The Determinants of Youth Activity Status in Mexico City. A Mixed - - PDF document

1 The Determinants of Youth Activity Status in Mexico City. A Mixed Methods Analysis Gabriela Sánchez-Soto Andrea Bautista León Department of Demography The University of Texas at San Antonio Contact: gabriela.sanchez-soto@utsa.edu Paper submitted for the 2017 International Population Conference, Cape Town, South Africa DRAFT: Please do not cite without authors’ permission Abstract Recently, the literature on youth in Mexico has focused the lack of educational and occupational

• pportunities as they transition from adolescence into adulthood. A concern is the increasing

proportion of youth who are not in school or work. This group is known as “NINIs” (an abbreviation for “not studying, not working”). Despite increased public interest, little research has focused closely on this group. In this paper we describe the prevalence and determinants of being a NINI in the Mexico City Metropolitan Area using data on youth ages 18 to 25 from the National Survey of Occupation and Employment, and data from 40 in-depth interviews collected in 2015-2016. Findings show that NINIs are more likely to be children or spouses of the householder, to come from households where the head is more educated, and to be more educated males or less educated females. Among the college educated there is a bifurcated path, where some are more likely to be NINIs while others to be full-time students. Overall, youth in more disadvantaged households are more likely to work full-time. Our qualitative evidence provides some nuance to understanding the barriers to school and work, and supports our quantitative analysis. NINIs are inactive when work opportunities are scarce or while they try to gain admission to college. Barriers to employment relate to discrimination of less experienced workers and the undesirability of low-wage employment. Barriers to school relate to an inability

• f universities to absorb most applicants and to a mismatch between school and vocation.

Keywords: Children and Youth; education; inclusion and exclusion; Qualitative methodology; Quantitative methodology;

2 Introduction In recent years, the literature on the outcomes of youth in Mexico has focused on their activity status and the lack of educational and occupational opportunities as young people transition from adolescence into adulthood (Giorguli Saucedo 2009; International Labor Organization 2010; Pérez-Baleón 2012). A specific concern is the increasing proportion of young people who are not engaged in work or education. This group of young people is widely known among researchers, journalists and the general public in Mexico as “NINIs”, an abbreviation derived from the phrase “ni estudia, ni trabaja” (which roughly translates into “not studying, not working”). Both popular culture and the media have shown great interest in NINIs, particularly focusing on their perceived unrealized potential and the consequences of their status for the future labor force in the country (Murayama 2010; Rosique Cañas 2013). Exact figures about NINIs in Mexico are hard to come by, but some estimates put them at around 7.8 million, or 22% of the population ages 12 to 29 (Negrete Prieto and Leyva Parra 2013; Rosique Cañas 2013). Existing estimations vary due to differences in the definition of work and education activities, and on whether they include informal and sporadic work, unpaid labor, or care and domestic activities (Leyva Parra and Negrete Prieto 2014). Besides the challenges in measurement and definition, little is still known about the determinants of becoming a NINI, and even less attention has been paid to their motivations, family life experiences, aspirations, and expectations for the future. Some explanations suggest that becoming a NINI is related to socioeconomic status and to the characteristics of the household of

• rigin (Arceo-Gómez and Campos-Vázquez 2012; Murayama 2010; Rosique Cañas 2013).

Others also consider this trend is a consequence of decreasing labor opportunities despite increased enrollment in high school and college across Mexico (Arceo-Gómez and Campos-

3 Vázquez 2012; Rosique Cañas 2013). Existing research suggests that the proportion of NINIs has been overestimated due to the assumption that NINIs are inactive or “do nothing” when many of them are actually engaged in unpaid housework or providing family care (Leyva Parra and Negrete Prieto 2014). These kinds of family obligations often go unrecognized and may be a barrier to schooling or to more traditional employment. In addition, it is likely that sporadic, informal and unpaid work has not been properly reflected in employment data available from quantitative surveys, so that participation of NINIs in productive activities may be underestimated. Given all this, our research has two specific aims. The first is to understand the prevalence of NINI status relative to other activities over the observation period and the characteristics of youth who fall into this category. Second, we explore the determinants of being a NINI versus engaging in work and schooling at the individual, household, and community

• levels. We use quantitative data from the Encuesta Nacional de Occupación y Empleo (ENOE,

National Survey of Occupation and Employment) and data from in-depth interviews with NINIs we collected in Mexico City during 2015-2016. In combination, these data allow us to understand trends in the rates and determinants of youth activity, and the perceived barriers to

• ccupation and employment among NINIs.

Background and Previous Research According recent estimates from the Organization for Economic Cooperation and Development (OECD) Mexico occupies the fifth place among its member states in the proportion of NINIs. Close to 8 million Mexican youth are not working or attending school, which represents 22% of 15 to 29 year olds in the country, although only some of these declare themselves to be

4 unemployed, 19% of youth are not actively looking for a job and are detached from the labor market altogether (OECD 2016). The presence of NINIs in popular culture goes beyond academics and newscasters, the acronym has made it into colloquial language as a short-hand for youth’s inability to fulfill their potential, which is often perceived to be due to their lack of motivation and their disregard for hard-work. A quick review of national and regional news media shows an increase in usage and awareness during the last decade. The rhetoric goes from alarmist news that speak of NINIs as a “lost generation,” of how NINIs are stopping the progress of the country, and how their laziness stifles their potential –and the demographic dividend for Mexico. The public discourse centers on the waste of human capital that inactivity status represents (Murayama 2010; Zepeda 2013). Other more measured media speaks of NINIs as victims of their circumstances and of the structural disadvantages in the educational system and the labor market, and as a result of poor policy-making on the government’s part (Miranda 2015; Roldan 2016; Rosique Cañas 2013). More recent narratives are likely to see the issue as one derived from the increasing lack of educational and occupational opportunity in the country (Poy Solano 2010). Existing research on Mexican NINIs focuses on two main areas. The first focuses on discussions of definition and measurement, as well as prevalence in nationally representative

• data. For instance, Leyva and Negrete (2014), argue that the proportion of NINIs has been
• verestimated. The authors question the widespread assumption that NINIs are inactive or

“doing nothing,” and they suggest many of the young people in this category are not attending school or working due to family and care commitments (Leyva Parra and Negrete Prieto 2014). In addition, it is likely that sporadic and unpaid work has not been properly reflected in employment data available from cross-sectional quantitative surveys, so that participation of

