2.1 The automotive and textile industries in Mexico

To understand the impact of international trade on economic activity and employment, it is essential to examine the evolution of GVCs, which refers to the cross-border movement of goods, investment, services, knowledge and individuals associated with an international production network (Amador and Cabral 2015). The study of GVCs allows for a better comprehension of the geographical dispersion of economic activities, the agents involved and the tasks performed in these processes by both developed and developing economies (Meyer-Stamer and Wältring 2007). Perhaps the most significant development in GVC analysis has been the availability of quantitative information that distinguishes between trade flows in value-added terms. Notably, the generation of global input–output matrices, when combined with Leontief’s (1936) inverse matrix, has made it possible to disaggregate export data and concepts that, in trade statistics, erroneously count the value of intermediate goods more than once.

By now, many investigations have documented how much of a country’s trade corresponds to production fragmentation or GVC production. Several authors have put forward methodologies for this analysis (Hummels, Ishii and Yi 2001; Johnson and Noguera 2012; Stehrer 2012; Koopman, Wang and Wei 2014; Arto et al. 2015). In this study, we follow the methodology developed by Wang, Wei and Zhu (2013), which includes a full decomposition of export data at bilateral and sector levels into four categories: domestic value-added (DVA); foreign value-added (FVA); value-added re-exported to home country; and pure double counting. DVA corresponds to the value generated in the country of analysis (in our case, Mexico) and can refer to final or intermediate goods further processed in destination countries. FVA, meanwhile, originates in a foreign country and is then “returned” to the country of analysis as imports of either intermediate or final goods. Following Koopman et al. (2011), who combine DVA and FVA, we use both categories to measure the degree to which countries participate in shared production. DVA measures the country’s participation as a “supplier” of value-added, whereas FVA captures the country’s position as a “user” of value-added incorporated from other countries. Both indicators are measured as a proportion of exports. We use these indicators to focus on two features: Mexico’s relevance in the global trade network for the two selected industries, and Mexico’s level of participation in this type of trade relative to its total exports.

To that end, we relied on the 2014 release of the World Input-Output Database (WIOD) (Timmer et al. 2015). This is a database of inter-regional input–output tables,12 which contains information on 40 countries13 and serves as a model for the rest of the world for the period between 1995 and 2011. The database comprises 35 sectors, following the International Standard Industrial Classification of All Economic Activities Revision 3 (ISIC Rev.3), of which 14 correspond to manufacturing. We concentrated on two industries (textiles and textile products, and transport equipment) and their performance across four years (1996, 2001, 2008 and 2011, this being the most recent year for which data are available).14 It should be noted that the data provided by WIOD are not further disaggregated.

In ISIC Rev.3, transport equipment covers other types of transport in addition to automobiles, such as ships and boats, railway and tramway locomotives, aircraft and spaceships. However, according to INEGI (2020), in Mexico the automotive industry represented, on average, 93 per cent of the GDP of the transport equipment sector throughout the years covered in the study. Moreover, according to data on firms that are mainly exporters, the value added in the automotive industry represented, on average, 98 per cent of the total value added in the transport equipment sector between 2003 and 2011 (INEGI 2018).

The decomposition approach used by Wang, Wei and Zhu identifies the bilateral flow of value-added between a pair of countries for each sector, which allows us to present this interchange as a global network of value-added. This type of analysis highlights the relationship among countries as well as the relative importance of individual countries with regard to the whole ensemble of participants. In the network graph, each country (or group of countries, as in the EU15 and EU13 (see footnote 17)) represents a node, and the domestic value-added exported from county $i$ to country $j$ ($w_{ij}$) is captured by the line linking both countries. Thus, the resulting network is directed and weighted, because it incorporates the direction and intensity of commercial relationships between countries. We use the following filter: if $w_{ij}<0.01 \mathrm{p}\mathrm{e}\mathrm{r}\mathrm{ }\mathrm{c}\mathrm{e}\mathrm{n}\mathrm{t}$of the total DVA in exports, then there is no link in the graph. The size of the nodes represents the weighted “outdegree”15 or the sum of DVA that the country exports to other countries. The colour shows the weighted “in-degree”16 or the sum of DVA received from other countries. Finally, the thickness of the line captures the magnitude of the DVA exchanged between the two countries. In Mexico’s case, the outgoing lines are purple, and the incoming lines are blue.

Figure 1.a shows the global network of DVA in textile exports for 1996. At that time, the EU15 group was the leading exporter and importer of DVA, mainly because of internal trade within the EU region, and China was the second significant exporter of DVA, trading with nearly all the economies included in the graph. The United States, meanwhile, was the leading importer of DVA. The picture changed drastically by 2011 (Figure 1.b), however, by which point China was the principal exporter within the network; its outgoing lines are considerably thicker than those for the other countries in the graph, and those directed towards the EU15 and the United States are particularly important. In 1996, Mexico exported DVA primarily to the USA and, to a lesser extent, to Canada, the EU and Brazil. Mexico also received DVA from the United States, as well as Taiwan Province of China, Canada, China, Australia, the Republic of Korea and the EU15 (in descending order of importance). Mexico’s position in the global network of DVA in textiles was concentrated in trade with the United States. By 2011, however, even its linkages with the United States had diminished in favour of trade with China.

China’s entry into the World Trade Organization in 2001, the termination of the Multifibre Arrangement in 2005 (thereby eliminating the associated quotas) and the long-running preparations undertaken by Asian countries – and especially China – with a view to liberalizing trade created the conditions for the impressive growth in global Chinese exports, in particular to the United States. In general terms, Mexico has been displaced by China as the main supplier to the United States since 2002. The competition between China and Latin American countries such as Mexico lies not only in the low cost of the labour force, but also in the ability to generate essential economies of scale, low-cost inputs (in many cases produced internally), increasing technological advances, financing and experience of full package processes, rather than their proximity to the markets or the ability to respond quickly to changes in demand and buyers (Cárdenas Castro and Dussel Peters 2007).

Figure 1. DVA in the global export network for textiles, 1996 and 2011

Note: For the country codes, see Annex 2.

Source: Own elaboration based on WIOD data.

Figure 2 shows the global network for DVA in transport equipment. Although the EU15, Japan and the United States were the most relevant actors in this network, Canada and Mexico also played a significant role. Particularly meaningful are the connections between Japan and the United States and between Canada and the United States, as shown by the thickness of the lines linking them. By 2011, it is noticeable that, unlike the textile network, these linkages diversified and more countries participated in the network, as shown by the thickness of the lines. In this sense, the graph reflects the spatial readjustments experienced in the automotive industry between the 1990s and 2010, when the presence of the former main actors declined, while Brazil, India, Mexico, the Republic of Korea and, most importantly, China increased their participation, with the latter ultimately becoming the leading global automobile producer. Nonetheless, companies based in Japan, the United States and the EU15 (such as Toyota, General Motors, Volkswagen, Ford and Hyundai) continue to dominate (Dussel Peters and Gallagher 2013).

Figure 2 reflects these trends. While the most important actors remained the same, the linkages between them became thinner, and those between other countries grew larger in comparison with their position in 1996. One such country is Mexico; all its connections are thicker by 2011, especially that with the United States. In the automotive industry, Mexico exports more DVA than it receives from other countries, mainly because of its exports to the United States. In fact, the automotive industry has the largest trade surplus in the Mexican manufacturing sector for the whole period, amounting to US$ 88.867 billion in 2019 (Banxico 2020). Unlike the textile industry, the Mexican automotive industry is at an advanced stage of integration with the United States, to such a degree that some authors refer to it as a regional auto parts/automobile chain with regional models, inputs, products and processes (Dussel Peters and Gallagher 2013).

Figure 2. DVA in the global export network for transport equipment, 1996 and 2011

Source: Own elaboration based on WIOD data.

