ANALYSIS THE FUTURE OF WORK IN MEXICO: Exploring Key Actors’ Perceptions on the Socio-Economic and Policy Implications of Workplace Transformations Achim Kemmerling, Melina Altamirano and Ma. Susana Rosales November 2023 The discussion about workplace transformations brings together questions about technological change, from physical and virtual automation processes by bots and robots to platform economies, big data, and artificial intelligence. New forms of work have followed the digital technological turn. Globally, thousands of businesses linked to digital technology now depend on algorithms built to sustain service, transport, telecommunications, and manufacturing activities. The debate about the implications of these changes has created both positive and negative scenarios. In Mexico, the shock of digital transformation is asymmetric, affecting some regions, some sectors, and some vulnerable groups much more than others. In this document, we present the findings of an expert focus group discussing the challenges and opportunities brought by digital change in Mexico.

THE FUTURE OF WORK IN MEXICO: Exploring Key Actors’ Perceptions on the Socio-Economic and Policy Implications of Workplace Transformations Achim Kemmerling, Melina Altamirano and Ma. Susana Rosales November 2023

Contents Introduction: Debating the future 4 of work in Mexico 1. Technological change and workplace 6 transformations in Mexico 2. Participants and procedure 9 3. Results 10 The socio-economic impact of new technologies 10 Policy and political implications 12 Political obstacles to better policies 14 4 Discussion and concluding reflections 15 5. References 17

FRIEDRICH-EBERT-STIFTUNG- The Future of Work in Mexico INTRODUCTION: DEBATING THE FUTURE OF WORK IN MEXICO * The recent release of ChatGPT, an artificial intelligence-powered chatbot seemingly able to perform various tasks associated with qualified jobs, has fascinated millions of users and created shockwaves for those who reflect on the future of work worldwide. The discussion about workplace transformations brings together questions about technological change, from physical and virtual automation processes by bots and robots to platform economies, big data, and artificial intelligence. New forms of work have followed the digital technological turn. Globally, thousands of businesses linked to digital technology now depend on algorithms built to sustain service, transport, telecommunications, and manufacturing activities. The debate about the implications of these changes has created both positive and negative scenarios. Among the optimistic views is that new technologies will lead to more productivity, freeing people from menial or outright dangerous tasks and allowing them to lead more meaningful and creative lives. These new labor dynamics may also allow workers to emancipate from careers tied to particular employers, offering them the possibility of providing their services at will(Srnicek, 2018). Many observers are more alarmed than hopeful, however. In the negative scenario, automation will lead to net job losses across all economic sectors. It will lead to more surveillance at the workplace, less autonomy, and less social protection and regulation, with potential additional negative effects on workers’ mental health. Some analysts argue that platform economies can also lead to extreme market concentration and enormous levels of income inequality (e.g., Brynjolfsson& McAfee, 2014; Guellec& Paunov, 2017). Thus, public discussions on the future of work have significant policy and political implications, ranging from questions of infrastructure(power grit and servers) to economic and industrial policy, education, and, most importantly, social and labor market policies. In Mexico, these debates are relatively new. Similar to other middle-income countries’ economies, the Mexican labor market is characterized by significant disparities in productivity and capital access(Rivas Valdivia& Gaudin, 2021), marking an unequal pace in technological-led transformation across sectors. Many observers argue that the future of work is still far away and that companies see much less need to invest in potentially labor-saving technologies since labor costs are so low. Nevertheless, some academic researchers, international organizations, and business consultancies suggest that risks of job loss due to automation might affect as much as half of all jobs in middle-income countries(e.g., Nedelkoska& Quintini, 2018). While some of these projections might overestimate the impact of technological change, it seems safe to argue that significant workplace transformations are imminent in Mexico. Previous works have started studying the policy implications of workplace transformations in a comparative perspective, with a particular concentration on high-income countries. Current studies reveal the complexity and varying roles of social actors intervening in the processes linking technological change, labor markets, and policy reforms. An assessment of the political consequences of new technologies thus requires analyzing the perceptions, expectations, and preferences of the mass public, but also a particular focus on actors directly related to sectors dealing with technological change, civil society, and practitioners. In this paper, we report and discuss the results of an expert focus group bringing together actors from several sectors of Mexican society: government, industry, trade unions, the technological community, and academia. The objective of this workshop was to foster a dialogue among sectors, collect a diverse set of voices, and stimulate the discussion around two main questions: 1) what is the socio-economic impact of new technologies for your sector? And 2) what are the policy and political implications of the future of work for your sector? * We are deeply thankful to the FES for its financial support, and particularly to Carlos Cabrera, Ebert Foundation´s Economic Dialogue Coordinator in Mexico, who generously contributed with his expertise to the workshop’s design and made it possible for us to have several of the participating actors at the event. We also thank the Center for International Studies and the Seminario de Estudios sobre el Futuro at Colmex for their contribution in terms of the event’s logistics. Johanna Moeller, research intern at the FES, and Calep Pimienta, postdoctoral fellow at PolDigWork, provided valuable assistance. 4

Exploring Key Actors’ Perceptions on the Socio-Economic and Policy Implications of Workplace Transformation Our objectives of the workshop were multiple, but our focus was on identifying common topics or even common visions among participants, as well as points of divergence and controversy. We were particularly interested in four main transversal topics: First, the multiple dimensions of inequality created, enhanced, or maybe also mitigated by technological change. The usual point of departure in this regard is discussions about digital divides. Who has access to and benefits from using new technologies, and who is left behind? Such digital divides can be further split between genders, age groups, or maybe also ethnic groups, especially when we think about vulnerable minorities. There are also significant spatial inequalities, for instance, those between urban centers and the countryside or the extensive Mexican macro-regions. The second transversal topic is the divide between the formal and informal sectors. While this divide is notoriously hard to define, with many grey zones in between, it is also clear that informality matters. The informal sector in Mexico is large(around half of the labor force according to INEGI 2020a), and people working informally have little or no access to social protection. The effect of digitalization on formalization in the labor market is still uncertain. In turn, the formal-informal divide relates to the third transversal topic, the role of the state in managing the technological revolution. Against the background of multiple divides and inequalities, government actions could help or worsen technology’s impact on society. State capacity is crucial in assessing changes, regulating new technology, compensating losers, and incentivizing positive developments. Our final transversal topic was the role of narratives and perceptions of technology and its links to the social world. How accurate are the tropes we tend to tell ourselves about technology? Are we prone to scandalizing, or are we perhaps still underestimating the sheer size of the shock? Do we forget essential parts of the stories and focus on very visible companies, sectors, or aspects of technological change while ignoring others? We summarize our findings in the following sections. We start with a discussion of what participants thought about the socio-economic impact of technology. In the next section, we will summarize our participants’ positions on the policy and political implications of the future of work. In the last section, we analyze and evaluate our findings in light of the transversal questions we posed above. 1 The workshop took place on March 9th, 2023, at El Colegio de México. One part, a roundtable discussion was streamed live and is available here: https://www.youtube.com/watch?v=IWD46fIEmLE 5