5 NINIs in productive activities may be underestimated. Another possibility is that activity status varies throughout the year as employment and educational opportunities fluctuate. The second area of research has focused on the causes and consequences of youth inactivity, and how the structure of educational and employment opportunities have caused an increase in the number of NINIs and serious limitations to their ability to get back to school or work (Rosique Cañas 2013). This type of research mostly relies on national data and there is not as much work focusing on local labor markets and educational contexts, or on how the context of local economic opportunity affects youth’s likelihood to be in school or work. This work also has not yet explored the barriers perceived by youth to enter school or work, and how household context influences their activity status. Research Questions and Expectations Given the background provided by previous research and the gaps that remain to be filled, the main research questions we aim to address in this paper are: How prevalent is NINI status relative to other activities among Mexican youth? What are the determinants youth inactivity at the household and community level? How do youth perceive their status as NINIs? And, what are the barriers to education and employment they perceive? Due to our understanding of the determinants of activity status, we would expect fluctuations in the prevalence of activities over the observation period. But overall, being a NINI would be more prevalent among women, less educated youth, and children of the household head living in two-parent households. Engagement in full-time work will be more likely among men and among the most disadvantaged, while school attendance will be probably associated with higher economic resources in the household. Given our understanding of the public discourse

6 around NINIs, we expect that youth who identify as NINIs will have a clear understanding of the term and its connotations, in addition, we expect they will be well aware of the impediments they face to enter the educational system or to obtain employment. In these narratives about their limitations we expect them to demonstrate the important role of their socioeconomic status and the context of work opportunities in the local context. Specifically, we focus on the Mexico City Metropolitan Area as it encompasses a varied context of work opportunity and occupational diversity, making it an excellent case study to explore these patterns of instability and precarious employment that characterize large urban areas across the developing world. Mexicans growing up in Mexico City have access to a variety

• f educational and employment opportunities, although access is highly unequal.

Further, NINIs’ precarious relationship with the labor market can extend well beyond the early twenties and have important consequences for their long-term career development and income security, understanding the determinants of this status is essential to design timely interventions to overcome these disadvantage. Young NINIs may have a harder time obtaining stable employment because previous work experience is highly valued in the labor market, feeding into a vicious cycle where they cannot get a job because they do not have job experience, and cannot get job experience because they do not have a job. Young Mexicans also are at a higher risk of ending up in dead-end, low quality, precarious jobs, where they do not have access to benefits, health insurance, stable schedules, or long-term contracts, because their relative inexperience and youth leaves them with slim pickings in the local labor market. Data and Methods

• Data. We use a mixed methods approach to explore the activity status and the determinants of

7 inactivity among youth in Mexico City. First, we use data from five quarterly waves of the ENOE from April 2015 to June 2016 to explore the activity patterns of youth from the Mexico City Metropolitan Area over the observation period (N=4,512 men and 4,453 women) (Instituto Nacional de Estadística Geografía e Informática 2016). The ENOE is a quarterly survey with a rotating panel that collects information on employment dynamics in the country, each selected household remains in the survey for five consecutive panels, this gives us the opportunity to compare the prevalence of different activity statuses throughout the time covered in the panels and provides an idea of the stability of the activities youth in the sample participate in. Another advantage of the ENOE data is that we can account for household structure and family socioeconomic characteristics besides individual-level indicators. We selected people ages 18 to 25 and who have completed middle school. In addition to this information, we matched the individual records to labor force composition data at the delegation or municipality level.1 Second, we use data from in-depth interviews with young NINIs in the Mexico City Metropolitan Area that we collected in 2015 and 2016. We collected 40 interviews with young people ages 18 to 25 who have finished middle school, who do not attend school, do not work in paid employment, and have been doing so for at least three months before the interview.2 We recruited interviewees using a variety of methods; we designed recruitment flyers that we sent to

• rganizations –public libraries, art collectives, etc.– that cater to young people in the City, we

distributed the flyer to colleagues and contacts in the city, and we created a public Facebook group to cast a wider net of local respondents. We also made new contacts through respondents

1 Mexico City is divided into administrative units called delegaciones, as the metropolitan area extends outside of

the City’s territory, parts of the surrounding Metropolitan area include municipalities from the neighboring states.

2 This qualitative data collection was conducted thanks to funding from the Department of Demography and the

College of Public Policy at The University of Texas at San Antonio. The data collection received human subjects research approval from the Institutional Review Board from The University of Texas at San Antonio.

8

• themselves. Our use of this convenience sample and a snowball approach allowed us to access a

diverse group of young people from different areas and socioeconomic strata in the city. Our sample includes 23 women and 17 men. These interviews centered on the respondents’ experiences as NINIs and on their perceived barriers to education and employment. Our interview also includes questions on demographic and socioeconomic characteristics of the interviewee and their families, along with an account of their experiences with the educational system and the labor market, and their time

• use. In addition, we included questions to gauge their sources of support among family and

social networks. Lastly, we collected information on expectations and aspirations for the future and careers. In this paper we focus on the parts of the interview that collected youth’s narratives about being a NINI, and on the barriers they perceive to attending school and engaging in employment. Quantitative Analysis Strategy. Using the ENOE data, we describe the proportion of people who can be identified as a NINI across the panel and explore the prevalence of different activities throughout the observation period to understand the stability in the proportions of the sample in each activity status over the period of observation. We also describe the characteristics

• f youth who are NINIs and those who belong to other activity categories. We estimate a

multinomial logistic regression model to predict the risk and determinants of being a NINI, a full-time student, or combining school and work, relative to being a full-time worker. Standard errors are adjusted for clustering at the individual level to account for a small proportion of the sample who participated in more than one wave (The survey re-interviews about 20% of each wave’s sample from one quarter to the next for 5 quarters, after which the respondent is not interviewed again).

9 Independent variables: Our analysis controls for the age of the individual; educational attainment; a combined measure that indicates the sex of the household head and whether the household is single-headed; the educational attainment of the household head; relationship of the individual to the household head; whether the respondent is in a marital or cohabiting union;3 household income per capita, and a dependency ratio for the household. These variables account for relevant individual- and household-level characteristics. We use economic characteristics at the delegation or municipality level, which include the proportion of the labor force working in the secondary sector, in commerce, and in the services sector to represent the conditions of the local labor markets in the area of residence of our sample. We also account for the period of the ENOE each observation belongs to. Qualitative Analysis Strategy. We use information from our qualitative interviews to understand how well the quantitative definitions of activity status reflect the experiences of young NINIs. Our qualitative analysis aims to identify and represent the main narratives associated to the activity status of youth and how socioeconomic status and family background are associated to experiences with school and work. We use NVIVO 10 to explore and analyze this data. These interviews will complement and help explain the patterns found from the ENOE

• data. The analysis presented in this paper mostly relies on the questions on perceived barriers to

education and employment, and on questions about how young people see their situation as

• NINIs. However, we use the insights we acquired through the entire interview to inform model

specification, interpretation of findings, and conclusions. We expect that our unique insights will

3 We only have fertility information for the female respondents in the sample. We tested the indicator in preliminary

models and, as expected, we found that having children is associated to a decreased risk of being a full-time student

• r a student/worker, but it had no significant effect on the risk of being a NINI among the women in our sample. We

decided to use union status as our family formation indicator because it is available for both sexes and its effects are similar.