While it is crucial to account for the role played by Mexico in the network for each of these two industries, we also need to identify how Mexico participates in the related GVCs. One way to do this is by looking at FVA as a percentage of exports. Plotting the FVA as a proportion of each country’s gross exports, as in Figure 3, reveals interesting patterns. Almost all countries saw this proportion increase between 1996 and 2011, except for Canada. Furthermore, it appears that this proportion is higher for emerging economies such as Mexico, Taiwan Province of China and the Republic of Korea than for advanced economies such as the United States and Japan. Brazil stands out for its low percentage of FVA, reflecting its reliance on primary product exports, which are naturally high in local content. For all countries, the transport equipment sector saw a higher level of FVA than the textile industry. The changes in the proportion of FVA between 1996 and 2011 are also more substantial in the transport equipment sector.

Of this group of economies, Mexico presented the largest share of FVA in manufacturing exports and did not experience significant changes between 1996 and 2011. Typically, emerging countries with large shares of exports in segmented industries need more imported inputs in order to generate exports, leading to high levels of FVA, as is Mexico’s case. In the textile and transport equipment industries, Mexico had the second highest level of FVA as a percentage of exports, after the EU13. This indicator may reflect how countries participate in the network, with those that incorporate a higher level of FVA in their exports possibly being located in the assembly phase of the chain, meaning that they need to import more in order to export, while countries with a lower proportion of FVA may be located in, for example, the design phase of the chain, in particular in the automotive industry.

Figure 3. FVA as a percentage of exports in the overall manufacturing sector and in the textile and transport equipment industries, 1996–2011, selected countries (per cent)

Source: Own elaboration based on WIOD data.

However, to fully understand a country’s involvement in GVCs, we also need to consider DVA, in particular that corresponding to exports of intermediate goods used by third countries, referred to as $\mathrm{D}\mathrm{V}{\mathrm{A}}_{\mathrm{G}\mathrm{V}\mathrm{C}}$.17 The higher the level of FVA as a percentage of exports and the higher the level of $\mathrm{D}\mathrm{V}{\mathrm{A}}_{\mathrm{G}\mathrm{V}\mathrm{C}}$, the higher the level of participation of a given country in the GVC (Taglioni and Winkler 2016). However, this does not mean that all countries present high levels for both indicators. In reality, the nature of a country’s participation in a GVC varies depending on the type of goods exported and their place in the global production process. In the case of textiles or other technologically simple consumer goods, their value depends on their level of sophistication, which translates (or not) into a trademark. In the case of highly sophisticated products, the company that owns the brand appropriates a significant part of its value. Were the production process to be fragmented internationally, the assembly country would take only the part of the value that corresponded to wages, while the owner of the brand would appropriate the value of the design, production organization, distribution and marketing.

Automotive products, on the other hand, are classified as higher technology products, for which the production process is broken down into several phases. At the head of the process is a company in charge of the most complicated phases, which allows it to generate and appropriate a large part of the value of the goods produced. A country’s position in the trade network and the division of the value of exports will depend on the phases in which the country specializes. If it specializes in the assembly of parts produced in other countries for which a highly skilled workforce is not required, the value of the exported product will have a high imported component (meaning that the internal incorporated DVA will be low).

Figure 4. Participation in GVC by FVA and DVA_GVC as a percentage of exports, selected countries (per cent)

Source: Own elaboration based on WIOD data.

Figure 4 contains data on FVA and $\mathrm{D}\mathrm{V}{\mathrm{A}}_{\mathrm{G}\mathrm{V}\mathrm{C}}$ as a percentage of exports for the overall manufacturing sector and for the textile and transport equipment industries specifically. In general, we can observe that countries present either a high FVA or a high $\mathrm{D}\mathrm{V}{\mathrm{A}}_{\mathrm{G}\mathrm{V}\mathrm{C}}$, but generally not both. Countries with high FVA present a lower level of $\mathrm{D}\mathrm{V}{\mathrm{A}}_{\mathrm{G}\mathrm{V}\mathrm{C}}$, signalling that they might be specialized in assembly. Conversely, those with lower FVA have a higher level of $\mathrm{D}\mathrm{V}{\mathrm{A}}_{\mathrm{G}\mathrm{V}\mathrm{C}}$. There are some exceptions; for example, Taiwan Province of China has a high level of $\mathrm{D}\mathrm{V}{\mathrm{A}}_{\mathrm{G}\mathrm{V}\mathrm{C}}$ for the transport equipment industry, but its level of FVA is not significantly low. However, Taiwan Province of China is a minor player in that network. We observe a slightly different linear trend in the textile industry; while negative, the linear slope is less steep than that of the other two categories, indicating that, on average, countries present a higher level of $\mathrm{D}\mathrm{V}{\mathrm{A}}_{\mathrm{G}\mathrm{V}\mathrm{C}}$in this industry.

Mexico’s position is similar across all three categories. It registers a high level of FVA and a low level of $\mathrm{D}\mathrm{V}{\mathrm{A}}_{\mathrm{G}\mathrm{V}\mathrm{C}}$, although the latter did increase between 1996 and 2011. The $\mathrm{D}\mathrm{V}{\mathrm{A}}_{\mathrm{G}\mathrm{V}\mathrm{C}}$ incorporated into its textile exports is lower than that incorporated into its transport equipment exports and its total manufacturing exports. This might be due to the fact that the Mexican textile industry relies more heavily on the import of parts and components, and it exports a higher proportion of finished products (assembled garments) than the Mexican automotive industry (the leading supplier of auto bodies and parts, including chassis, electrical components and engines, for the United States), even though it specializes mainly in assembly and therefore has also a high proportion of FVA (Dussel Peters and Gallagher 2013).

Mexico plays a relatively unimportant role in the DVA trade network for global manufacturing exports, with the exception of its link to the United States, which is one of the three most important actors in the network. Mexico’s trade relationship is heavily concentrated towards its northern neighbour, although, during the period of analysis (1996–2011), a few other countries gained a greater presence, most noticeably China and European countries.

Mexico’s textile industry – like that of many other countries – suffered a decline over this period as Chinese textile manufacturing boomed and became the major global exporter, while Mexico reduced its number of trade partners and the magnitude of value-added traded with them. The picture for the transport equipment sector was entirely different, however. Unlike the other networks analysed, more countries participate in the GVC for transport equipment. In most cases during the period of analysis, the connections became thinner, revealing greater diversification. Nevertheless, between 1996 and 2011, Mexico’s importance in the network grew, reflected in the number of links with other countries and the relative weight of those connections. Unlike in the overall manufacturing sector and the textile industry specifically, Mexico’s “exporter” role is more important than its role as an importer.

2.2 Jobs related to trade

Having analysed the trade performance of the Mexican manufacturing industry, we will now proceed to address the issue of employment with the aim of quantifying how many jobs were created by, among other things, Mexico’s increased participation in foreign trade, both in the case of final exports and trade in GVCs.

Figure 5 shows the proportion of employment provided by each industry. In 1996, the service sector provided 61 per cent of total employment, while agriculture and mining provided 20.5 per cent and manufacturing 18.2 per cent. The textile industry was the second biggest employer in the manufacturing sector, with a share of 2.5 per cent, while transport equipment was fourth with 2 per cent. By 2011, the service sector had increased its share of total employment, reaching 68.8 per cent, implying a reduction in both the primary and the manufacturing sectors, which dropped to 14.3 per cent and 16.9 per cent respectively. The proportion of employment provided by the textile industry remained practically unchanged (2.6 per cent), and although the percentage of jobs provided by the transport equipment industry did not significantly increase (rising only to 2.3 per cent), it became the third most important source of jobs in the manufacturing sector.

Conversely, in 1996 the manufacturing sector represented roughly 60 per cent of total DVA in exports, while the service sector represented 25 per cent and agriculture and mining the remaining 15 per cent (Figure 6). Transport equipment, the most relevant manufacturing industry with regard to DVA, generated 18 per cent of total value-added in exports, representing more than one quarter of the value added by the manufacturing sector as a whole. The textile industry, meanwhile, came in fourth place with 5 per cent. In 2011, manufacturing continued to represent two thirds of total DVA; while transport equipment contributed to one third of that total, textiles had dropped to 3.5 per cent. Clearly, transport equipment was the main source of DVA generated in the manufacturing sector in Mexico in the period under study, even increasing its share in the total economy, but it was not the principal employer. In textiles, however, even though its contribution to total DVA in exports dropped during that same period, the proportion of jobs generated by the industry increased.