FRIEDRICH-EBERT-STIFTUNG- The Future of Work in Mexico 1 TECHNOLOGICAL CHANGE AND WORKPLACE TRANSFORMATIONS IN MEXICO Current technological transformations are developing in several dimensions of the economy with heterogeneous workplace impacts(e.g., Introduction in Busemeyer et al. 2022). We focus our exploration of crucial actors’ perceptions on the following major types of technological change: a) Digitization, conceptualized as the conversion of information into a digital format and its implications for harvesting big data and excluding unmeasurable or‘analog’ content. b) Artificial intelligence and algorithms that capable of processing and analyzing this information, often without direct human interference after the initial stage of programming and training of those algorithms. c) Processes of physical and virtual automation, usually with the help of bots(algorithms) and robots, with all the implications for the aggregate number of jobs, old versus new tasks, as well as the nature and quality of jobs. d) Platform economies as new forms of engagement between those who run the platform, those who produce, sell, or buy on those platforms, and the implications of this process for the creation of new roles(e.g., prosumers), new dependencies, new forms of surveillance and new forms of productivity. e) Processes of digitalization which include all the dimensions mentioned above and their consequences for the quantities of work, new modalities, and qualitative characteristics of work. How are these processes taking place in the Mexican labor market? Although, as in other middle-income countries, the global pandemic accelerated the adoption of technologies and new work arrangements, in Mexico, their scale is still limited by 2023. According to the Survey on the Economic Impact Generated by COVID-19 on Companies in Mexico(INEGI, 2020), 5.7% of economic units have adopted teleworking, particularly those in the sectors of business administration, systems engineering, computer science, accounting, legal, finance and information technology. Available estimates suggest that those who benefit primarily from teleworking are people with high educational levels, stable employment relationships, and professional, managerial, and administrative occupations. According to available evidence, in Mexico, the probability of teleworking is only 10.6%(Leyva& Mora, 2021). 2 Nonetheless, there are already visible trends of adjustment in workplace arrangements. Recent studies estimate that 40% of Mexican companies are starting to provide flexible working options to their employees: 19% have access to home office schemes some days per week, and 14% can work remotely at any time(Borjas, 2019). Since young people who were born between 1995 and 2010 will be a 100% digital generation, scholars and informed observers expect an expanding pool of workers with developed skills and fluidity to join the Gig Economy(also called the sharing economy, the fourth industrial revolution, platform capitalism, electronic-computer capitalism, connective capitalism, digital capitalism, siliconization, the age of techno-imperialism(Radetich, 2022). The extraction of algorithmic data is central to production in a growing set of companies. It has generated new relationships between workers, consumers, and other capitalists, giving way to business models that capitalize on large amounts of data(Srnicek, 2018), potentially creating new job positions. In Mexico, the scarcity or absence of regulation regarding these relatively recent digitized modes of production might facilitate flexible work arrangements but also lead to precarious and unprotected labor conditions. Significant adjustments in the production processes might modify the demand for specific skills, creating new jobs but also potentially displacing workers in certain occupations. 2 Leyva and Mora(2021) find that Mexico City concentrates 19% of the jobs that can be done remotely, followed by the state of Nuevo León(14.2%) and Querétaro(12.2%). Other states offer a smaller percentage of potentially remote jobs, such as Guanajuato(7.6%), Guerrero and Veracruz(7.3%). 6

Exploring Key Actors’ Perceptions on the Socio-Economic and Policy Implications of Workplace Transformation While precisely forecasting which sectors will experience more significant disruption based on available indicators is quite challenging, recent contributions have started estimating the risk of automation in Latin America. Figure 1 shows the proportion of jobs with a high risk of automation by region in Mexico according to the calculations of the Center for Distributive, Labor, and Social Studies(CEDLAS) based on microdata from the Mexican National Survey of Household Income and Expenditure(ENIGH). The plot illustrates the varying risk across Mexican states(albeit moderate in this estimation), linked with the predominant skills in the distinct subnational economies. 3 When discussing these estimates, Brambilla et al.(2021, 2022) note that technological change is more likely to affect national income distributions than overall rates of employment in Latin America. This result is because“unskilled and especially semi-skilled workers are likely to bear a disproportionate share of the adjustment costs since the automatability of their occupations is higher compared to skilled workers” (Brambilla et al., 2022, p.251). Perhaps the most visible dimension of the technological revolution in Mexico is the expansion of the platform economy. According to the National Occupation and Employment Survey(ENOE Survey, from the National Institute of Statistics and Geography, INEGI 2021), in Mexico, 250,000 people are working as delivery drivers. As noted, the implications of the increase in platform jobs for labor security and social protection are still unclear. The formal economy generates registered jobs with a minimum of legal protections and associated tax contributions. Informal jobs lack such labor protections and are sometimes performed in unregistered economic units (INEGI, 2020a). Beyond a binary formal-informal conception, the characteristics and consequences of these forms of economic subsistence conform to a complex sector with varying dynamics referred to as solidarity economies, popular economies, feminist economies, alternative economies, and economies from below(Rosales, forthcoming). 4 4 Alba(2023) proposes the use of the term popular economy , instead of informal sector . The author notes that the latter term comes from the Latin verb secare , which means“to cut” implying that there are two separate parts in the economy, one formal and the other informal. However, in reality there is no such distance between “formal and informal people, nor is there an a priori unity between the various segments of the supposed informal sector since they are extraordinarily heterogeneous. Thus, the popular economy alludes to the great heterogeneity of self-employed workers, micro-enterprises, street vendors, etc. that condense popular sectors that do not have access to the informal economy.” 3 The authors consider the risk of automation across occupations based on a previous study(Arntz et al. 2016, 2020). They then use the number of workers in each occupation as weights(Brambilla et al. 2021) Although moderate in this estimate, this risk is higher than the expected for OECD countries(see Arntz et al.(2016)). 7

FRIEDRICH-EBERT-STIFTUNG- The Future of Work in Mexico Figure 1. Proportion of jobs with high risk of automation by region in Mexico Risk of automation 0.193 0.163 Powered bi Bing ©GeoNames, Microsoft, TomTom Source: Authors’ elaboration with estimates from Brambilla et al.(2021) from the Center for Distributive, Labor, and Social Studies(CEDLAS), based on microdata from the Mexican National Survey of Household Income and Expenditure(ENIGH). Recent workplace transformations have also started to show gendered patterns. Currently, women tend to be significantly underrepresented in platform jobs. A study among delivery drivers found that at least 80% of the delivery workers are men(Colmex 2021—estimate using a sample of 1,000 workers, Inmujeres, 2021). Another investigation carried out by Radetich(2022) on platform drivers in Mexico found that only 6.4% were women(out of 315 drivers who participated in a survey). These gender inequalities in the platform labor market reveal a significant digital divide and a continuation of the low-level trend of female labor force incorporation (particularly in the formal sector). Another study pointed out that, in 2019, 59.2% of women in Mexico did not use a computer, laptop, or tablet either at home or outside of it(vs. 54.6% of men). Leyva and Mora(2021) point out that 15.3% of the work carried out by women could be done remotely, yet inequality in women’s employment persists. The main barriers are the occupational structure in Mexico and the lack of robust safety nets, which would balance the productive and reproductive work that women perform inside and outside the household (e.g., early childhood education and care policies; See Altamirano, 2020). 5 This gap deepens in rural areas, since it is recorded that 77.7% of rural women did not use any of these devices(78.1% men). This means that the impact of women's participation in technological jobs at the national level is also reduced. CEPAL suggests that inclusive digitization would be a favourable action for the Mexican case where 32.1% of women who do not have digital access. 8