10 also contribute to improved definitions of youth activity status, a more nuanced understanding of its determinants, and the development of appropriate policy responses. Characteristics of the ENOE Sample To illustrate the distribution of activity status and the transient nature of inactivity status

• vertime, Figures 1 and 2 show the prevalence of the four activity status categories over the

period of observation for men and women respectively. Most of the men in our sample are working full-time, full-time student is the second most prevalent status. Being a NINI is not as prevalent among men, though the prevalence increases in the summer months and at the beginning of the year. Likely, these changes may represent transition moments between work or school or school desertion after the partial completion of a grade. Among women, Figure 2 shows that working full-time is not as prevalent as it was for men, and being a NINI or a full-time student alternate as the second most prevalent activity

• status. There is variation in status prevalence across periods, similar to the pattern seen for males.

A notable finding is that for a couple of the periods, being a NINI is a more prevalent status than being a full-time student among women. Overall, the data shows that NINI status is not stable throughout the period of observation, which is consistent with the trends observed in our qualitative interviews, where we found that activity status is not fixed across time and that youth enter and exit different activities frequently due to instability in the labor market and the precariousness of the jobs available to them. We also found that being a NINI is more prevalent among women overall, and that women are also less likely to work or be students full-time relative to men. Table 1 presents the distribution of our variables of interest by activity status and sex.

11 The left side panel presents information for men and the right for women. Among men, NINIs and full-time students are younger on average that full-time workers and those combining school with work. Among women, full-time students are the youngest, followed by NINIs, while full time workers are the oldest. Male NINIs are highly educated, in contrast to female NINIs who are more likely to only have a middle or high school level of education. Most full-time workers have a middle or high school education among both sexes. For both men and women, those studying full-time and those combining school and work have higher levels of schooling. In terms of union status, there are important differences across activity status and sex, among NINIs only 4.5% of men are in a union, in contrast to 53% of women. Among men, full- time workers have the highest percentage in a union (29%). Overall, students, both full-time and in combination, are the least likely to be in a union, which is consistent with the role incompatibility between being a spouse and being a student. Most of our respondents come from a household with a male head and where both parents are present, the lowest proportion is among full-time working females and the highest proportion is among NINIs and full-time students. About one quarter of respondents live in a household headed by a single mother, among these, the proportion of NINIs is lower, and the proportion of workers is higher, especially for females. there is some variation in the education

• f the heads of household, full-time workers live in households where the heads have lower

education, while higher education among heads is found among students, both full- and part- time. Most male NINIs in our ENOE sample are the children of the household head, while female NINIs are, in descending order, daughters, spouses or daughters-in-law of the household

• head. Overall, most respondents selected in this sample are the children of the household head,

12 independently of their activity status. The household dependency ratio goes from 0.2 to 0.5, the dependency ratio is greater in the households of youth who work full time and among female

• NINIs. The median household income per capita is higher among youth who combine school and

work, followed by youth working full-time, which makes sense as in these cases, the respondent’s contributions to the household economy increase their total income, and for those who combine school and work, this may be a strategy to continue financing their education. Relative to youth in the other activity categories, NINIs live in households with lower levels of household income. As already shown in Figures 1 and 2, the proportion of NINIs is greater in the summer months and at the beginning of the year, though this temporal differences are less pronounced for

• women. At the community level, the average proportion of workers in the secondary sector is

18%, in commerce is 22%, and in services is 56%. Multivariate Analysis Results Table 2 presents the relative risk ratios from the multinomial logistic regression model we estimated to predict the determinants of being in each of the status categories. We find that age is negatively related to all our outcomes relative to being a full-time worker. For men, being a NINI is associated with higher levels of education, college educated youth are almost three times more likely to be NINIs relative to those with some high school, while those with middle school education or less have a decreased risk of being NINIs. For women, the effects of education are different, we only observe a decreased risk of being a NINI among women with a high school education. Similarly, relative to working, full- and part-time school attendance are positively associated with a college education and negatively related to

13 lower education. It seems that among our sample, higher education leads to a bifurcated path where some become NINIs and others continue to pursue their education. More research needs to be done to better understand who takes the path of becoming a NINI and whether a higher risk of inactivity among the more educated is related to the transition into first job post-graduation. Men who are in a union have a decreased risk of being in all activity categories, relative to work. This is consistent with the expectation that family formation and full-time work are more compatible roles among men. Among women, being in a union is associated with an increased risk of being a NINI, and to a decreased risk of being a student or combining school and work. This is consistent with the increased incompatibility between the role of wife and the role of student, and shows that even though female labor force participation has increased in Mexico, conflict between family responsibilities and work outside the home remains significant. We find that household headship characteristics have effects that are similar for men and

• women. Among males, in a house headed by a woman (whether or not there is a second parent),

is related to a lower risk of being a NINI and a lower risk of being a student compared to two- parent, male headed households. No significant effect is found for households headed by single fathers, except for lower risks of being a full-time student. For women, we find that relative to those living in two-parent, male headed households, women living in other types of households have a decreased risk of being enrolled in school full time, and that those living with a single mother are less likely to be NINIs relative to working full-time. Further, higher education of the household head is related to an increased risk of males being NINIs, for women, there is an increased risk of being a NINI only when the household head has a high school education. For both sexes, living in a household where the head has more than a high school education is associated with a higher relative risk of studying full-time vs.

14 working full-time. These findings may seem counterintuitive, but it is possible that in these cases, the higher SES of the household is related to flexibility in activity status for some, while for others is related to a longer time spent as a full-time student. Ultimately, it is likely that being a NINI is something that the children of families with some resources can afford, and that the poorest youth likely cannot afford to be NINIs and have to take whatever job they can possibly

• get. Further exploration of the barriers for school enrollment among NINIs would provide more

nuance to our understanding of what keeps some youth in school and what excludes others. The relationship to the householder is significantly related to activity status. Sons of the household head have a higher risk of being NINIs (RRR=2.28), other relatives are also at an increased risk of inactivity (RRR=2.33), though the largest risk is among those unrelated to the householder (RRR=3.04). These same groups are also more likely to be students. This echoes the trend observed for the effect of household head education, it appears that in terms of household characteristics there is a bifurcation between youth who are NINIs and youth who are students. A different picture emerges for women, the only relationship categories that associated to a higher risk of being a NINI are daughters-in-law (RRR=2.09) and daughters of the householder (RRR=1.85). Daughters and other relatives of the household head are more likely to be full-time students relative to working, while spouses have a decreased risk attending school. A higher dependency ratio in the household is associated with lower risk of being a NINI for men. For women higher dependency ratios in the household are associated with a decreased risk of full- time school attendance, and school and work combination, and increased risk of being a NINI. Higher household income is associated with decreased risk of being either a NINI or a full-time student for both sexes. Lastly, regarding the characteristics of the place of residence, the composition of the