The above data do not capture the relationship between the value added in exports in each industry and its corresponding employment level. An input–output analysis is a common and useful approach for estimating the contribution of exports to job creation. Our analysis builds on the standard methodology used in the literature (Miller and Blair 2009). It relies on the classification of productive activities into the following broad types: domestic activities, and activities that involve production sharing between two or more countries (Wang et al. 2017). The details are presented in Annex 3.

Figure 5. Employment by industry as a proportion of total employment, Mexico, 1996 and 2011 (per cent)

Source: Own elaboration based on WIOD data.

Figure 6. DVA in exports by industry as a proportion of total DVA, Mexico, 1996 and 2011 (per cent)

Source: Authors’ calculations based on WIOD data.

We calculated the number of jobs created directly and indirectly for each type of production. This does not only include jobs created within each industry (namely direct employment within the industry itself) but also includes those indirectly supported by domestic suppliers (backward and forward linkages) both for domestic production and for production related to international trade in both traditional (related to exports of final goods) and GVC activities (exports of intermediate goods).

As Table 1 shows, in 1996 domestic production accounted for around 40 per cent of total jobs in both the textile and the transport equipment industries; this was almost 20 percentage points lower than the rate in the overall manufacturing sector, and almost half the rate in the overall economy. Traditional trade accounted for the majority of jobs in these two industries (46.4 per cent for textiles and 40.7 per cent for transport equipment). Only 11 per cent of total jobs in the textile industry were in GVC activities. This proportion reached almost 19 per cent in the transport equipment industry, just a little above the corresponding average for the manufacturing sector as a whole. In this sense, the textile industry more accurately reflected the Mexican economy as a whole, which accounted for 8.7 per cent of jobs in GVC activities.

The picture changed radically in the following years. In textiles, domestic production became the principal source of jobs (except for 2001, where traditional trade took first place) and traditional trade lost importance (dropping to 36.3 per cent). GVC activities recorded a small increase to 12.4 per cent. On the other hand, jobs related to trade activities in the transport equipment industry increased. The share of total jobs generated by final exports reached 52.5 per cent in 2011, while the proportion generated by GVC production reached 29.6 per cent. In other words, by 2011 domestic production had lost importance in terms of employment generation in the transport equipment industry, representing only 17.9 per cent of jobs, less than half that in 1996. Nonetheless, transport equipment remained the leading industry for job creation.

Table 1. Jobs associated with domestic production, traditional trade and GVC trade as a percentage of total jobs (per cent)

Source: Authors’ calculations based on WIOD data.

Table 2 shows whether the jobs generated by an industry corresponded to the industry itself or whether they were related to inputs from other industries, namely indirect jobs. In GVC trade, we cannot record the jobs generated inside and outside the industry, as it would involve not only Mexican-originated inputs but also other countries’ inputs that were “embedded” in intermediate imports.18 Instead, we registered the total percentage of jobs generated by Mexican exports of intermediate inputs that other countries subsequently incorporated into their production. Job creation over the 1996–2011 period is further defined as the difference in the number of jobs between 2011 and 1996.

In the textile industry, there is a significant difference between domestic production and traditional trade in terms of employment generation. At the start of the period of analysis, two thirds of jobs in domestic production were generated within the industry itself, while one third came from other industries (both upstream and downstream). In contrast, in traditional trade, approximately 10 per cent of jobs came from other sectors – a sign of the low capacity of final exports to stimulate more jobs outside the sector. Although in 2008 and 2011, the proportion of jobs created within each industry was reduced, the difference between domestic production and traditional trade remained constant. As mentioned above, GVC activities in the textile industry are not particularly relevant to job generation, unlike the automotive industry, in which GVC activities have become as crucial to generating jobs as traditional trade.

Between 1996 and 2011, the textile industry created 477,000 jobs, the vast majority of which (305,000) came from domestic production within the sector. Traditional trade generated 86,000 jobs. It is interesting to note that traditional trade generated more jobs outside the industry (23,700) than domestic production (14,900). The textile industry did not, therefore, increase its demand for additional inputs from other industries during this period, but rather demand actually decreased in relative terms.

Table 2. Number of jobs generated within or outside the industry according to production destination, Mexico, 1996–2011 (thousands)

Note: GVC trade covers exports and imports of all goods and services that are part of a GVC.

Source: Authors’ calculations based on WIOD data.

As noted above, in the transport equipment industry, traditional trade activities played a central role in employment. These jobs were mainly created within the industry. A negligible number of jobs came from other industries, which reflects the disconnect between this activity and the rest of the economy with regard to employment generation. There was also a clear decrease in the importance of domestic production; while in 1996 domestic production created almost as many jobs as traditional trade, in 2011 it was the least relevant area of production in terms of the number of jobs generated. The transport equipment industry created 521,000 jobs between 1996 and 2011, mainly in trade activities; this included a loss of 59,000 jobs in domestic production.

The analysis points to a distinction between the textile and transport equipment industries in terms of employment creation. While textiles has an inclination for domestic production with an essential connection to other industries, transport equipment is trade-oriented with almost no need for domestic inputs or post-production services. We can corroborate this trend through typical linkage analysis. (See the methodology described in Annex 3.) Linkages measure the effect of an increase in the output in any particular industry on other activities. As we are concerned with the impact on employment, Figure 7 shows the relative position of the two industries in terms of the number of indirect jobs created, in accordance with the taxonomy commonly used in the literature (see Table 3).

Table 3. Industrial activity taxonomy, according to forward and backward linkages

Source: Own elaboration based on Rasmussen (1957), inter alia.

In Figure 7, we present the results for both industries. It is no surprise that the transport equipment industry is classified as an “island” over the four years of study. In contrast, textiles is located in the key sector quadrant. This means that, relatively speaking, the Mexican textiles industry is an activity in which linkages with other industries are central to the generation of employment.

The textile industry is a crucial generator of employment in the Mexican economy. At the start of the period of analysis, production destined for final exports represented a source of employment that was almost as important as domestic production. However, by 2011 this situation had changed, and the industry mainly employed workers in domestic production. In addition to generating jobs within the industry, textile production destined for the domestic market generated jobs indirectly by demanding inputs from other sectors. The relative importance of these indirect jobs declined towards the end of the period of analysis, however. In activities related to foreign trade, the number of indirect jobs was much lower, although it remained an area relevant to job generation. The study of forward and backward linkages with regard to job creation corroborates the central role played by textiles in employment, which was found to be a key industry according to both indicators.

Figure 7. Backward and forward linkages in the textile and transport equipment industries based on the industrial activity taxonomy, Mexico, 1996–2011

Note: See Annex 3 for the methodology for classifying forward and backward linkages.

Source: Authors’ calculations based on WIOD data.

Although the transport equipment industry is the main generator of value-added in export production, it does not have the same relevance to job creation. Employment creation within this industry is concentrated in foreign trade activities, mainly those related to final exports, although the relative importance of GVC did increase during the period of analysis.

As we have seen, a significant proportion of the inputs necessary to produce exports in the transport equipment industry come from abroad, and therefore a significant number of the jobs created are based abroad. This seems to be related to the fact that the industry generates almost no external employment in other industries. In other words, it does not have significant backward or forward linkages with the domestic economy and thus is located in the “island” quadrant. This does not mean that the industry is not labour-intensive, rather that those jobs are most likely generated outside the country.

As mentioned above, the data provided in WIOD do not allow for a deeper analysis of trade and employment in relation to decent work. The next section proceeds to address this fundamental issue through the construction of multidimensional indicators and qualitative aspects of employment, thereby facilitating an approximate assessment of the evolution of decent work in the textiles and automotive industries in Mexico.