Exploring Key Actors’ Perceptions on the Socio-Economic and Policy Implications of Workplace Transformation 2 PARTICIPANTS AND PROCEDURE The individuals included in the focus group were 20 experts(17 men, 3 women, aged 30-65) working in sectors with varied exposure to technological change (7 researchers in academia, 6 labor union representatives, 4 consulting experts in the private sector, 1 civil society expert, 1 public sector expert). On the basis of the focus group, we also had a roundtable discussion with another 2 female and 4 male experts. 6 The organizing team contacted prospective participants via email and informed them about the content and objectives of the workshop. Participants were told that their interventions would be anonymous and confidential in our report. We prepared a topic list and questions to guide the discussion, which the authors moderated. The meeting lasted approximately 5 hours, including breaks. The total conversation time among participants was around 3 hours total. Two assistants and the organizers transcribed participants’ remarks for later analysis. Before we start reporting results, it is worth highlighting who was represented and who was not. We managed to reach out to most of our target sectors. Nonetheless, some actors could not participate—either because they were too busy or not willing to take part in the discussion group—. For instance, we could not include representatives from the agricultural sector and the countryside. While many experts also mentioned these sectors in the discussions, we had no direct representatives among us. Some representatives of big companies might have been busy, but we also noted that representatives of big tech companies tended to be more elusive. Beyond the findings of our expert focus group, the lack of participation of these key agents of change limits the prospects of having broad social discussions about the implications of workplace transformation. Similar things hold for the gender balance, which was also due to a number of last-minute dropouts. 6 See here: https://www.youtube.com/watch?v=IWD46fIEmLE&t=18s 9

FRIEDRICH-EBERT-STIFTUNG- The Future of Work in Mexico 3 RESULTS THE SOCIO-ECONOMIC IMPACT OF NEW TECHNOLOGIES In the first part of the conversation, we asked our participants to assess—from the perspective of their organization, their sector, or their individual opinion— the socio-economic impact of technological change, not only in objective quantitative and qualitative terms but also in terms of subjective assessment, perhaps even emotional responses. We start with these overall assessments before we turn to our transversal questions. Overall assessment of the socio-economic impact A clear majority of our participants said that they already feel the shock, and they think it will only increase in the following decades. Physical automation already happens, for instance, in the case of the automobile industry in central and northern Mexico or self-check-out cash desks in supermarkets. This implies that companies look for different skill profiles and might hire fewer workers in the near future. In general, the employment effects of foreign direct investment might change and be smaller if companies invest more in robots and automation. While the expected size of automation depends on how we estimate its impact, some prominent studies have calculated that about forty percent of all formal-sector jobs are susceptible to automation risks(Cárdenas& Ruelas, 2018; Silva Taylor et al., 2022), similar to what others studies have found for Latin America(McKinsey Global Institute, 2017)(McKinsey Global Institute, 2017) and the OECD(Frey& Osborne, 2017). Remarkably, while the number of jobs at risk has remained roughly stable over the last 15 years, the types of occupations have changed. This suggests that occupations and sectors adapt and that being exposed to risks of automation is not the same as jobs being expunged. For Europe, detailed analyses on the level of specific tasks(Arntz et al., 2016) have shown that the actual risks of redundancy are much lower since occupations often combines tasks that are easier to automate with those who are less at risks. Available estimates for Latin America suggest a higher risk(16.7% on average for the six largest economies in the region according to Brambilla et al.(2022, p.243), compared to an average risk of 9%. in OECD countries in Arntz et al.(2016)). In our discussions, it was noticeable that negative scenarios and, at times, even pessimistic feelings were more frequent than positive or even optimistic assessments, mainly when we talked about the most recent bouts of technological change, such as artificial intelligence(AI). Nonetheless, there were notable exceptions. Some experts delivered a more sanguine picture of the near future. For instance, technological change has not—so far—led to solid and long-term disruptions in the activties for major sectors. There seems to be remarkable continuity in the qualitative properties of the Mexican labor market(Bensusán Areous, 2017; Bensusán Areous& Florez Vaquiro, 2020), with important exceptions, especially in the first two years of the pandemic. Even those experts, however, admitted that a considerable part of the problem is that we systematically know less about new jobs, activities, and sectors. The entire statistical architecture is biased towards the traditional formal sector, and big tech companies do little to increase transparency for their businesses. As an example, it is hard to assess the size of the platform or‘gig’ economy. While prominent companies such as Uber or Rappi are on everybody’s mind, reliable statistics are hard to come by. Recent studies report estimates of a sizable growth of platform businesses in the last three years(31 million delivery service users and a market value of 650 million pesos in Mexico by mid-2021), with approximately 500,000 persons registered as delivery workers, according to IMSS(Silva and Jaramillo-Molina, 2022). Also, for other OECD countries, estimates must be treated with caution. For the average European country, the gig economy rarely exceeds 1-3 percent of the workforce (Drahokoupil& Piasna, 2022; Huws et al., 2018), and even for those active on platforms, those incomes are relatively minor compared to other economic activities. The picture for specific sectors, such as ride-hailing or food distribution, might be very different. Prominent platform companies companies such as Uber or Rappi have attracted a lot of academic and policy attention(e.g., Alba et al., 2021; Herrera et al., 2019; Silva Taylor et al., 2022). These studies have shown that for most platform‘workers’ —especially since the pandemic—, these jobs are the main form of permanent economic activity and income. 10