15 labor market is not related to the risk of being a NINI, but higher proportions of workers in the secondary and tertiary sectors are related to a decreased risk of being a full time student for men, while only the proportion in services is negatively related to women’s risk of being full-time

• students. As shown in the descriptives, men are more likely to be NINIs during the summer

months, and women are less likely during the fall and spring quarters. Qualitative Analysis Results To better understand the experiences of youth identifying as NINIs, we analyzed narratives related to the perceived barriers to work and school among our interviewees. Our in-depth interview respondents are 22 years old on average, 57% of them are women, almost all live with at least one of their parents, are single and most have no children (two of our female respondents were mothers and one of our male respondents was a father). About 38% of our respondents have less than a high school education, 22% have a high school diploma, 28% some college, and 9% completed college. Many of our respondents left school while attempting to complete high school or in the early semesters of college. Among the high school educated we had many respondents who did not “leave school” per se, but rather were in the process of seeking admission to university, others left college early when they discovered their chosen path was not what they expected.5 To better understand how family background affects being a NINI, we collected detailed

5 In Mexico, students have to choose their major before applying to college. Students apply for admission directly to

the department of their choice, and though some programs have a common core (i.e. engineering) and then split into more specific specializations from there (i.e. civil engineering, electrical engineering, etc.), for a student who wishes to change programs, the process may require them to drop out and re-apply to another program. There is also a high degree of vocational tracking, starting in high school, students choose a track related to the area they are interested in pursuing in college (i.e. social sciences, life sciences, humanities, business and administration, physics and math, and so on). Thus, for a student who wants to radically change majors, this process may even require taking additional high school credits before being able to apply to a new program in college.

16 information on socioeconomic status. We ascertained our respondents’ socioeconomic status in three ways. First, we asked about assets (computers, laptops, cell phones, etc.) and services (cable TV, internet, landline phone, etc.) available in their household. Then we asked how they would classify their families’ socioeconomic status relative to others in their community. Lastly, we asked what was the approximate monthly income of their household. We find that our respondents come from varied socioeconomic statuses as evidenced by their response to the first and third questions. But when asked to self-identify their social class, most of them said their families were “middle class.” If we consider the education of their parents, about half of the parents have at least some college education, whereas fewer have a high school education or less. The occupations of the parents ran the gamut from agricultural worker to managerial positions, but the majority concentrated in jobs in the secondary and tertiary economic sectors, with a few in white-collar occupations. Most of our respondents lived with their parents in nuclear

• households. Only a few of our respondents lived with other relatives, those who live with other

relatives are likely to have moved from somewhere else in the country with the purpose of studying or working in Mexico City. The Meanings of Being a NINI Across our conversations with our respondents, we found that the label “NINI” is well understood by them. To avoid biasing their responses on the meanings of the term when recruiting and interviewing them, we made an effort to not use the term at all, especially because we wanted to know whether this label came up organically in our conversations and to assess our respondents’ awareness of its connotations in public discourse. When we asked them about their educational and occupational trajectories, and their perceived barriers to success in these areas,

17 some of them eventually used the term to refer to their situation or to explain how others think of them due to their current circumstances. For those who did not bring it up on their own, we asked directly about the respondents’

• pinions on the situation of youth who “do not study and do not work.” Our respondents are well

aware that the use of NINI by the public and the media is pejorative and that it is associated with decreased social status. Tania,6 a female respondent, is 24 and she dropped out of high school to train as a tattoo artist, she thinks about this issue a lot:

The issue of calling a guy “NINI” is derogatory, no? In the first place, it’s is very derogatory, because you do not know really the kind of life that guy has. No, really, there are guys who do thousands of things to make money. So they really aren’t NINIs. […] These people, they are looking for a way, they are looking for their path, so calling them that is like you are stopping them. No? Some of them sell their drawings, paintings, so, they are making some money. But people look down on this because they don’t work for a business, because they don’t work for someone important, or… whatever. Then I see, how should I call it? Depression, sometimes, in these guys. […] They say: “society is not appreciating what I do, they don’t see I am really trying to do what I like to do, and live of what I like to make.”

Despite identifying as NINIs, many of our respondents expressed frustration with the public’s misunderstanding of their circumstances and motivations. These respondents believe the public thinks being a NINI is a result of personal failure, while ignoring institutional constraints and lack of opportunities that youth in Mexico City face today.7 Laura (female, 21 years old) and

6 To protect the anonymity of our respondents, all the names we use in this text are pseudonyms. 7 In the course of our research we also got to experience this negative discourse first hand, often when talking to

• ther people about our project in the hopes of obtaining respondents, we were told that our project was a good thing,

maybe we could finally “find out what is wrong with these children,” or that maybe we “can find a way to motivate them to be productive members of society.” This point of view was not shared by all adults we encountered, some did recognize that economic opportunity in the country is limited, but we understood the differing opinions as part of the dominant public conversation we have observed around youth in these circumstances.

18 Ramiro (male, 25 years old) remembered the advertisements for a program they became involved with in their delegacion to use art to promote activity among inactive youth, Laura says:

Even the name… it was “Youth in State of Alert” no? And in small lettering it said it was for NINIs, but it gave this idea that we were all thugs, no? drug addicts […] [Laura laughs] many of us do not look like that, so they looked at us funny, […] for them it would have been perfect if all of us showed up on drugs. […] Yeah, and it is only that you haven’t had the opportunity to go study or work.

They were aware of the prevalence of inactivity in their social environment and mentioned that many of their own friends, classmates or relatives were (or had been) NINIs. Daniel (male, age 23) thinks the popularity of the term comes out from the increase of people without

• pportunities, of those who want to study but do not find easy access to school. Sara (female, age

22) reflected on the situation, she sees how people speak badly of NINIs all around, but she thinks it is because they “don’t know.” We ask her, what would she tell this people? And what sort interventions would she like to see?

[They should know] that those who are NINIs, is not precisely because they are lazy, or because they are taking a year off, or that they don’t want to study anymore, rather it is that Mexico is a country that is going down in terms of opportunity, in terms of resources, in terms of work […] In public schools, teachers could be less strict and less mean; in private schools it would be, I don’t know, lowering tuition for high school and college, because they rise it up to the sky […] or that they had more scholarships, more opportunity. […] At work they should give them at least one-month trial, give them the opportunity to demonstrate they can do the work and not just leave them out.