3.1 Selected indicators of decent work

In accordance with the above, in this study we constructed a battery of decent work indicators based on the ILO guidelines in order to analyse employment conditions at national level and in the automotive and textile industries specifically (see Table 4). Taking into consideration the ILO (2013a) report on the situation of decent work in Mexico, these indicators are divided into four axes:

Employment access is a key pillar in the promotion of decent work. The core principle is the promotion of “opportunities for women and men to obtain decent and productive work in conditions of freedom, equity, security, and human dignity” (ILO 1999). On the one hand, this notion implies the existence of employment opportunities for all who seek work; in other words, it measures access to employment from a positive perspective against the relevant population base. On the other hand, employment opportunities can also be measured in a negative sense in terms of unemployment, underemployment and the lack of employment opportunities (ILO 2008). Besides these indicators, it is necessary to know the overall conditions of employment, based, for example, on fair wages and the fulfilment of labour rights. To do so, we also considered the following aspects: insufficient income, rate of critical occupancy conditions, labour informality rate and the wage labour rate.

Employment security addresses the protection of the worker against sickness and disease (whether or not work-related) and occupational injuries, which is considered a fundamental element of social justice.20 In 1985, the International Labour Conference adopted the Labour Statistics Convention, 1985 (No. 160), which provides that: “Statistics of occupational injuries shall be compiled in such a way as to be representative of the country as a whole, covering, where possible, all branches of economic activity.” In Mexico, the Mexican Institute of Social Security provides social security cover for formal workers employed in private firms. It compiles and publishes information about the population covered, including injuries and illnesses at work.

Table 4. Decent Work indicators by axis

Source: Own elaboration based on ILO (2013a) and ILO (2013b).

Equal opportunity and treatment in employment: Another of the pillars established by the Millennium Development Goals was the promotion of equal opportunity and treatment in employment for all men and women. As the Declaration on Social Justice for a Fair Globalization states, gender equality and non-discrimination must be considered essential characteristics of decent work. This is a cross-cutting issue within the SDGs. The selected indicators provide information about women in the workplace, child labour and type of contract.

Social dialogue covers all types of negotiation, consultation or simply information exchange between representatives of the government, employers and workers on issues of common interests relating to economic and social policy (ILO 2008, 86). In the world of labour, social dialogue requires effective freedom of organization and trade union association for workers and employers and guaranteed periodic collective bargaining (ILO 2013b, 303). Selected indicators for social dialogue at national and industrial levels include strikes, summons to strikes, conciliations and collective labour agreements. However, the data must be interpreted with caution as, for example, the absence of strikes may result from excellent democratic negotiation processes between the relevant parties, or it may be a reflection of an authoritarian regime that has banned all demonstrations and strikes.

In Annex 5, we include the methodology and data sources used for computing these indicators.

3.2 Decent work indicators for the automotive and textile industries

3.2.1 Employment access, and equal opportunities and treatment in employment

According to the National Occupation and Employment Survey, in 2019 the labour force participation rate in Mexico was close to 60 per cent, while the unemployment rate was just 5.3 per cent. Both figures are relatively similar to those in 1996. However, in general, Mexico’s working conditions are far from laudable: 59 per cent of the employed population are in informal employment, and just under half of all workers have sufficient income. As shown in Table 5, between 1996 and 2019, the precarious employment rate (characterized by few hours of work or excessive hours with insufficient income) increased by 6.5 percentage points.

Over the period of analysis, the percentage of workers in the automotive industry with an income equivalent to less than two times the minimum wage (the insufficient income indicator) stalled around 53 per cent. Although the precarious employment rate for the industry almost tripled during the same period, it remained half the national level. The automotive industry had a high proportion of salaried workers, representing 99 per cent of employed personnel, and the informal employment rate for 2019 was 2.1 per cent, which was 26 times lower than the national rate.

Table 5. Axis I: Employment access

Source: Authors’ calculations based on the National Employment Survey and the National Occupation and Employment Survey conducted by INEGI and the Labour Statistical Information System maintained by the Secretariat of Labour and Social Welfare.

On the other hand, in the textile industry, job insecurity increased during the period of analysis. The proportion of informal workers increased from 58.9 to 62.8 per cent (above the national rate), and the precarious employment rate similarly surpassed the national level.

At national level, between 1996 and 2019, the percentage of workers with a formal contract decreased from 25.5 per cent to 22.8 per cent. While, in the automotive industry, most workers (63.9 per cent) possess a contract, only 21.6 per cent of workers in the textile industry have a contract.

The role of women in both industries is entirely different (see Table 6). In the automotive industry, only 36.9 per cent of employees were women in 2019, a rate which is remarkably similar to the rest of the economy (37.5 per cent). In the textile industry, meanwhile, women represent 64 per cent of employees. In both industries, some gender inequalities prevail. At national level, the wage gap decreased from 0.35 to 0.67 between 1996 and 2019, meaning that some progress was made in closing the gap between female and male earnings. However, we find a greater gap and some indications of the reverse trend in both the textile and automotive industries. In the automotive industry, although the gap diminished (from 0.16 in 1996 to 0.36 in 2019), some improvement achieved in the previous years was lost. (In 2008, the gender gap was 0.40.) In the textile industry, the wage gap was 0.52 in 1996, but it widened to 0.24 by 2008. While by 2019 it had recuperated somewhat, reaching 0.39, it remained at a lower level than in 1996.

Table 6. Axis II: Equal opportunity and treatment in employment

Source: Authors’ calculations based on the National Employment Survey and the National Occupation and Employment Survey conducted by INEGI and the Labour Statistical Information System maintained by the Secretariat of Labour and Social Welfare.

3.2.2 Employment security and social dialogue

The figures in Table 7 reveal an improvement in employment security in both the automotive and textile industries between 2008 and 2016. The first indicator (the percentage of labour injuries) shows a decreasing trend. Moreover, the percentage of permanent workers increased. The labour injuries indicator dropped from 1.9 per cent in 2008 to 1.1 per cent in 2016 in the automotive industry, and from 1.9 per cent in 2008 to 1.4 per cent in 2016 in the textile industry. Both figures are below the national rate of 2.3 per cent.

Table 7. Axis III: Employment security

Source: Authors’ calculations based on data from the Mexican Institute of Social Security.

While the percentage of permanent workers in the automotive industry had increased to 70.3 per cent by 2016, in the textile industry it had risen to only 44.6 per cent by the same year, representing a 25-percentage point gap between the two industries. Both industries were still above the national rate of 31 per cent, however. While opportunities remain to enhance the proportion of permanent workers, there has been an improvement in employment security in both the automotive and the textile industries.

The social dialogue indicators show a weakening in collective bargaining. Both nationally and in the automotive and textile industries specifically, the rate of strikes and collective bargaining agreements fell dramatically, in particular in the post-crisis period (see Table 8). As mentioned in Section 2, for decades labour policy was oriented to “preserve labour peace”, even through repression. In 2014, the Labour Secretary, Alfonso Navarrete Prida, celebrated that there had been no strikes or summons (Vergara 2015). Bensusán (2019) states that: “It was recognized, even from international economic organizations such as the World Bank and the International Monetary Fund, that the loss of union power or the weakening and atomization of collective bargaining are factors that explain the increase in income inequality and the slow economic recovery in Mexico.”

Table 8. Axis IV: Social dialogue

Source: Authors’ calculations based on data from the Secretariat of Labour and Social Welfare.

While the decent work indicators show general improvements with regard to employment security and labour rights, the industries present two different realities. On the one hand, the automotive industry experienced better results in employment generation, formal employment with defined contracts, and better wages. On the other hand, the textile industry showed poor performance in its path towards decent work. At national level, the precarious employment rate increased, which implies that employment conditions became more adverse as a result of lower incomes and fewer hours of work among the employed population, which warns of possible increases in underemployment.

In terms of gender equality, the results achieved in both industries are far from satisfactory. In the automotive industry, the proportion of female workers decreased. Although the proportion increased in the textile industry, the conditions were worse than the national average. In both cases, there is a need for men’s and women’s participation in workspaces to be more balanced and for workers to enjoy fairer conditions that provide sufficient income. Progress on social dialogue moved in the opposite direction to decent work, with a weakening of trade unions. The decline in the relevance of collective contracts and the reduction in summons to strike reflect the repressive policy towards such actions.