Exploring Key Actors’ Perceptions on the Socio-Economic and Policy Implications of Workplace Transformation The question remains as to whether those highly visible companies are representative of a larger trend, perhaps being the vanguard of a platformization of entire sectors or a relatively limited segment compared to other parts of the labor market. While most of the participants thought that platform work would only grow in the future, for countries such as Mexico, we do not yet fully understand how big the gig really is and how much growth potential there will be. More generally, some experts were skeptical about the predominant narratives about automation. While the shock is real, it should not be exaggerated, and we should instead think about how to retrain people and make sure that nobody is left behind rather than lamenting impending doomsday scenarios. The future of work and the pandemics also have changed important qualitative aspects of the Mexican labor market. The pandemic, in particular, has pushed digitalization and made informal muddling acceptable in numerous ways. It has, for instance, made working from home acceptable in many sectors(Giray Aksoy et al., 2022). This has created new challenges for regulators, some of which the Ley de Teletrabajo(a law regulating home office schemes) aimed to address(see below). Either as a response to the new law or the subsiding of the pandemic, many sectors have reverted to the status quo before 2020. Still, there are also many sectors and types of employment(e.g., self-employed digital workers) who remain firmly in their home offices. Many observers saw the quality of working standards in the new economy critically. Platforms and algorithms lead to new, unheard levels of surveillance. Time pressure for people working for platform companies like Uber or Rappi is a constant issue, and mental health is a serious concern. Some participants emphasized that rather than freeing us from heavy work, it seems that algorithms are capable of overworking people even more(Huesca et al., 2022), making calls for a“right of laziness” a remote possibility. Some favorable aspects should not be neglected, though. Technology might tip the balance even further away from platform workers, but technology can also be used in their favor, for instance, reporting abuse or security issues. New technologies have also created numerous new jobs(see also Busemeyer and Glassmacher in Busemeyer et al. 2022), from YouTubers to professional dog walkers. There is also increasing demand for medium- to high-skilled workers(e.g., Chua, 2018 expects a race with- rather than against the machines. See also Gallego, Aina et al., fc.). Inequalities and the Informal Sector Most experts agreed, however, that the shock is very asymmetric, hitting some sectors, occupations, and tasks much harder than others. Some see a pattern of polarization in the workforce: while typical medium-skill routine tasks are very much in danger of being automated, both very high skill levels and straightforward physical tasks are much more demanding for robots or algorithms to do efficiently. This will affect certain sectors more than others(e.g., banking versus cleaning streets). We should, however, also be aware that these processes are highly dynamic, and what seems inevitable today might rapidly change in the following decades. Just as an example, one big hope to mitigate the consequences of deindustrialization was the service sector and social services in particular(Bell, 1974). While it is true that some personnel services are challenging to automate, for instance, in the care sector, even in these sectors, the use of robots is expanding in some countries(Eggleston et al., 2021). Other personnel services such as gastronomy or retail are already being automated in Mexico. Several experts mentioned the significant gender divides created by digitalization in combination with the pandemic. For instance, platform work in ride-hailing and food distribution was one of the few sectors growing despite the pandemic. Still, these are predominantly male activities(Radetich, 2022). For instance, Alba Vega et al. (2021) found that more than 75% of food distribution workers are male. As noted above, women are also more vulnerable when it comes to social protection and skill attainment. Another clear divide is between different age cohorts. Young people entering the labor market may have better training than previous cohorts, but they face even higher labor market risks. They were also particularly hard hit by the pandemic. A particularly frequently mentioned divide was between urban and rural regions of the country and the vast differences between macro-regions. Several experts highlighted the fact that there is not one, but several‘Mexicos.’ While industrialized and growing regions in the center and the north of the country face already high levels of penetration when thinking about new technologies, other parts of the country in the South, for instance, are being left behind by much of the technological transformation. Most attention, however, fell on the differences between formal and informal activities. There is much academic discussion about the socio-economic and political relevance of the informal-formal divide(Altamirano et al., 2022; Baker& Velasco-Guachalla, 2018; Berens, Sarah& Kemmerling, Achim, 2019). As already discussed, some authors have therefore preferred to refer to the informal sector as the popular economy(Alba Vega 2023), highlighting the fact that there are many grey zones between formal and informal activities and that this sector is an essential pillar of the Mexican economy. Setting these definitional issues aside, formality matters in terms of access to labor market regulation, social protection, and the obligation to pay taxes, licenses, etc. The role of new technologies in shifting this divide is somewhat ambivalent. Take platform work as an example. While Uber drivers are registered and pay taxes, they have little access to social benefits and employment protection. The reason is that Uber et al. have created a new model of labor relations. They claim that those offering services on their platforms are independent, self-employed entrepreneurs rather than dependent employees(Bensusán Areous, 2017; Silva Taylor et al., 2022). There is a significant legal debate about this topic in many countries(e.g., Crouch, 2018; Rani& Grimshaw, 2019), and it will be particularly informative to watch the public discussion on this issue in Mexico. 11

FRIEDRICH-EBERT-STIFTUNG- The Future of Work in Mexico In the case of platforms, payment of the Value Added Tax (IVA) is mandatory, and those that generate profits for workers must withhold the Income Tax(ISR). However, many companies have decided to charge said taxes to the final consumer. And in the case of those that generate profits for workers, they have seen fit to transfer it to them under ambiguous concepts. In Mexico, the approximate percentage of taxes ranges from 16% VAT, between 2% and 8% ISR, 2% and 10% for lodging services, and 0.4% and 5.41% for the sale of goods and provision of services such as Mercado Libre and Amazon. Currently, the Tax Administration Service(SAT) has registered 186 companies providing digital services in the Federal Taxpayer Registry. Some of them are the most prominent players in the digital service sector, such as Didi, Uber, Amazon, Airbnb, Zoom, Apple, Domestika, and Spotify. What seems clear is that platform workers tend to work mainly for one or few companies, have repeated interactions with the owners of the platforms, and have few other forms of income. They also face new types of risks(for instance, traffic accidents, insecurity, and, of course, income losses due to health issues) while having little social protection to show for themselves. In the words of Carlos Alba, most of these workers have one foot in the 19 th century with little labor or social protection, while the other foot firmly rests in the new economy of the 21 st century with new types of risks(Alba, 2023). Narratives about Technological Change Our final transversal topic was the importance of narratives about technology. If we look at the discussions, one key narrative was about mechanisms of control and surveillance, invoking the specter of surveillance capitalism as described by Zuboff(2019) and others. There might also be the possibility of using technology to level the power imbalance, but it seems that this is an underdeveloped idea in the Mexican context so far. Surveillance also plays a considerable role in concerns about data privacy and cybersecurity. While some experts hailed Mexico’s relatively liberal stance towards digitizing information, others were clearly concerned about big tech’s use of this type of information. We are only beginning to understand how much data is harvested and how much this has shifted the informational asymmetries between individuals, commercial companies, and the government. Some authors argue that these shifts fundamentally undermine the functioning of public goods provision and social and welfare arrangements(Iversen& Rehm, 2022). On a more abstract level, several experts complained about the fact that neither in society nor in government is there any actual national‘project,’ let alone a digital agenda to confront the technological changes and their socio-economic effects. Hence, we turn to our second main question: the role of public policy in mitigating the potentially negative sides of technology. Policy and political implications We split this section into two parts. The first part deals with the range of policy areas and how they are affected by the future of work. The second part deals with political implications and the questions of limitations and obstacles to adjusting public policy to the new realities. Technology experts emphasized the implications of technological change for infrastructure management and availability (ranging from access to essential commodities such as water and electricity to more technical aspects such as server ownership and access point control and censorship). Ever since the extensive privatizations in the 1980s and 1990s, the default mode seems to be for larger companies to run private infrastructure, which creates a whole suite of regulatory issues(e.g., Juárez et al., 2022). Infrastructure is also crucial in assuring who gets access to what quality of services, who controls the data, and who makes the infrastructure safe against attacks and issues of cybersecurity. Regulation issues relate to infrastructure ownership, but it goes beyond. Even staunch libertarians have recently called for a moratorium in the development of AI-power chatbots such as GPT4(De Vynck, 2023). These statements show that regulators chronically and dangerously lag behind the most recent technological developments. Among the dangers that need urgent codification of ethical and regulatory guidelines are the type of racial, gender, and, in general, discriminatory biases in algorithms. Such biases have led some experts to speak of artificial dumbness rather than artificial intelligence(Juárez et al., 2022). While these positions mainly intend to spark debate, they also illustrate framing disputes and, on a deeper level, the possibility of creating myths around technology. While technology can be a force for good, the balance clearly tilts towards already powerful interests. So far, platforms have successfully eschewed most forms of regulatory scrutiny. Regulation needs to include improved transparency in the use of algorithms and business numbers of big tech companies to see how many people they serve, how they use data, and how much profit they make. Profitability is also crucial for fiscal revenues, which, in turn, could be used to compensate some of those people clearly losing against the machines. So far, taxing the new economy has been a challenge worldwide (Gelepithis, 2022). Amazon, for instance, pays a fraction of profits each year in countries like France and Germany. Given that the digital economy has created a footloose form of capitalism, taxation needs to occur on many different levels. Indeed, some Mexican communities have successfully introduced local charges for Uber. In the state of Nuevo León, there was an initiative to charge 3% to transfer services from platform companies such as Uber or Didi. However, digital companies can assign that charge to the drivers themselves. Although these measures might elicit positive interpretations, the reality is that they reveal digital platforms’ power capacity to circumvent tax compliance and their ability to transfer the burden to consumers via alternative pricing strategies(e.g., dynamic rates). 12

