies established in Mexico between 1982 and 1993, surpassing others that arrived recently (see Table 9).

The estimated wage gaps in Table 9 reveal a paradox between General Motors and BMW. Although they share a geographical location, General Motors has a -90% living wage difference, while BMW has a -49% living wage difference.

However, the evidence presented suggests that the wage gap between the paid wage and the living wage increases when motor vehicle manufacturing companies arrived more recently in the country, suggesting greater precariousness with trade liberalization.

The wage gap between General Motors and BMW can be attributed to several factors. Intense competition among workers for jobs, driven by the entry of new companies such as GM, can put downward pressure on wages. In addition, these new firms may not comply with established labor and wage standards, resulting in less favorable conditions for workers. Trade liberalization has facilitated company mobility and labor flexibility, allowing wages to adjust to market conditions, which can lead to further wage insecurity, especially if labor regulations are not strong enough to protect workers. In summary, the widening wage gap is due to the time companies have been operating and factors such as labor competition, the commitment of companies to labor standards and flexibility due to trade liberalization.

5. Conclusions

The study clearly shows that although the contractual wages are competitive and exceed the minimum wage and extreme urban poverty, they still do not reach the level necessary to guarantee a dignified standard of living. This finding has implications for wage policy in the motor vehicle industry, which is crucial to Mexico's growth and investment.

The methodology based on ^{Anker and Anker (2017)} provides a robust tool for measuring the living wage and a guide for wage policies that improve the living conditions of workers in the motor vehicle manufacturing sector. It is essential that companies, both established and emerging in Mexico, consider these findings and adopt wages that meet the needs of their workers.

Implementing the T-MEC offers an opportunity to improve working conditions in Mexico by promoting union participation and bargaining. Workers need to be involved in wage discussions, especially in the presence of transnational corporations such as BMW, which have raised the cost of living in cities such as San Luis Potosi.

The study suggests extending the methodology to the auto parts sector, offering opportunities for future research and a more complete understanding of wage dynamics throughout the automotive supply chain, facilitating more effective policies to benefit a greater number of workers.

In summary, the T-MEC makes the discussion of the living wage critical. The labor clauses in the agreement could encourage a transition to fairer wages in the motor vehicle industry. Companies and unions must proactively improve wage conditions to ensure a fair future for workers in the sector.

Policy proposals based on this article and research on living wages (^{García-Jiménezet al., 2021} and ²⁰²²) in the motor vehicle industry are as follows:

These proposals are designed to ensure that workers in the motor vehicle industry move closer to receiving decent, fair and adequate compensation that reflects their contribution to the sector's productivity and allows them and their families to improve their well-being.