One crucial element in the road towards decent work is the increasing prevalence of formal labour relations and the development of a more competitive structure. More precisely, it makes a difference whether the key players in an industry tend to be big firms, operating in the formal labour markets, or if they are mainly small- and microenterprises in informal markets. In the case of the textile industry, 95 per cent of companies are small- and medium-sized enterprises (SMEs), meaning that they do not have more than ten employees, and only 0.28 per cent of companies have more than 250 employees. In the automotive industry, microenterprises represent only 38 per cent of all companies, and large companies represent 26 per cent. Over the past 20 years, foreign direct investment in the auto industry has exceeded that in textiles by a multiple of 15 or more. This difference may be crucial in explaining their distinct patterns of insertion in the world market and the evolution towards decent work.

Annex 1. Main provisions of trade agreements applicable to Mexico

Source: Based on DOF, 2018.

Annex 2. World Input–Output Database (WIOD) countries and territories

Acronym	Country	Group		Acronym	Country	Group
AUS	Australia	AUS		IRL	Ireland	EU15
AUT	Austria	EU15		ITA	Italy	EU15
BEL	Belgium	EU15		JPN	Japan	JPN
BGR	Bulgaria	EU13		KOR	Republic of Korea	KOR
BRA	Brazil	BRA		LTU	Lithuania	EU13
CAN	Canada	CAN		LUX	Luxemburg	EU15
CHN	China	CHN		LVA	Latvia	EU13
CYP	Cyprus	EU13		MEX	Mexico	Mexico
CZE	Czechia	EU13		MLT	Malta	EU13
DEU	Germany	EU15		NLD	Netherlands	EU15
DNK	Denmark	EU15		POL	Poland	EU13
ESP	Spain	EU15		PRT	Portugal	EU15
EST	Estonia	EU13		ROM	Romania	EU13
FIN	Finland	EU15		RUS	Russia	RUS
FRA	France	EU15		SVK	Slovakia	EU13
GBR	United Kingdom	EU15		SVN	Slovenia	EU13
GRC	Greece	EU15		SWE	Sweden	EU15
HUN	Hungry	EU13		TUR	Turkey	TUR
IDN	Indonesia	IDN		TWN	Taiwan, China	TWN
IND	India	IND		USA	United States	USA

Annex 3. Methodology: Measuring job creation related to domestic and trade activities

Input–output analysis is a common and useful approach for estimating the contribution of exports to job creation. The model in matrix form is described as $\mathit{X} = \mathit{A}\mathit{X} + \mathit{Y}$, where X is the value of production, $\mathit{A}$ is a matrix of technical coefficients and $\mathit{Y}\mathrm{ }$is a vector of final demand in the economy. The model is solved using matrix algebra with this result:$\mathrm{ }\mathit{X} = {\left(\mathit{I} – \mathit{A}\right)}^{-1} \mathit{Y} = \mathit{L}\mathit{Y}$, where ${\left(\mathit{I}-\mathit{A}\right)}^{-1} = \mathit{L}\mathrm{ }$is known as the Leontief inverse or the total requirements matrix (Miller and Blair 2009). As is well known, the Leontief inverse captures the total impact of changes in final demand, taking account of both direct and indirect effects. Total effects include the need for inputs from the first change in final demand from productive sectors; this, in turn, generates additional effects (indirect effects) from the inputs required to meet that change in demand and so forth. In our case, we are not interested in changes in production but rather changes in employment; instead of analysing the total final demand, the focus will be on exports only.

For this purpose, we calculated employment coefficients for each sector by dividing $i_{th}$ sector employment by output in that same sector. This information, when arranged as a diagonal matrix $\widehat{\mathit{E}}$ (with employment coefficients placed along the main diagonal and zero elsewhere), pre-multiplies the Leontief inverse$\mathrm{ }\widehat{\mathit{E}}{\left(\mathit{I}-\mathit{A}\right)}^{-1}$, allowing us to obtain the number of jobs directly and indirectly needed for the production of one unit of final demand in a national set-up. If we also post-multiply this result by a matrix of exports, $\mathit{T}$, we obtain a measure of the number of jobs needed to produce net exports: $\widehat{\mathit{E}} {\left(\mathit{I}-\mathit{A}\right)}^{-1} \mathit{T}$. The labour coefficients matrix, $\mathit{E}$, may represent different categories of employment, for example according to skill level or gender. In this sense, we can track labour outcomes in terms of the value-added paid to the labour force in the production of goods according to different classifications of jobs (gender or skill level, among others).

This idea, however, must be reviewed in the light of GVCs; if we want to calculate the number of jobs generated by a country’s exports, we have to take into account that those exports include value-added from inputs that come from other countries. The result can be misleading because we would be ignoring cross-border production linkages, which are better accounted for by trade in value-added statistics.

Jiang (2013), for example, used the 2014 release of the WIOD tables to analyse five categories of effects that trade activities had on employment during the 1995–2009 period, specifically the labour content in exports, in imports, in the import content of exports, in the export content of imports, and in intermediates contained in imports from a third country, the latter three categories representing GVC activities. Jiang’s paper focused on jobs created through the labour demand of a country, irrespective of whether they were created within the country or elsewhere.

Given our interest in estimating the number of jobs created in Mexico by domestic production versus those created by trade-related activities, we adopted the suggestions of Wang et al. (2017) and classified productive activities into broad types depending on whether they were domestic or whether they involved production-sharing between two or more countries. The matrix equation $\widehat{\mathit{E}}\mathit{B}\widehat{\mathit{Y}}=\widehat{\mathit{E}}{\mathit{L}\widehat{\mathit{Y}}}^{\mathit{D}}+\widehat{\mathit{E}}\mathit{L}{\widehat{\mathit{Y}}}^{\mathit{F}}+\widehat{\mathit{E}}\mathit{L}{\mathit{A}}^{\mathit{F}}\mathit{B}\widehat{\mathit{Y}}$ is the number of jobs involved in selected activities from a source country (or sector) that are directly or indirectly involved in final production in a given country or sector. Where $\mathit{B}$ is the global inverse Leontief matrix, ${\widehat{\mathit{Y}}}^{\mathit{D}}$ represents a diagonal matrix with information on final production for domestic consumption, ${\widehat{\mathbf{Y}}}^{\mathbf{F}}=\mathrm{ }\widehat{\mathbf{Y}}-{\widehat{\mathbf{Y}}}^{\mathbf{D}}$ is the diagonal matrix of final products exports, and ${\mathit{A}}^{\mathit{F}}=\mathit{A}-{\mathit{A}}^{\mathit{D}}$ is an off-diagonal block matrix of imported input coefficients.

The equation $\widehat{\mathit{E}}\mathit{B}\widehat{\mathit{Y}}=\widehat{\mathit{E}}{\mathit{L}\widehat{\mathit{Y}}}^{\mathit{D}}+\widehat{\mathit{E}}\mathit{L}{\widehat{\mathit{Y}}}^{\mathit{F}}+\widehat{\mathit{E}}\mathit{L}{\mathit{A}}^{\mathit{F}}\mathit{B}\widehat{\mathit{Y}}$ identifies, for each country-sector, three types of jobs associated with production activities:

1. $\widehat{\mathit{E}}{\mathit{L}\widehat{\mathit{Y}}}^{\mathit{D}}$: Jobs used to produce and consume goods domestically, in which there is no cross-border trade.

2. $\widehat{\mathit{E}}\mathit{L}{\widehat{\mathit{Y}}}^{\mathit{F}}$: Employment incorporated into exports of final goods. This domestic factor content crosses national borders only for consumption (of goods and services, as well as capital) in the other country. We classify these as jobs generated in “traditional” trade.

3. $\widehat{\mathit{E}}\mathit{L}{\mathit{A}}^{\mathit{F}}\mathit{B}\widehat{\mathit{Y}}$: Employment included in exports/imports of intermediate goods and services. Production activities in different countries use this value, exported from the origin country, to produce their exports. Hence, the jobs created in this category are part of the interchange of production between countries or GVC activities.