Exploring Key Actors’ Perceptions on the Socio-Economic and Policy Implications of Workplace Transformation On the international level, there are ample debates about a digital service tax and mechanisms to combat harmful forms of tax competition(for instance, the OECD’s so-called BEPS, the base erosion and profit shifting— initiative). The U.S., arguably the country benefiting most from significant tech revenues in the Western hemisphere, has recently accepted the introduction of a global minimum of 15% tax on corporate income. In Mexico, there is so far a relatively muted debate on taxing the new economy(see Cabrera et al., 2021). Participants did not discuss redistributive or compensation measures at length. When prompted about social policy and taxation instruments, participants quickly gravitated to non-contributory measures such as universal cash transfers designed following a Universal Basic Income(UBI) scheme. The apparent centrality of UBI in the participants’ top of mind resonates with similar narratives and debates in Europe(e.g., Chrisp& Martinelli, 2022). However, we know that the new risks associated with digitalization will require an integral bundle of social policy interventions, starting with access to comprehensive healthcare services. The new forms of work also call for new labor standards and regulations in the digital economy(Bensusan and Santos 2021). The discussions about the Ley de Teletrabajo highlighted the need but also the problems related to this issue. Jurisprudence, so far, has been very slow in changing to the new realities. It has not been easy to convince legislators and judges that platform work is a form of dependent work, but there are notable exceptions(e.g., Belgium) that might serve as a template. A big problem of regulating labor standards in Mexico is, of course, the many evasive actions the target population can take. If strengthening labor codes leads to more informal activities, the main effect of regulation evaporates. Some experts have argued that the Ley de Teletrabajo is an example, as it might incentivize companies to shift back to previous work formats or refuse to implement regulations. Therefore, government interventions require careful calibration to avoid such problems. Of all policy areas, education provoked most discussions (see also Schatan, 2018). The debates quickly were bifurcated in two directions. On the one hand, representatives of industries and some academics mainly focused on specific skill needs dealing with new technologies in advanced industries such as automobile, aerospace, or financial services. Here, competence needs are so extreme in some parts of the country(e.g., Baja California and Queretaro) that companies have started ‘importing’ skilled labor force from countries like the Philippines or Bangladesh. On the other hand, we had discussions about how to adapt the education system in general. Representatives of government have highlighted the need to focus on vulnerable segments such as newcomers in the labor market, young people in general, the rural population, and women. Tech and government experts highlighted the importance of also improving tertiary education. There is a consensus that Mexico needs people with general and creative skills to participate in the technical revolution and to shape it rather than implement and administer it. Several experts referred to the importance of looking at the ‘bigger picture.’ A frequently made connection was not only to industrial policy but also to environmental policies and the issue of sustainability. The supply chain connects agricultural producers to transport, gastronomy, and, eventually, food distributors. While digitalization improves the connections, it also anonymizes them further. A crucial pending issue is to jointly reflect on the twin transitions of digitalization and mitigating climate change. Finally, state capacity is crucial for accompanying the digital transition. State agencies in several Latin American countries often do not even dispose of the statistical information and a statistical architecture to process all this information, let alone the human capital to use this information strategically. According to our participants, in Mexico, knowledge of new technologies, regulations, as well as the social dimension of the future of work is minimal. Therefore, a national digital strategy is lacking. There are some legal initiatives worth mentioning(e.g., the Ley de Teletrabajo and the latest attempts to regulate platforms and online work), as well as new laws raising the minimum wage and the number of vacation days for those working in the formal sector. However, it is clear that some previous attempts also backfired, for instance, in the education sector. One former government official mentioned the fact that Mexico had bought a costly satellite system for distance education in the 1990s but that this technology was never really used to total capacity. These points brought us to the final round of discussions. Political obstacles to better policies If improving state capacity is perhaps one of the most fundamental prerequisites to dealing with the future of work effectively, weak state capacity is probably the most critical political obstacle. Historically, Mexico’s governments have had problems with generating sufficient public revenue, recruiting staff meritocratically, and delivering high quality public services. In recent years, state capacity has further deteriorated, especially in relation to the ever-increasing gap in technological know-how compared to the private sector. Weak state capacity affects the digital transition in numerous ways, from controlling the borders to designing intelligent policies. Put under duress, the state has always responded in a two-pronged way: regulating and taxing the formal sector while tolerating numerous informal or perhaps illegal activities((Alba Vega et al., 2015; Holland, 2017; López-Cariboni, 2018). Yet, the gap becomes particularly obvious vis-à-vis big national and international tech companies. Already, the privatizations of the 1990s showed that the Mexican government had severe issues regulating new private quasi-monopolies in telecommunication. The entry of new tech companies might lead to similar results. 13