Sara’s thoughts reflect what we heard from many of our other respondents, they do not want scolding, they want opportunity. They have plans and desires to improve themselves. An

19 important finding from our qualitative study is that for many of our respondents, their time use reflects strong educational or occupational ambitions. We find that many of our interviewees are NINIs not because of a disdain for school or work as media portrayals and public opinion would have it. Rather, many are preparing for college entrance exams, learning a skill online, or doing

• dd jobs to further their skills. This is a stark contrast to the popular perception that NINIs lack

the ambition and motivation to further their opportunities. Some of our respondents expressed very clearly that they left the traditional educational system because they found it lacked a clear connection to the particular career they are interested in pursuing, others are simply interested in a less conventional path to an occupation, while many others are simply stuck in a transition period between high school and university, or between the end of university and their first job. The educational and occupational histories we collected show great complexity in the barriers to schooling and working among this population. We would like to focus more specifically on some of the most commonly discussed barriers to school and work that our respondents articulated in our conversations. Barriers to Education and Work Many of our respondents encountered bureaucratic hurdles and institutional limitations to continuing their education. Two of the most discussed limitations are: 1) the mismatch between their occupational aspirations and the education available to them and the difficulty to transfer credits across educational systems (i.e. from a Mexico State high school to a National University high school); and 2) the discrepancy in the number of students applying for higher education and the number of places available at universities. The inflexibility of the educational system left many of our respondents dissatisfied with

20 their options. Some did not like the schools they entered first and wanted a change, but knew that changing would imply losing some of the progress already made. Many others simply became discouraged by what they saw as a rigged system; while others were dissatisfied with the quality

• f the education they received. A few of our respondents cited the lack of vocational guidance

and mentoring in high school as a reason to become disenchanted with higher education. For many others it was hard to find a connection between the skills obtained in school and the requirements of the labor market. Felipe, a 20 year-old who dropped out of high school to work in a department store felt work gave him more practical skills that he was not attaining in school:

I worked in customer service, sales, stocking merchandise, inventory, all the things they normally do in department stores. […] I realized that I really enjoy working, I am more of a worker than a student. […] I felt more comfortable. It’s just that I don’t like school, like it bores me a lot, I don’t like it being so mechanic, so square.

Even though his family would like him to get a college degree, he is interested in becoming a volleyball coach, he hopes to finish high school and apply to a sports training school. His school trajectory has been interrupted by a lack of motivation and a preference for work, though work has not come steadily for him. Some of our respondents left school after not being admitted in the high school or college

• f their choice, but many others did after not being admitted to one at all. Not being admitted

into public high schools and universities means that their only educational opportunities are either the more expensive but good quality private schools –which they most likely cannot afford– or the cheaper but poor quality private for-profit schools that prey on the students rejected by the public educational system. There is definitely a discrepancy in the number of

21 applicants and the number of places available in public higher education institutions. To illustrate the lack of access to higher education, for the academic year 2015-2016 the National Autonomous University of Mexico (UNAM), the largest and most prestigious public university in the country, only admitted 9% of their applicant pool, of 128,000 applicants, only 11,470 were accepted (Olivares Alonso 2015). Many of our respondents were NINIs while preparing to retake admissions exams, and their parents agreed that their temporary inactivity would be worth it if they could eventually get into their desired university. Antonio (male), is 20 years old and hopes to go to UNAM to study medicine, but so far, he has taken the admission exams twice without making it through. He and his family decided he should not work while he continues preparing to take his exams one more time. He tells us about his and his peers’ experience trying to get into college. For those who did not go to a UNAM high school and gained direct admission, it can be very difficult to get in. He thinks that in medical school “there should be more places, last year among thousands who took the exam,

• nly 11 got in.” Among his friends, half of those who were trying to get into college already

gave up. Irene (female, 19 years old), wants to be a nurse, she has taken 3 admissions exams, 2 at UNAM and one at another local university, without success. She has tried to find work but her lack of experience has been a barrier. Even though she has a high school degree, she has not succeeded in finding a job; because she has never worked, she cannot demonstrate the 6 months to 1-year experience most jobs ask for. Her family is also not doing very well financially; her father makes $1,500 pesos a week (~80 USD) which is simply not enough to support a family of

• 5. Irene does not think the family has enough to pay for her school.

They’ve told me that they will support me [to cover the costs of school] but I am looking for work right

22

now because […] For what I see, the school requires uniforms, I would need daily transportation fare… but also in school the advantage is that I can apply for a scholarship, that wouldn’t be so difficult.

Besides Irene, many others mentioned economic problems as barriers to further educational attainment, many parents could not help pay for school, even if they supported their children’s desire for higher education. Others talk about having responsibilities in their homes, encompassing both housework and caregiving. Mara (female, age 22) was working on a bachelor’s degree in psychology but left school to take care of her baby. She says that, on one side her family pressures her to get a job and not be “lazy,” but on the other she receives criticism if she leaves the baby for too long when she goes out to look for work. She feels sad and confused at hearing these mixed messages. Unfortunately, she now sees her college life further and further away from her, and between her child and her family’s expectations, the chances that she will ever go back are slim. A few more of our respondents talked about having to care for younger siblings, elder relatives, and doing housework because both their parents had to work to make ends meet. These respondents were less likely to be actively looking for a full- time job, and more likely to engage in sporadic work. But even those with responsibilities at home remarked on the scarcity of well-paid formal employment, their job search had shown them that most of the opportunities for young people are in precarious, unstable, and low-wage jobs. Ernesto (male, age 25), had to work because his family could no longer support his education, he dropped out of high school to work full time at a cell phone company, but despite working long hours, his salary was too low, so he quit. Sadly, he still has not found a new job. Now he feels stuck between looking a new job and putting together enough money to go back to school. Many of them cite blatant discrimination in the labor market due to their age, experience,

23

• r even their looks, style, and identity. Ageism is present in job ads that require applicants to be

within a specific age range, but also expect them to be educated and have several years of

• experience. This was the most common example described by our respondents, who were

particularly exasperated to see how their credentials do not always translate into better jobs. This is articulated by Antonio, the 20 year-old who has been trying for 2 years to gain admission to the National University, when talking about employment opportunities for him and his peers. He argues that getting a job is difficult even for high school graduates as employers may want higher education, but also something most of them do not have: work experience.

They ask for a lot! […] They want experience… professional experience […] It is a bit illogical, right? Because if you were dedicated to your studies, how are you going to have work experience? […] they [young people] make an effort, or their effort is not enough, even if you thought it would be! Even if you came out [of school] with a 10 grade average! [10 is the highest grade].