Using WIOD tables for the chosen years, we can calculate the $\mathit{B},\mathrm{ }\widehat{\mathit{Y}, }{\mathit{L},\mathrm{ }\widehat{\mathit{Y}}}^{\mathit{D}},\mathrm{ }{\widehat{\mathit{Y}}}^{\mathit{F}}\mathrm{ }\mathrm{a}\mathrm{n}\mathrm{d}\mathrm{ }\mathrm{ }{\mathit{A}}^{\mathit{F}}$ matrices. $\widehat{\mathit{E}}\mathrm{ }$is obtained using data on the number of persons in employment taken from the Socio-Economic Accounts of the 2014 WIOD release and the output vector from WIOD. More detailed information on the construction of the Socio-Economic Accounts is available in Gouma, Timmer and de Vries (2014).

The resulting terms $\widehat{\mathit{E}}{\mathit{L}\widehat{\mathit{Y}}}^{\mathit{D}},\widehat{\mathit{E}}\mathit{L}{\widehat{\mathit{Y}}}^{\mathit{F}}\mathrm{ }\mathrm{a}\mathrm{n}\mathrm{d}\mathrm{ }\widehat{\mathit{E}}\mathit{L}{\mathit{A}}^{\mathit{F}}\mathit{B}\widehat{\mathit{Y}}$ correspond to matrices of dimensions (41x35, 41x35); to account for the number of persons engaged in each type of activity, we therefore obtained the sum of each row. For the first two terms, it is possible to assess whether those jobs correspond to the industry under analysis or whether they relate to inputs from other industries –making them indirect jobs – because they correspond to diagonal matrices. In the case of the third term, GVC trade, we cannot differentiate between jobs generated inside and outside each industry, as such jobs include not only those pertaining to inputs that originated in the country of analysis but also those pertaining to inputs from all other countries and to those embedded in intermediate imports.

Another way to measure direct and indirect job creation is through linkage analysis. A typical backward linkage analysis shows the degree to which the output of sector j depends directly on inputs from other sectors, which can be measured through matrix$\mathrm{ }\mathit{A}$. To also consider indirect effects, the analysis should be done using matrix $\mathit{L}$ (local Leontief inverse) (Miller and Blair 2009).

As our interest lies in employment generation, we pre-multiplied the Leontief matrix by the employment coefficients present in $\widehat{ \mathit{E}}$. So, to calculate linkages, it is enough to perform the sum for each column of the elements j of the indicated matrix: $\mathit{b} \left(\mathit{t}\right)=\widehat{ \mathit{E}}{\mathit{i}}^{\text{'}}\mathit{L}$. Rasmussen (1957) proposed a standardized interpretation of these indicators: $\mathit{b}\left(\mathit{t}\right)=\frac{ \widehat{ \mathit{E}}\mathit{n}{\mathit{i}}^{\text{'}}\mathit{L}}{{\mathit{i}}^{\text{'}}\widehat{ \mathit{E}}\mathit{L}\mathit{i}}$. In other words, this indicator measures the ratio of the jobs created in all industries for good j and the employment generated in industry j, independently of the prevailing final demand. This indicator can also approximate the economic importance of the selected industries (or key sectors). If the importance of one is higher than another, then that industry has a more significant impact than the average of all the other industries. In the case of backward linkages, this is known as the dispersion index, and if we calculate it with the Ghosh inverse matrix, we obtain the forward linkages.

Annex 4. Automotive and textile industries: Table of equivalences between the 1993 Mexican National System of Accounts (SNA) and the 2002 North American Industry Classification System (NAICS)

Annex 5. Methodology: Mexico’s decent work indicators (1996–2019)

The estimation of decent work indicators for the Mexican economy and, more precisely, for the textile and automotive industries was based on the 2013 ILO manual of decent work indicators and the first edition of the ILO biannual newsletter entitled Situación del trabajo decente en México. Based on both documents and on the statistical information available, we created a set of indicators across four axes: employment access, employment security, equal opportunity and treatment in employment, and social dialogue. To that end, it was necessary to incorporate variables obtained from the Mexican National Institute of Statistics and Geography (INEGI), the Mexican Secretariat of Labour and Social Welfare (STPS) and the Mexican Institute of Social Security (IMSS).

Axis 1. Employment access indicators

Labour force participation rate (LFPR)

Measurement objective: Percentage of the working age population (15 years and over) that participates in the labour force.

Data sources: INEGI (National Employment Survey and National Occupation and Employment Survey)

Method of computation:

$LFPR=\left(\frac{\left(\mathrm{N}\mathrm{u}\mathrm{m}\mathrm{b}\mathrm{e}\mathrm{r}\mathrm{ }\mathrm{o}\mathrm{f}\mathrm{ }\mathrm{e}\mathrm{m}\mathrm{p}\mathrm{l}\mathrm{o}\mathrm{y}\mathrm{e}\mathrm{d}\mathrm{ }+\mathrm{ }\mathrm{n}\mathrm{u}\mathrm{m}\mathrm{b}\mathrm{e}\mathrm{r}\mathrm{ }\mathrm{o}\mathrm{f}\mathrm{ }\mathrm{u}\mathrm{n}\mathrm{e}\mathrm{m}\mathrm{p}\mathrm{l}\mathrm{o}\mathrm{y}\mathrm{e}\mathrm{d}\right)}{\mathrm{T}\mathrm{o}\mathrm{t}\mathrm{a}\mathrm{l}\mathrm{ }\mathrm{n}\mathrm{u}\mathrm{m}\mathrm{b}\mathrm{e}\mathrm{r}\mathrm{ }\mathrm{o}\mathrm{f}\mathrm{ }\mathrm{t}\mathrm{h}\mathrm{e}\mathrm{ }\mathrm{w}\mathrm{o}\mathrm{r}\mathrm{k}\mathrm{i}\mathrm{n}\mathrm{g}\mathrm{ }\mathrm{a}\mathrm{g}\mathrm{e}\mathrm{ }\mathrm{p}\mathrm{o}\mathrm{p}\mathrm{u}\mathrm{l}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}}\right)\times 100$

Unemployment rate (UR)

Measurement objective: Percentage of the labour force that is not working but is searching for a job.

Data sources: INEGI (National Employment Survey and National Occupation and Employment Survey)

Method of computation:

$UR=\left(\frac{\mathrm{N}\mathrm{u}\mathrm{m}\mathrm{b}\mathrm{e}\mathrm{r}\mathrm{ }\mathrm{o}\mathrm{f}\mathrm{ }\mathrm{u}\mathrm{n}\mathrm{e}\mathrm{m}\mathrm{p}\mathrm{l}\mathrm{o}\mathrm{y}\mathrm{e}\mathrm{d}\mathrm{ }}{\mathrm{T}\mathrm{o}\mathrm{t}\mathrm{a}\mathrm{l}\mathrm{ }\mathrm{n}\mathrm{u}\mathrm{m}\mathrm{b}\mathrm{e}\mathrm{r}\mathrm{ }\mathrm{o}\mathrm{f}\mathrm{ }\mathrm{p}\mathrm{e}\mathrm{r}\mathrm{s}\mathrm{o}\mathrm{n}\mathrm{s}\mathrm{ }\mathrm{i}\mathrm{n}\mathrm{ }\mathrm{t}\mathrm{h}\mathrm{e}\mathrm{ }\mathrm{l}\mathrm{a}\mathrm{b}\mathrm{o}\mathrm{u}\mathrm{r}\mathrm{ }\mathrm{f}\mathrm{o}\mathrm{r}\mathrm{c}\mathrm{e}}\right)\times 100$

General pressure rate (GPR)

Measurement objective: Percentage of the labour force represented by the unemployed population, plus the employed population in pursuit of another job.