FRIEDRICH-EBERT-STIFTUNG- The Future of Work in Mexico Tech companies themselves do not only have better-trained programmers and lawyers, but they often also have the consumers/ prosumers on their side. It is not only a matter of quality and price, though many consumers prefer cheaper services to better labor standards. It is also soft factors that make tech companies popular. Through public campaigns, tech companies create an image of providing efficient, cheap, and modern services despite the fact that they explicitly avoid regulation or taxation(Davies et al., 2022). They make a host of myths around themselves(Silva Taylor et al., 2022), for instance, giving those who work on their platforms flexibility, autonomy, and freedom, even if the actual working standards often look very different. This image campaign makes new investments, such as Tesla in the northern state of Nuevo León, a sign of success for local governments, with little reflection on what types of socio-economic consequences will follow and under what kind of conditions production in these new plants will occur. In part, this imbalance results from the fact that civil society in Mexico is not very strong. Trade unions tend to be weak, especially in the new sectors, and the image of unions in Mexican society is not always positive. For all these reasons, few platform workers seem to favor the introduction of unions(Rosales, forthcoming). Still, there are notable exceptions, such as the Unión Nacional de Trabajadores por Aplicación. Given the lack of social mobilization, many legal initiatives do not really come into being because of intense pressure from citizens, unions, and party members. Instead, many legal reforms are introduced‘from the side and from above,’ as Graciela Bensusan stated in the roundtable discussion. 14

Exploring Key Actors’ Perceptions on the Socio-Economic and Policy Implications of Workplace Transformation 4 DISCUSSION AND CONCLUDING REFLECTIONS The workshop was quite valuable in bringing together different voices, forms of expertise, and knowledge. While we kept the topic deliberately broad, we saw agreement on many significant issues as well as some divergent opinions. Most experts agree that the future of work is a crucial topic whose salience will only rise in the following years. Negative projections dominated the discussion, although this also could speak to the type of voices we collected as well as—to some degree— the professional need in many sectors to highlight risks rather than opportunities. In other areas, there was not so much dissent but a divergence in priorities. Education seems a good example, where some experts found medium-skill technical workers to be a key concern. In contrast, others focused more on the needs of young people or on creating IT intelligence for future rounds of technological progress. As for our four transversal topics, our picture has only partially sharpened following the workshop. Quite clearly, the shock is asymmetric, affecting some regions, some sectors, and some vulnerable groups much more than others. Inequality, already a central issue in Mexico, will both affect the digital transition and will be enhanced by it. For the informal-formal divide, the image gets blurry. While some trends could lead to a higher degree of formalization (e.g., by registering and taxing platform work), this requires political will and intelligence. Otherwise, digitalization might mainly mean that, once again, the informal sector will be the primary way the Mexican economy absorbs shocks. Whether or not the Mexican state and society will react strategically when confronted with new technologies also depends on the narratives we tell ourselves. If we act like rabbits staring in the snakes’ eyes, fascinated only by doomsday scenarios, we will achieve very little. Similarly, if we naively believe what new technologies make us feel, the outcome will be the same. Only an active and informed society, as well as the governments it chooses to elect, will make a difference between being modernized or modernizing(see conclusion in Busemeyer et al., 2022). We end with some recommendations. 1) Break down policy silos The workshop has shown how important it is to bring experts and representatives of different sectors together. Policy silos are a pervasive problem in many political systems(Popiel, 2022). In our case, there are clearly weak networks between those experts working on technology, hard skills, and technocratic ideas about regulation on the one hand and experts in social and labor affairs on the other. As an example, a recent and very accessible book about robots and regulation does not include perspectives from the field of social and labor market policies(Juárez et al., 2022). In turn, most discussions on those policy issues do not go deeply into technological complexities. These two sides need to be brought together in order to make tech discussions more sensitive to the social reality and turn discussions on social and labor issues more tech-savvy. Hence, we need to create a dialogue between sectors and disciplines, bringing more voices from the industry and tech companies. Even within sectors, there are interesting differences between disciplines. For example, many economists tend to be much more sanguine or pragmatic about technological change(Mokyr et al. 2015). They argue that historically speaking, doomsday scenarios of technology have always been exaggerated. In the case of Mexico, in particular, the argument would be that labor costs are low, so the pressure to automatize is not immediate or substantial. Contrast this to tech experts and(to a lesser degree) representatives of business who are more alarmed. They tend to look more at consumer demand and how it tends to favor automation. Even more pessimistic are those working on platforms and the gig economy. Again, fostering discussions between these different perspectives will allow us to get a better and more holistic picture. 15

FRIEDRICH-EBERT-STIFTUNG- The Future of Work in Mexico 2) Avoid tunnel vision and narrow discussions The workshop showed that while we need to work on concrete solutions for specific problems—e.g., how to regulate labor relations on platforms— we must not forget more significant questions. What about basic infrastructure, access to electricity/energy, and how do we ensure more autonomy from work and technology instead of getting more and more outworked by them? Raising one’s gaze is also essential to see certain types of biases. Many discussions and political debates focus on particular platforms, companies, and problems. We learned a lot about companies like Rappi or Uber but little about Mercado Libre or other platforms. The latter equally took off during the pandemic and have made many physical retailers either redundant or moving into the cloud. The new labor forms must recognize workers’ fundamental rights and advance regulations protecting their labor conditions. For women, crucial requirements are decent working conditions and being able to combine paid and unpaid work in violence-free environments. Related to this, we need to avoid fixation with certain narratives or perhaps even myths. The debate about new forms of social policy in the wake of digitalization and automation is a fascinating example. Arguments on this topic almost automatically invoke discussions about UBI, even in middle-income countries. While there is nothing wrong with discussing UBI, it is clear that it is not a panacea. In Mexico, for instance, a move towards universal health care would be at least as important as making cash transfers more readily available. An agenda in favor of new instruments of redistribution, social policy, and labor regulation requires more thoughtful strategies in how to advocate for them and to build larger support coalitions. In many countries, there is low support for the distribution of unconditional social policy benefits in cases of no apparent need or no work intention(Schwander& Vlandas, 2020). A crucial political challenge is to achieve a consensus among voters around social policy universalization as a reasonable intervention to address social risks. 3) Create agency and capacity To adequately address the forthcoming challenges associated with technological transformation, Mexico requires a clear digital strategy. Such a project would require a professional and autonomous state administration as well as an active and vigilant civil society. While perhaps hard to achieve overnight, gradual steps towards investing in public administration and fostering NGOs, watchdogs, and trade unions would go a long way in this direction. A capable government is not a small government but one that knows how to deal with profound transformations. In a nutshell, there are two approaches to technological modernization(see Busemeyer et al. 2022). Either governments merely and belatedly react to the digital transition that is going on in private companies and markets, or the state administration and civil society embrace new forms of technology critically and modernize themselves. The digital revolution can deliver many innovative solutions, from e-health to robot judges, smart tax offices, or internet regulators. It depends on informed coalitions and strategic political action to make the state smart enough to handle the future of work. 16