Another of our respondents, Carlos (male), who is 24 and has a B.Sc. in Biotechnology, thinks there are opportunities for those with a college education, but they often offer low wages and also require the work experience youth do not have:

I think there are opportunities, but the problem is that most of them require experience, experience that most of the time we do not have, and unfortunately, the few offers available offer very low wages. […] The

• ther time I was looking for job opportunities on OCC [job search website], there was an ad for a job at a

higher education institution that would be compatible with my profession, they offered to pay $3,000 pesos a month [~160 USD] and that’s too little. [I: With a bachelor’s?] With a bachelor’s, a master’s… and if it is a master’s they will say you are overqualified. […] Often there are positions available as a research fellow at the university but those pay even less. […] though for many of us these are useful just to gain experience, […] I want to start gaining experience more than the money itself.

24 Of course, Carlos says, you can also walk away from your profession and get a job in a call center, but this job will be even more exploitative. Many of our respondents spoke badly of these kinds of customer service, marketing, or sales jobs that commonly fall under the call center

• umbrella. The common story is one of long working days, unpaid training periods, and very low

salaries because you are expected to work on commission. In our respondents’ experiences, salaries at call centers barely covered their transportation and meal expenses. Alas, for many young people in Mexico City, these call centers are some of the only non-manual jobs you can access without a college education. Daniel (male, 23 years-old) has a Bachelor’s degree in Political Science and has applied to numerous jobs he finds online, though he often does not get a call back because he lacks work

• experience. Jobs at this level often expect applicants to have one or two years of experience,

even though they have just graduated college. Daniel would like to get work as a political analyst and a writer, but most of the jobs he finds online for political scientists are to work in government, as a bureaucrat, which would keep him out of his desired path. And even for those, he does not have the desired experience. We ask him whether universities could do more to provide work experience and job opportunities before graduation, he says:

They do job fairs and events like that, but they are for employers that are… you know? Like Starbucks… like, they are very generic jobs that truthfully have nothing to do with our specialization, […] the job advertisements at UNAM have not helped very much either, I think there should be some kind of arrangement, maybe with employers…. In my area I just haven’t found it…

A small group of our respondents experienced discrimination because of their alternative looks

25 and style, or even their identity. We heard young men with piercings and tattoos tell stories of being turned away from jobs only because of how they looked. One respondent even spoke of moving forward with the interviewing process until his previously covered tattoos were visible. One of our respondents spoke of being excluded from opportunities for being a transgender woman, despite being qualified and experienced, because her official documentation does not match her gender identity. A few of our respondents spoke of being interested in less- conventional career paths (i.e. art related occupations, tech occupations) but not always being able to find entry-level opportunities in their desired area. The narratives of discrimination were varied, but all resulted in stunted educational and

• ccupational opportunities for our respondents, though some of them adapted and worked around
• them. After dropping out of high school because his image clashed with teachers and school

regulations, Cristian (23 years-old) went to technical school for a certificate in electronics repair and taught himself to fix computers. He now repairs computers, cell phones, and other electronics from home and continues to teach himself how to repair equipment. Despite his talent and his desire to learn more, he does not feel he fits in the conventional educational system or in a regular job. He says “I have tattoos, piercings, my hair is red! My whole physical look blocks my entry into many places.” He has encountered prejudice in school and in the labor market, where people rely on “good presentation” too much. “Good presentation” is a very commonly listed requirement in job advertisements, most of the time it means having a conventional clean- cut look, for women it also means being attractive and well-groomed. When we ask Carlos if there is a place where he does not face this kind of discrimination he responds without hesitation: “at home!” Despite their disappointment with the educational system and the labor market, most of

26

• ur respondents agreed there is great value in schooling, many desire career paths that require

extensive training and school attendance. A few think that the value in education relies mainly on having a diploma you can show to employers to get a job, this view was especially true of those who failed short of getting a high school diploma and were now frustrated by the slim pickings in the labor market for those without a high school education. We heard many stories of even the most menial jobs requiring more education these days. An example we often heard was that “even cleaning staff were now expected to have a high school diploma.” A few others expressed that the value of education to them relies on the skills and knowledge acquired. These were more likely to be respondents who aspired to attain higher schooling and to do more specialized work. Discussion Our interviews reveal that NINIs have a clear understanding of their situation and how it disadvantages them. We find that many of them have faced institutional barriers and lack of vocational support in school. The perceived structural and institutional obstacles are inconsistent with popular perceptions of NINIs. However, despite setbacks, many of them express desire to access education or employment in the future. Our quantitative data analysis also shows that there is temporal variation in the risk of being a NINI, and that activity status is not constant over time even within the relative short-term explored here. Women are more likely to be NINIs than men. Most of the youth in our samples live with their parents, though many in the ENOE sample also lived with their in-laws. Men with some college education are more likely to be NINIs while higher education does not have a significant effect for women. Our data analysis shows a bifurcation in the paths of youth, for some, higher education leads to increased risk of being a NINI, while for others, higher

27 education is associated with being a full-time student. Given what we see in the experiences of our interviewees, more work is needed to explore if perhaps a difficult transition from college into work is related to these findings. Most

• f our interviewees who had college degrees described their struggles getting jobs, on the other

side of the spectrum, we also had many NINIs who could not find a place in the labor market because they did not have a high school diploma. Another problematic transition evident in our findings is between high school and college. Many of our respondents were stuck in this transition period and found it difficult to gain entrance to university even after years of trying. While being in a union is associated to working full time among men, women in unions were more likely to be NINIs this, combined with the low rates of female labor force participation relative to men’s, shows that family transitions among youth are strongly related to women’s drop out from school and work. We saw a similar pattern among the few mothers we

• interviewed. Further, we find that household level characteristics have an important effect in the

activity status of youth, we see that position in the household relative to the householder and the socioeconomic status of the family are related to activity. Furthermore, a key finding is how much these patterns and determinants differ by sex. We find that higher education of the household head is associated to an increased risk of being a NINI, this is consistent with the characteristics of our interviewees as well. Interestingly, it seems that a significant proportion of NINIs are inactive because they can “afford” to be, some could afford to not work as the prepared for college admissions, others could afford to hold off until they found a satisfactory

• job. We believe this is not a luxury that poor youth can allow themselves, the informal labor

market is full of youth from disadvantaged backgrounds doing menial and dead end jobs. These types of situations are not well reflected in our qualitative analysis (though we do have a few

28 respondents from lower SES backgrounds), because poor youth are less likely to have finished middle school and are more likely to have been working since their early teenage years, albeit most likely in low status, informal, and dead end jobs. The context of opportunities in the community is important as well, we see that more workers in the secondary and tertiary sectors discourage staying in school full-time. However, we should acknowledge that our measure of labor context is limited. In Mexico City, the labor markets that people have access to often span large areas of the city. People are used to long commutes, and to living in the fringes of the city and traveling to the core every day for their