Data sources: INEGI (National Employment Survey and National Occupation and Employment Survey)

Method of computation:

$GPR=\left(\frac{\mathrm{N}\mathrm{u}\mathrm{m}\mathrm{b}\mathrm{e}\mathrm{r}\mathrm{ }\mathrm{o}\mathrm{f}\mathrm{ }\mathrm{p}\mathrm{e}\mathrm{r}\mathrm{s}\mathrm{o}\mathrm{n}\mathrm{s}\mathrm{ }\mathrm{u}\mathrm{n}\mathrm{e}\mathrm{m}\mathrm{p}\mathrm{l}\mathrm{o}\mathrm{y}\mathrm{e}\mathrm{d}\mathrm{ }+\mathrm{ }\mathrm{n}\mathrm{u}\mathrm{m}\mathrm{b}\mathrm{e}\mathrm{r}\mathrm{ }\mathrm{o}\mathrm{f}\mathrm{ }\mathrm{p}\mathrm{e}\mathrm{r}\mathrm{s}\mathrm{o}\mathrm{n}\mathrm{s}\mathrm{ }\mathrm{e}\mathrm{m}\mathrm{p}\mathrm{l}\mathrm{o}\mathrm{y}\mathrm{e}\mathrm{d}\mathrm{ }\mathrm{i}\mathrm{n}\mathrm{ }\mathrm{p}\mathrm{u}\mathrm{r}\mathrm{s}\mathrm{u}\mathrm{i}\mathrm{t}\mathrm{ }\mathrm{o}\mathrm{f}\mathrm{ }\mathrm{a}\mathrm{n}\mathrm{o}\mathrm{t}\mathrm{h}\mathrm{e}\mathrm{r}\mathrm{ }\mathrm{j}\mathrm{o}\mathrm{b}}{\mathrm{ }\mathrm{L}\mathrm{a}\mathrm{b}\mathrm{o}\mathrm{u}\mathrm{r}\mathrm{ }\mathrm{f}\mathrm{o}\mathrm{r}\mathrm{c}\mathrm{e}}\right)\times$100

Underemployment rate (UER)

Measurement objective: Percentage of the employed population represented by working age individuals who need, and are available, to offer more hours of work than their current occupation allows them to perform.

Data sources: INEGI (National Employment Survey and National Occupation and Employment Survey)

Method of computation:

$UER=\left(\frac{\mathrm{N}\mathrm{u}\mathrm{m}\mathrm{b}\mathrm{e}\mathrm{r}\mathrm{ }\mathrm{o}\mathrm{f}\mathrm{ }\mathrm{p}\mathrm{e}\mathrm{r}\mathrm{s}\mathrm{o}\mathrm{n}\mathrm{s}\mathrm{ }\mathrm{u}\mathrm{n}\mathrm{d}\mathrm{e}\mathrm{r}\mathrm{e}\mathrm{m}\mathrm{p}\mathrm{l}\mathrm{o}\mathrm{y}\mathrm{e}\mathrm{d}}{\mathrm{ }\mathrm{E}\mathrm{m}\mathrm{p}\mathrm{l}\mathrm{o}\mathrm{y}\mathrm{e}\mathrm{d}\mathrm{ }\mathrm{p}\mathrm{o}\mathrm{p}\mathrm{u}\mathrm{l}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}}\right)\times 100$

Precarious employment rate (PER)

Measurement objective: INEGI defines persons in “precarious employment” as: employed persons who work less than 35 hours a week for market reasons; employed persons who work more than 35 hours a week and have a monthly income below the minimum wage; and employed persons who work more than 48 hours a week and earn the equivalent of no more than two times the minimum wage. According to the ILO (2013c), the precarious employment rate is the percentage of the employed population in precarious employment (as defined by INEGI).

Data sources: INEGI (National Employment Survey and National Occupation and Employment Survey)

Method of computation:

$PER=\left(\frac{\mathrm{N}\mathrm{u}\mathrm{m}\mathrm{b}\mathrm{e}\mathrm{r}\mathrm{ }\mathrm{i}\mathrm{n}\mathrm{ }\mathrm{p}\mathrm{r}\mathrm{e}\mathrm{c}\mathrm{a}\mathrm{r}\mathrm{i}\mathrm{o}\mathrm{u}\mathrm{s}\mathrm{ }\mathrm{e}\mathrm{m}\mathrm{p}\mathrm{l}\mathrm{o}\mathrm{y}\mathrm{m}\mathrm{e}\mathrm{n}\mathrm{t}}{\mathrm{ }\mathrm{E}\mathrm{m}\mathrm{p}\mathrm{l}\mathrm{o}\mathrm{y}\mathrm{e}\mathrm{d}\mathrm{ }\mathrm{p}\mathrm{o}\mathrm{p}\mathrm{u}\mathrm{l}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}}\right)\times 100$

Informal employment rate (IR)

Measurement objective: Proportion of the employed population who are labour-vulnerable owing to the nature of the economic unit in which they work and/or whose labour relationship or dependency is not recognized by their source of employment.

Data sources: INEGI (National Employment Survey and National Occupation and Employment Survey)

Method of computation:

$IR=\left(\frac{\mathrm{N}\mathrm{u}\mathrm{m}\mathrm{b}\mathrm{e}\mathrm{r}\mathrm{ }\mathrm{i}\mathrm{n}\mathrm{ }\mathrm{i}\mathrm{n}\mathrm{f}\mathrm{o}\mathrm{r}\mathrm{m}\mathrm{a}\mathrm{l}\mathrm{ }\mathrm{e}\mathrm{m}\mathrm{p}\mathrm{l}\mathrm{o}\mathrm{y}\mathrm{m}\mathrm{e}\mathrm{n}\mathrm{t}}{\mathrm{ }\mathrm{E}\mathrm{m}\mathrm{p}\mathrm{l}\mathrm{o}\mathrm{y}\mathrm{e}\mathrm{d}\mathrm{ }\mathrm{p}\mathrm{o}\mathrm{p}\mathrm{u}\mathrm{l}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}\mathrm{ }}\right)\times 100$

Insufficient income indicator (III)

Measurement objective: Proportion of the employed population that earns the equivalent of no more than two times the minimum wage.

Data sources: INEGI (National Employment Survey and National Occupation and Employment Survey)

Method of computation:

$III=\left(\frac{\mathrm{N}\mathrm{o}.\mathrm{ }\mathrm{o}\mathrm{f}\mathrm{ }\mathrm{e}\mathrm{m}\mathrm{p}\mathrm{l}\mathrm{o}\mathrm{y}\mathrm{e}\mathrm{d}\mathrm{ }\mathrm{p}\mathrm{e}\mathrm{r}\mathrm{s}\mathrm{o}\mathrm{n}\mathrm{s}\mathrm{ }\mathrm{e}\mathrm{a}\mathrm{r}\mathrm{n}\mathrm{i}\mathrm{n}\mathrm{g}\mathrm{ }\mathrm{n}\mathrm{o}\mathrm{ }\mathrm{m}\mathrm{o}\mathrm{r}\mathrm{e}\mathrm{ }\mathrm{t}\mathrm{h}\mathrm{a}\mathrm{n}\mathrm{ }\mathrm{t}\mathrm{w}\mathrm{i}\mathrm{c}\mathrm{e}\mathrm{ }\mathrm{m}\mathrm{i}\mathrm{n}\mathrm{i}\mathrm{m}\mathrm{u}\mathrm{m}\mathrm{ }\mathrm{w}\mathrm{a}\mathrm{g}\mathrm{e}}{\mathrm{E}\mathrm{m}\mathrm{p}\mathrm{l}\mathrm{o}\mathrm{y}\mathrm{e}\mathrm{d}\mathrm{ }\mathrm{p}\mathrm{o}\mathrm{p}\mathrm{u}\mathrm{l}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}}\right)\times 100$

Wage labour rate (WLR)

Measurement objective: Percentage of the employed population that receives a salary from the economic unit for which they work or for which they carry out activities.