REFERENCES Alba, C., Graciela, B.,& Vega, G.(2021). El trabajo del futuro con derechos laborales. Diagnóstico y estrategia de política pública para el reconocimiento de derechos laborales de trabajadores de plataformas digitales. El Colegio de México. https://ces.cdmx.gob.mx/storage/app/media/publicaciones2020/Estudio_d erechos_laborales.pdf Alba Vega, C., Lins Ribeiro, G.,& Mathews, G.(2015). La globalización desde abajo: La otra economía mundial. Fondo de Cultura Económica. Alba, C.(2023, de próxima publicación) Economías populares en México en tiempos de Covid-19. El Colegio de México. Altamirano, M.(2020). Política social e igualdad de género en México, 2012-2018. Foro Internacional, 755–789. https://doi.org/10.24201/fi.v60i2.2738 Altamirano, M., Berens, S.,& Deeg, F.(2022). Varieties of Economic Vulnerability: Evidence on Social Policy Preferences and Labor Informality from Mexico. Latin American Politics and Society, 64(2), 139–168. https://doi.org/10.1017/lap.2022.9 Arntz, M., Gregory, T.,& Zierahn, U.(2016). The Risk of Automation for Jobs in OECD Countries: A Comparative Analysis(Working Paper 189; OECD Social, Employment and Migration Working Papers). OECD. https://doi.org/10.1787/5jlz9h56dvq7-en Baker, A.,& Velasco-Guachalla, V. X.(2018). Is the Informal Sector Politically Different?(Null) Answers from Latin America. World Development, 102(C), 170–182. Bell, D.(1974). The Coming of Postindustrial Society. A Venture in Social Forecasting(Heinemann). Bensusán Areous, G.(2017). Nuevas tendencias en el empleo: Retos y opciones para las regulaciones y políticas del mercado de trabajo. https://hdl.handle.net/11362/42613 Bensusán Areous, G.,& Florez Vaquiro, N.(2020). Cambio tecnológico, mercado de trabajo y ocupaciones emergentes en México. https://hdl.handle.net/11362/46181 Berens, Sarah& Kemmerling, Achim.(2019). Labor Divides, Informality, and Regulation: The Public Opinion on Labor Law in Latin America. Journal of Politics in Latin America, 11(1), 23–49. Borjas, M.(2019, July 3). La gig economy llega para quedarse: ¿Está México preparado para este fenómeno? eSemanal- Noticias del Canal. https://esemanal.mx/2019/07/la-gig-economy-llega-para-quedarse-esta-m exico-preparado-para-este-fenomeno/ Brambilla, I., César, A., Falcone, G., Gasparini, L.,& Lombardo, C. (2021). The Risk of Automation in Latin America. Documentos de Trabajo Del CEDLAS. CEDLAS-Universidad Nacional de La Plata, No 281. Brambilla, I., César, A., Falcone, G., Gasparini, L.,& Lombardo, C. (2022). The Asymmetric Risks of Automation in Latin America. Desarrollo Económico, 62(235), 234–253. Brynjolfsson, E.,& McAfee, A.(2014). The Second Machine Age: Work, Progress, and Prosperity in a Time of Brilliant Technologies. W. W. Norton & Company, Inc. https://www.amazon.de/-/en/Erik-Brynjolfsson/dp/0393350649 Busemeyer, M., Kemmerling, A., van Kersbergen, K.,& Marx, P. (2022). Digitalization and the Welfare State. Oxford University Press. https://doi.org/10.1093/oso/9780192848369.001.0001 Cabrera, C., Casar, J. I., Raymundo, M., Campos, M., Urzúa, C. M., Vázquez, D.,& García, A.(Eds.).(2021). Posibilidades para una reforma al impuesto sobre la renta de las personas: Hacia un nuevo pacto fiscal. Friedrich-Ebert-Stiftung. https://library.fes.de/pdf-files/bueros/mexiko/17774-20210525.pdf Cárdenas, S.,& Ruelas, I.(2018). Profesiones en riesgo. CREFAL. Chrisp, J.,& Martinelli, L.(2022). The Case for a Basic Income in the Emergent Digitalized Economy. In M. Busemeyer, A. Kemmerling, P. Marx, & K. van Kersbergen(Eds.), Digitalization and the Welfare State(pp. 174–191). Oxford University Press. Chua, K.(2018). The Future of Work: Trends to ride the wave of digital disruption. LinkedIn Article. Crouch, C.(2018). Redefining Labour Relations and Capital in the Digital Age. In Work in the Digital Age: Challenges of the Fourth Industrial Revolution(pp. 187–197). Rowman& Littlefield International. https://pure.mpg.de/pubman/faces/ViewItemOverviewPage.jsp?itemId=ite m_2601891_2 Davies, H., Goodley, S., Lawrence, F., Lewis, P., O’Carroll, L.,& Cutler, S.(2022, July 11). Uber broke laws, duped police and secretly lobbied governments, leak reveals. The Guardian. https://www.theguardian.com/news/2022/jul/10/uber-files-leak-reveals-gl obal-lobbying-campaign De Vynck, G.(2023, March 29). Elon Musk and a handful of AI leaders ask for‘pause’ on the tech. The Washington Post. Drahokoupil, J.,& Piasna, A.(2022). Internet and Platform Work in Europe. In M. R. Busemeyer, A. Kemmerling, K. Van Kersbergen,& P. Marx(Eds.), Digitalization and the Welfare State(p. 0). Oxford University Press. https://doi.org/10.1093/oso/9780192848369.003.0014 Eggleston, K., Lee, Y. S.,& Iizuka, T.(2021). Robots and Labor in the Service Sector: Evidence from Nursing Homes(Working Paper 28322). National Bureau of Economic Research. https://doi.org/10.3386/w28322 Frey, C. B.,& Osborne, M. A.(2017). The future of employment: How susceptible are jobs to computerisation? Technological Forecasting and Social Change, 114, 254–280. https://doi.org/10.1016/j.techfore.2016.08.019 Gallego, Aina, Kurer, Thomas,& Schöll, Nikolas.(fc.). Neither Left-Behind nor Superstar: Ordinary Winners of Digitalization at the Ballot Box. Journal of Politics. Gelepithis, M.(2022). The Politics of Tax Policy in the Digital Age. In M. R. Busemeyer, A. Kemmerling, K. Van Kersbergen,& P. Marx(Eds.), Digitalization and the Welfare State(p. 0). Oxford University Press. https://doi.org/10.1093/oso/9780192848369.003.0020 Giray Aksoy, C., Barrero, J. M., Bloom, N., Davis, S., Dolls, M.,& Zárate, P.(2022). Working from Home Around the World. Brookings Papers on Economic Activity. Guellec, D.,& Paunov, C.(2017). Digital Innovation and the Distribution of Income(w23987; p. w23987). National Bureau of Economic Research. https://doi.org/10.3386/w23987 Herrera, O. A., Keller, L.,& González, S.(2019). ¿Quiénes son los conductores que utilizan las plataformas de transporte en América Latina?: Perfil de los conductores de Uber en Brasil, Chile, Colombia y México. https://doi.org/10.18235/0001973 Holland, A. C.(2017). Forbearance as Redistribution: The Politics of Informal Welfare in Latin America. Cambridge University Press. https://doi.org/10.1017/9781316795613 Huws, U., Spencer, N., Syrdal, D. S.,& Holts, K.(2018). Working in the Gig Economy: Insights from Europe. In Work In The Digital Age(pp. 153–162). Rowman& Littlefield International. https://researchprofiles.herts.ac.uk/en/publications/working-in-the-gig-eco nomy-insights-from-europe INEGI.(2020). Survey on the Economic Impact Generated by COVID-19 on Enterprises(ECOVID-IE). National Institute of Statistics and Geography, INEGI. 17