• jobs. The same goes for educational opportunity, since Mexicans are not restricted to attending

schools in their neighborhoods, they may face a considerable commute to attend the school of their choice. The geographic influence of the city is larger than the individual municipality or delegation, in future work we hope to design an indicator that could more adequately measure these processes. Another potential limitation is that our qualitative evidence is only a snapshot of the experiences of NINIs and do not include youth in other activities. To overcome the first limitation, we focused on collecting as much retrospective information in our interviews, especially on trajectories of school and work experience. We also collected information on the social context of these individuals, to properly situate their experiences. Although our limited time and resources did not allow us to collect similar data on other youth, we decided that focusing on NINIs only was a good approach. Mainly because of their salience in public discourse, but also because their motivations and experiences are less well documented. We can conclude from our exploration of these narratives and data that being a NINI has more to do with lack of opportunity and structural constraints, than laziness and demotivation. Our respondents’ narratives show a desire to overcome this status and to move on with this stage

29 in their lives. For some this would mean finally being admitted in college, while for others it would mean finding a well-paying steady job. It is clear from our analyses that both economic

• pportunity and family background are close determinants of a successful transition into

adulthood. Many potential policy recommendations emerge from our findings. The priority for policy makers should be to create targeted programs that improve access to higher education and entry-level work opportunities for these youth. Access to school should include an increase in enrollments for local universities for the students who today are excluded. Programs could also include efforts to improve vocational training and advising in high school to improve the transition into college or into the labor market. Exclusionary and discriminatory practices in the labor market should be banned and salaries should also be better regulated to avoid the exploitation of young workers. Schools, employers, and government should design programs that promote the transition into a first job, using probationary periods instead of requesting unrealistic job experience for youth trying to build their resumes. School-to-work training programs and apprenticeships would be ideal to make connections between education and real-world experience and can serve as appropriate transitions between the skills learned in school and a job commensurate to these skills. Programs like these should also include training for those not interested in pursuing higher education after completing high school, this should ease the transition into the labor market for those interested in a particular trade or technical occupations. Our next step is to further explore our data on time-use and on professional aspirations of NINIs to try to demonstrate that even though youth’s trajectories in school and work are not always direct, they do follow specific goals and strategies to make the best out of their

• limitations. Many of our respondents demonstrated that they have a clear idea of what they need

30 to accomplish in order to achieve their goals in the long term and they spend their time

• accordingly. We would also like to explore time-use patterns to learn how NINIs engage in

productive activities that are not officially classified as employment in surveys. Preliminarily, we can say that time use patterns reflect interesting trends. On one hand, time use may be related to educational or occupational ambitions (i.e. studying for college entrance exams, learning a skill

• nline, doing odd jobs etc.), while in others, it reveals the burden of family obligations (i.e.

caring for a sick relative or taking care of household chores). We should explore these patterns more closely, especially to understand whether care activities burden lower SES youth more than

• thers and whether educational activities are more likely to be observed among higher SES

respondents.

31 References Arceo-Gómez, Eva Olimpia and Raymundo M. Campos-Vázquez. 2012. “Los NiNis En México.” MPRA Paper. Retrieved April 17, 2014 (http://www.inegi.org.mx/RDE/RDE_08/Doctos/RDE_08_Art6.pdf). Giorguli Saucedo, Silvia E. 2009. “Caminos Divergentes Hacia La Adultez En México.” 123–63. Instituto Nacional de Estadística Geografía e Informática. 2016. “Encuesta Nacional de Ocupación Y Empleo.” International Labor Organization. 2010. Trabajo Decente y Juventud en América Latina. Lima. Leyva Parra, Gerardo and Rodrigo Negrete Prieto. 2014. “NiNi: Un Término Ni Pertinente Ni Útil.” Coyuntura Demográfica (5):15–20. Miranda, Ana. 2015. “Sobre La Escasa Pertinencia de La Categoría NI NI : Una Contribución Al Debate Plural Sobre La Situación de La Juventud En La Argentina.” Revista Latinoamericana de Políticas y Administración de la Educación 2(3):60–73. Murayama, Ciro. 2010. “Juventud y crisis: ¿hacia una generación perdida?” Economía, UNAM 7(20). Negrete Prieto, Rodrigo and Gerardo Leyva Parra. 2013. “Los NiNis En México : Una Aproximación Crítica a Su Medición.” Revista Internacional de Estadística y Geografía 4(1):90–121.

• OECD. 2016. Panorama de La Sociedad 2016. Un Primer Plano Sobre Los Jóvenes.

Olivares Alonso, Emir. 2015. “La Jornada: Presentan 130 Mil Aspirantes Examen Para La UNAM; Se Quedarán 11 Mil.” La Jornada, March 2, 43. Pérez-Baleón, Guadalupe Fabiola. 2012. “Desigualdades de Género En El Inicio de La Vida Laboral Estable.” Papeles de Población 72:213–46.

32 Poy Solano, Laura. 2010. “La Jornada: Los Ninis, Fracaso Del Estado: Especialistas.” La Jornada, August 22, 2. Roldan, Nayeli. 2016. “Programas ‘nini’ Para Jóvenes: Ni Les Garantizan Empleo, Ni Educación.” Animal Politico. Retrieved April 18, 2017 (http://www.animalpolitico.com/2016/03/programas-nini-para-jovenes-ni-les-garantizan- empleo-ni-educacion/). Rosique Cañas, José Antonio. 2013. “Problemas estructurales de la educación superior en México: a los ‘nini’ ni los educan ni los contratan.” RIESED - Revista Internacional de Estudios sobre Sistemas Educativos 1(1–2):107–15. Zepeda, M. 2013. “Los ‘ninis’ Son ‘ninis’ Porque Quieren: 58% de Los Mexicanos.” Animal

• Politico. Retrieved April 18, 2017 (http://www.animalpolitico.com/2013/08/los-ninis-son-

ninis-porque-quieren-58-de-los-mexicanos/).

33 Tables and Figures Figure 1. Activity Status across the Observation Period, Men, Ages 18-25

Source: Encuesta Nacional de Occupacion y Empleo, Mexico 2015-2016

Figure 2. Activity Status across the Observation Period, Women, Ages 18-25

Source: Encuesta Nacional de Occupacion y Empleo, Mexico 2015-2016

6.2 12.2 6.5 8.1 4.7 31.3 23.7 30.5 29.3 33.8 8.3 8.2 10.4 8.5 9.4 54.2 55.9 52.6 54.0 52.1

0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 Apr-Jun 15 Jul-Sep 15 Oct-Dec 25 Jan-Mar 16 Apr-Jun 16

Percent

NINI Full-time student Student and worker Full-time worker

29.1 34.7 27.7 32.4 25.8 30.4 22.5 28.2 27.3 30.9 6.7 5.5 7.0 5.6 7.2 33.8 37.3 37.1 34.7 36.1

0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 Apr-Jun 15 Jul-Sep 15 Oct-Dec 25 Jan-Mar 16 Apr-Jun 16