Data sources: INEGI (National Employment Survey and National Occupation and Employment Survey)

Method of computation:

$WLR=\left(\frac{\mathrm{N}\mathrm{u}\mathrm{m}\mathrm{b}\mathrm{e}\mathrm{r}\mathrm{ }\mathrm{o}\mathrm{f}\mathrm{ }\mathrm{e}\mathrm{m}\mathrm{p}\mathrm{l}\mathrm{o}\mathrm{y}\mathrm{e}\mathrm{e}\mathrm{s}\mathrm{ }\mathrm{w}\mathrm{i}\mathrm{t}\mathrm{h}\mathrm{ }\mathrm{a}\mathrm{ }\mathrm{s}\mathrm{a}\mathrm{l}\mathrm{a}\mathrm{r}\mathrm{y}}{\mathrm{E}\mathrm{m}\mathrm{p}\mathrm{l}\mathrm{o}\mathrm{y}\mathrm{e}\mathrm{d}\mathrm{ }\mathrm{p}\mathrm{o}\mathrm{p}\mathrm{u}\mathrm{l}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}}\right)\times 100$

Axis 2. Equal opportunity and treatment in employment

Unemployment ratio by gender (URG)

Measurement objective: Ratio between the unemployment rate for women and the unemployment rate for men.

Data sources: INEGI (National Employment Survey and National Occupation and Employment Survey)

Method of computation:

$URG=\frac{\mathrm{U}\mathrm{n}\mathrm{e}\mathrm{m}\mathrm{p}\mathrm{l}\mathrm{o}\mathrm{y}\mathrm{m}\mathrm{e}\mathrm{n}\mathrm{t}\mathrm{ }\mathrm{r}\mathrm{a}\mathrm{t}\mathrm{e}\mathrm{ }\mathrm{f}\mathrm{o}\mathrm{r}\mathrm{ }\mathrm{w}\mathrm{o}\mathrm{m}\mathrm{e}\mathrm{n}}{\mathrm{U}\mathrm{n}\mathrm{e}\mathrm{m}\mathrm{p}\mathrm{l}\mathrm{o}\mathrm{y}\mathrm{m}\mathrm{e}\mathrm{n}\mathrm{t}\mathrm{ }\mathrm{r}\mathrm{a}\mathrm{t}\mathrm{e}\mathrm{ }\mathrm{f}\mathrm{o}\mathrm{r}\mathrm{ }\mathrm{m}\mathrm{e}\mathrm{n}}$

Gender wage gap (GWG)

Measurement objective: Ratio of employed women who earn three or more times the minimum wage to employed men who earn three or more times the minimum wage.

Data sources: INEGI (National Employment Survey and National Occupation and Employment Survey)

Method of computation:

$GWG=\frac{\mathrm{W}\mathrm{o}\mathrm{m}\mathrm{e}\mathrm{n}\mathrm{ }\mathrm{e}\mathrm{a}\mathrm{r}\mathrm{n}\mathrm{i}\mathrm{n}\mathrm{g}\mathrm{ }\mathrm{t}\mathrm{h}\mathrm{r}\mathrm{e}\mathrm{e}\mathrm{ }\mathrm{o}\mathrm{r}\mathrm{ }\mathrm{m}\mathrm{o}\mathrm{r}\mathrm{e}\mathrm{ }\mathrm{t}\mathrm{i}\mathrm{m}\mathrm{e}\mathrm{s}\mathrm{ }\mathrm{t}\mathrm{h}\mathrm{e}\mathrm{ }\mathrm{m}\mathrm{i}\mathrm{n}\mathrm{i}\mathrm{m}\mathrm{u}\mathrm{m}\mathrm{ }\mathrm{w}\mathrm{a}\mathrm{g}\mathrm{e}}{\mathrm{M}\mathrm{e}\mathrm{n}\mathrm{ }\mathrm{e}\mathrm{a}\mathrm{r}\mathrm{n}\mathrm{i}\mathrm{n}\mathrm{g}\mathrm{ }\mathrm{t}\mathrm{h}\mathrm{r}\mathrm{e}\mathrm{e}\mathrm{ }\mathrm{o}\mathrm{r}\mathrm{ }\mathrm{m}\mathrm{o}\mathrm{r}\mathrm{e}\mathrm{ }\mathrm{t}\mathrm{i}\mathrm{m}\mathrm{e}\mathrm{s}\mathrm{ }\mathrm{t}\mathrm{h}\mathrm{e}\mathrm{ }\mathrm{m}\mathrm{i}\mathrm{n}\mathrm{i}\mathrm{m}\mathrm{u}\mathrm{m}\mathrm{ }\mathrm{w}\mathrm{a}\mathrm{g}\mathrm{e}}$

Female share of employment (FE)

Measurement objective: Percentage of women of working age who are employed.

Data sources: INEGI (National Employment Survey and National Occupation and Employment Survey)

Method of computation:

$FE=\left(\frac{\mathrm{W}\mathrm{o}\mathrm{m}\mathrm{e}\mathrm{n}\mathrm{ }\mathrm{e}\mathrm{m}\mathrm{p}\mathrm{l}\mathrm{o}\mathrm{y}\mathrm{e}\mathrm{d}}{\mathrm{W}\mathrm{o}\mathrm{m}\mathrm{e}\mathrm{n}\mathrm{ }\mathrm{o}\mathrm{f}\mathrm{ }\mathrm{w}\mathrm{o}\mathrm{r}\mathrm{k}\mathrm{i}\mathrm{n}\mathrm{g}\mathrm{ }\mathrm{a}\mathrm{g}\mathrm{e}\mathrm{ }}\right)\times 100$

Axis 3. Employment security

Labour injuries as a percentage of total insured workers (LI)

Measurement objective: Percentage of workers insured by the IMSS who have had an accident at work.

Data sources: STPS and IMSS

Method of computation:

$LI=\frac{\mathrm{N}\mathrm{u}\mathrm{m}\mathrm{b}\mathrm{e}\mathrm{r}\mathrm{ }\mathrm{o}\mathrm{f}\mathrm{ }\mathrm{a}\mathrm{c}\mathrm{c}\mathrm{i}\mathrm{d}\mathrm{e}\mathrm{n}\mathrm{t}\mathrm{s}\mathrm{ }\mathrm{a}\mathrm{t}\mathrm{ }\mathrm{w}\mathrm{o}\mathrm{r}\mathrm{k}}{\mathrm{E}\mathrm{m}\mathrm{p}\mathrm{l}\mathrm{o}\mathrm{y}\mathrm{e}\mathrm{e}\mathrm{s}\mathrm{ }\mathrm{i}\mathrm{n}\mathrm{s}\mathrm{u}\mathrm{r}\mathrm{e}\mathrm{d}\mathrm{ }\mathrm{b}\mathrm{y}\mathrm{ }\mathrm{I}\mathrm{M}\mathrm{S}\mathrm{S}}\times 100$

Permanent workers as a percentage of the employed population

Measurement objective: Percentage of the employed population who are permanent workers registered with the IMSS.

Data sources: INEGI (National Occupation and Employment Survey) and IMSS

Method of computation:

$\frac{\mathrm{N}\mathrm{u}\mathrm{m}\mathrm{b}\mathrm{e}\mathrm{r}\mathrm{ }\mathrm{o}\mathrm{f}\mathrm{ }\mathrm{p}\mathrm{e}\mathrm{r}\mathrm{m}\mathrm{a}\mathrm{n}\mathrm{e}\mathrm{n}\mathrm{t}\mathrm{ }\mathrm{w}\mathrm{o}\mathrm{r}\mathrm{k}\mathrm{e}\mathrm{r}\mathrm{s}\mathrm{ }\mathrm{i}\mathrm{n}\mathrm{s}\mathrm{u}\mathrm{r}\mathrm{e}\mathrm{d}\mathrm{ }\mathrm{b}\mathrm{y}\mathrm{ }\mathrm{I}\mathrm{M}\mathrm{S}\mathrm{S}}{\mathrm{E}\mathrm{m}\mathrm{p}\mathrm{l}\mathrm{o}\mathrm{y}\mathrm{e}\mathrm{d}\mathrm{ }\mathrm{p}\mathrm{o}\mathrm{p}\mathrm{u}\mathrm{l}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}}\times 100$

Axis 4. Social dialogue

Summons

Measurement objective: Total number of summons.

Data sources: STPS

Strikes

Measurement objective: Total number of strikes.

Data sources: STPS

Collective bargaining agreements (CBA)

Measurement objective: Total number of collective bargaining agreements.

Data sources: STPS