INEGI.(2020a). Comunicado de prensa número 660/20. Resultados de la encuesta nacional de ocupación y empleo(nueva edición)(ENOEN). Cifras oportunas de noviembre 2020. Instituto Nacional de las Mujeres(2021) La brecha digital de género ¿Una expresión más de desigualdad? Año 7, abril, Boletín 4. http://cedoc.inmujeres.gob.mx/documentos_download/BA7N04%20VOB O_15072021.pdf Iversen, T.,& Rehm, P.(2022). Big Data and the Welfare State: How the Information Revolution Threatens Social Solidarity. Cambridge University Press. https://doi.org/10.1017/9781009151405 Juárez, J., Huesca, E.,& Arenas, P. C.(2022). Mi vecino es un robot. Debate. La brecha digital de género ¿Una expresión más de desigualdad?(2021). 2. Leyva, G.,& Mora, I.(2021). How High(Low) are the Possibilities of Teleworking in Mexico? Banco de México. https://doi.org/10.36095/banxico/di.2021.15 López-Cariboni, S.(2018). Informal Service Access in Pro-Cyclical Welfare States: A Comparison of Electricity Theft in Slums and Regular Residential Areas of Montevideo. Journal of Comparative Policy Analysis: Research and Practice, 21, 1–19. https://doi.org/10.1080/13876988.2018.1462604 McKinsey Global Institute.(2017). A future that works: Automation, employment, and productivity. Nedelkoska, L.,& Quintini, G.(2018). Automation, skills use and training. OECD. https://doi.org/10.1787/2e2f4eea-en Popiel, P.(2022). Regulating datafication and platformization: Policy silos and tradeoffs in international platform inquiries. Policy& Internet, 14(1), 28–46. https://doi.org/10.1002/poi3.283 Radetich, N.(2022). Cappitalismo. La uberización del trabajo. Siglo XXI Editores. Rani, U.,& Grimshaw, D.(2019). Introduction: What does the future promise for work, employment and society? International Labour Review, 158(4). Rivas Valdivia, J. C.,& Gaudin, Y.(2021). Diagnóstico de las brechas estructurales en México: Una aproximación sistémica general. https://hdl.handle.net/11362/47671 Rosales, Ma. S.(forthcoming). Economías populares y pandemia en la Ciudad de México. Apuntes Sobre El Futuro Del Trabajo, El Colegio de México. Schatan, C.(2018). Transformación productiva, empleos y retos para la formación de capital humano en México(p. 28). Friedrich-Ebert Stiftung. https://mexico.fes.de/fileadmin/user_upload/Publicaciones/Claudia_Schatan _Tranformacion_productiva.pdf Schwander, H.,& Vlandas, T.(2020). The Left and universal basic income: The role of ideology in individual support. Journal of International and Comparative Social Policy, 36(3). https://ora.ox.ac.uk/objects/uuid:85be792f-2dab-4d4d-844f-c1b8b12b2b3f Silva Taylor, M., Rivera González, R.,& Jaramillo-Molina, M.(2022). Reparto justo: Condiciones laborales de personas repartidoras de apps durante la pandemia(p. 54). Instituto de Estudios sobre Desigualdad, A.C. https://www.indesig.org/wp-content/uploads/2022/11/Reparto_Justo.pdf Srnicek, N.(2018). Capitalismo de plataformas. Caja Negra. 18 References

IMPRESSUM ABOUT THE AUTHORS IMPRESSUM Achim Kemmerling is a research professor at the Willy Brandt School of Public Policy at the University of Erfurt. Melina Altamirano is an associate professor at the Center for International Studies at El Colegio de México. Ma. Susana Rosales is a postdoctoral fellow of the Work and Inequalities Seminar at El Colegio de México. Corresponding author: Achim Kemmerling, achim.kemmerling@uni-erfurt.de. This working paper is a collaboration between the Willy Brandt School of Public Policy at Erfurt University, the Center for International Studies at El Colegio de México (Colmex), and the Friedrich Ebert Foundation in Mexico (FES). It is also the first product of the project PolDigWork—Politics and the Future of Work in Middle-Income Countries— funded by the Deutsche Forschungsgemeinschaft. The Friedrich Ebert Foundation in Mexico Yautepec 55| Col. Condesa 06140| Ciudad de México| México Responsible Yesko Quiroga Project coordinator Carlos Cabrera Telephone+52(55) 5553 5302 Fax+52(55) 5254 1554 https://mexico.fes.de Commercial reproduction of the material contained in this publication may be made only with prior written permission by an authorized representative of the FriedrichEbert-Stiftung(FES). The opinions expressed in this publication are those of the author. They do not purport to reflect the opinions or views of the Friedrich-EbertStiftung(FES) or its members. 19 ISBN 978-607-8887-09-5

THE FUTURE OF WORK IN MEXICO: Exploring Key Actors’ Perceptions on the Socio-Economic and Policy Implications of Workplace Transformations In this paper, we report and discuss the results of an expert focus group bringing together actors from several sectors of Mexican society: government, industry, trade unions, the technological community, and academia. The objective of this workshop was to foster a dialogue among sectors, collect a diverse set of voices, and stimulate the discussion around two main questions: 1) what is the socio-economic impact of new technologies for your sector? And 2) what are the policy and political implications of the future of work for your sector? Our objectives of the workshop were multiple, but our focus was on identifying common topics or even common visions among participants, as well as points of divergence and controversy. We were particularly interested in four main transversal topics: First, the multiple dimensions of inequality created, enhanced, or maybe also mitigated by technological change. The second transversal topic is the divide between the formal and informal sectors. While this divide is notoriously hard to define, with many grey zones in between, it is also clear that informality matters. In turn, the formal-informal divide relates to the third transversal topic, the role of the state in managing the technological revolution. Against the background of multiple divides and inequalities, government actions could help or worsen technology’s impact on society. State capacity is crucial in assessing changes, regulating new technology, compensating losers, and incentivizing positive developments. Our final transversal topic was the role of narratives and perceptions of technology and its links to the social world. Achim Kemmerling, Melina Altamirano and Ma. Susana Rosales November 2023 ISBN 978-607-8887-09-5