Percent

NINI Full-time student Student and worker Full-time worker

34 Table 1. Sample Characteristics by Sex and Activity Status, ENOE 2015-2016 Men Women NINI Full- time Student Student and Worker Full- time Worker NINI Full- time Student Student and Worker Full- time Worker % % % % % % % % Age (mean) 20.5 20.7 21.6 22.2 21.5 20.6 21.5 22.1 Education Middle school or less 16.4 4.4 8.8 37.1 39.7 3.1 4.4 29.7 Some high school 21.4 23.2 18.0 10.2 12.4 21.7 15.3 9.3 High school 26.8 21.8 23.5 35.3 30.7 22.4 19.7 37.5 Some college and more 35.4 50.7 49.8 17.4 17.2 52.8 60.5 23.5 In marital or cohabiting union 4.5 0.7 5.2 29.4 53.0 2.4 5.8 24.7 Household headship Male head, two parents 67.0 66.6 62.6 60.7 70.2 67.5 61.2 59.5 Female head, two parents 5.1 5.3 5.9 7.2 6.4 4.7 6.8 7.1 Male head, single parent 9.5 7.3 8.8 9.1 4.4 3.4 5.4 6.1 Female head, single parent 18.5 20.8 22.8 23.0 18.9 24.4 26.5 27.2 Education of household head Middle school or less 20.5 20.6 25.6 30.4 30.3 18.9 22.8 30.8 Some high school 33.3 22.3 28.4 39.3 35.6 27.9 27.2 36.7 High school 19.4 26.5 22.8 19.5 21.8 23.0 18.0 16.7 Some college and more 26.8 30.6 23.2 10.9 12.3 30.3 32.0 15.8 Relationship to household head Household head 3.3 1.6 4.3 14.4 1.5 1.1 4.4 3.4 Spouse 0.3 0.0 0.2 1.0 25.1 0.2 3.4 11.7 Son or daughter 77.7 80.6 79.2 63.5 48.4 84.5 77.2 64.1 Son- or daughter-in-law 0.6 0.0 0.2 6.0 15.6 0.9 0.7 6.1

35 Other 16.4 15.3 14.5 14.7 9.0 12.8 12.2 11.7 No relation 1.8 2.4 1.7 0.5 0.3 0.6 2.0 3.1 Household economic characteristics Dependency ratio (mean) 0.3 0.3 0.3 0.4 0.5 0.3 0.2 0.4 Household income per capita (median, MXP) 1,009 1,061 2,758 1,774 1,186 1,163 2,257 1,872 Period Apr-Jun 15 16.4 20.9 19.2 20.1 20.4 21.7 20.8 18.8 Jul-Sept 15 32.7 16.4 18.7 21.0 23.6 15.8 17.7 20.9 Oct-Dec 15 17.0 20.0 22.5 19.2 17.6 20.2 21.4 20.4 Jan-Mar 15 21.1 19.7 19.2 20.0 21.2 19.5 16.3 20.0 Apr-Jun 16 12.8 23.0 20.4 19.7 17.2 22.9 23.8 20.0 N= 336 1,352 422 2,402 1,305 1,241 294 1,613 % of sample 7.4 30.0 9.4 53.2 28.9 27.5 6.5 35.7 Total N= 4,512 4,453 Mean

• Std. Dev

Min Max Economic characteristics of municipality/delegation Proportion of LF in secondary sector 18.3 5.1 8.9 42.0 Proportion of LF in commerce 21.9 3.7 14.0 38.1 Proportion of LF in services 55.6 7.5 29.8 73.9

Figures represent percentages unless otherwise indicated Source: Encuesta Nacional de Occupacion y Empleo, Mexico 2015-2016

36 Table 2. Relative Risk Ratios from Multinomial Logistic Regression Predicting Activity Status, Ages 18-25, Mexico 2015-2016 Relative Risk Ratios relative to being a full-time worker Men Women NINI Full- time Student Student and Worker NINI Full-time Student Student and Worker Age 0.59 * 0.55 * 0.74 * 0.81 * 0.62 * 0.75 * Education Some high school (ref.) Middle school or less 0.42 * 0.11 * 0.20 * 1.00 0.11 * 0.15 * High school 0.72 ** 0.62 * 0.61 * 0.66 * 0.48 * 0.48 * Some college and more 3.53 * 6.59 * 3.61 * 1.03 2.87 * 2.88 * In marital or cohabiting union 0.38 * 0.10 * 0.38 * 4.16 * 0.42 * 0.35 * Household headship Male head, two parents (ref.) Female head, two parents 0.59 * 0.62 * 0.68 0.88 0.58 * 0.94 Male head, single parent 0.80 0.70 ** 0.85 0.87 0.39 * 0.78 Female head, single parent 0.71 * 0.79 * 0.92 0.75 * 0.77 * 0.84 Education of household head Middle school or less (ref.) Some high school 1.31 0.82 0.90 0.85 1.21 0.86 High school 1.49 * 1.63 * 1.25 1.26 * 1.78 * 1.02 Some college and more 2.31 * 1.91 * 1.31 0.94 2.07 * 1.19 Relation to the Householder Household head (ref.) Spouse 3.37 0.00 * 1.60 1.67 0.20 * 0.85 Son or daughter 2.28 * 3.04 * 1.62 1.85 * 2.19 * 0.52 Son- or daughter-in-law 1.39 0.00 * 0.29 2.09 * 1.87 0.30 Other 2.33 * 2.51 * 1.35 1.55 2.88 * 0.70 No relation 3.04 * 4.47 ** 2.28 * 0.18 * 0.63 0.51 Household economic characteristics Dependency ratio 0.60 * 0.93 0.83 1.35 * 0.68 * 0.57 * Household income (logged) 0.89 * 0.90 * 1.02 0.93 * 0.92 * 1.01

37 Economic characteristics of municipality/delegation Proportion of LF in secondary sector 1.00 0.95 * 0.97 0.99 0.98 1.05 Proportion of LF in commerce 1.01 0.96 * 0.96 1.01 0.97 1.02 Proportion of LF in services 1.01 0.96 * 0.99 0.98 0.96 * 1.05 * Period Apr-Jun 15 (ref.) Jul-Sept 15 1.62 * 0.63 * 0.80 1.09 0.47 * 0.58 * Oct-Dec 15 1.06 0.99 1.24 0.75 * 0.70 * 0.85 Jan-Mar 15 1.21 0.90 0.98 1.00 0.70 * 0.62 * Apr-Jun 16 0.78 1.11 1.11 0.80 ** 0.96 1.03 Constant 4092 9855024 998 135 83536 4.6 Prob>chi2 0.0000 0.0000 Pseudo R2 0.2588 0.2265 N= 4,512 4,453 *p<0.05 **p<0.10

Source: Encuesta Nacional de Occupacion y Empleo, Mexico 2015-2016