A DYNAMIC MODEL OF SELF-EMPLOYMENT AND SOCIOECONOMIC MOBILITY AMONG RETURN MIGRANTS: THE CASE OF URBAN MEXICO

Abstract

Return migrants engage in high rates of self-employment, which scholars commonly attribute to the accumulation of financial and human capital while working abroad. Central to this scholarship is the assumption that self-employment is positive and leads to upward economic mobility among return migrants. This scholarship is limited, however, because it relies on large surveys and cross-sectional census data that treat self-employment as a single uni-dimensional status measured at one point in time. To improve conceptualization and measurement of self- employment, we engage three bodies of research that have thus far had little cross-fertilization: the literature on work and self-employment in Latin America, the scholarship on return migration and self-employment, and developments in economic theories of international migration. Drawing on results from the first longitudinal analysis of the labor market trajectories of Mexican return migrants in a large urban area in central Mexico, we identify three types of self-employment – survivalist, temporary, and prosperous. To explain these divergent self-employment pathways, we draw on biographical narratives and identify two sets of mechanisms – human capital formation and life-course stage. Overall, our investigation of self-employment types suggests a complex relationship between international migration experiences and the labor market mobility of return migrants which cannot be understood without taking into consideration migrants’ social and economic circumstances before, during, and after migration. Consequently, our study yields insights into economic theories of international migration and provides direction for future research on return migration and labor market reintegration.

Introduction

Commonly thought of as a “nation of emigrants” (Fitzgerald 2008), Mexico is now experiencing unprecedented levels of return migration. While circular migration has long been characteristic of the U.S.-Mexico migration system, improved economic conditions in Mexico, the U.S. recession, and increased interior and border enforcement by the U.S. Government have escalated return migration to Mexico. Between 2005 and 2014, 2.4 million Mexicans returned home from the United States, double the figure from the previous decade (Gonzalez-Barrera 2015). Mexican returnees are increasingly opting not to go back to the small rural hamlets from which they emigrated (Masferrer and Roberts 2012), and are instead settling in urban areas with diverse labor markets where they can apply the skills they acquired in the U.S. and pursue pathways to economic mobility that include self-employment (Garip 2016; Hagan, Hernández-León, and Demonsant 2015; Hernández-León 2008).

Scholars have long documented high rates of self-employment among return migrants, which they commonly attribute to migrants’ accumulation of financial and human capital while working abroad (Hagan and Wassink 2016; Massey and Parrado 1998; Mezger Kveder and Flahaux 2012; Dustmann 1999).¹ Indeed, migration theories view international movement as a strategy through which would-be entrepreneurs travel abroad to accumulate savings, thus overcoming barriers in local labor markets such as limited access to business loans (Lindstrom 1996; Lindstrom and Lauster 2001; Parrado and Gutierrez 2016). Central to this scholarship is the assumption that self-employment upon return leads to economic mobility. Yet, labor market research on Mexico and other developing countries reveals that self-employment is a heterogeneous job status that includes both successful entrepreneurs and marginalized workers unable to find alternative jobs (Levy 2008; Perry et al. 2007; Gindling and Newhouse 2014). Migration scholars also document similar heterogeneity among international migrants, identifying those who migrate with a specific earnings target in mind, but also those who migrate as a result of low wages and blocked mobility in their local economies (Massey and Espinosa 1997; Villarreal and Blanchard 2013).

To date, scholars studying return migration and self-employment in Mexico have relied primarily on large datasets like the Mexican Migration Project and the Mexican Census to examine the labor market integration of return migrants. They find that relative to non-migrants, return migrants exhibit a high rate of self-employment (Massey and Parrado 1998; Sheehan and Riosmena 2013; Lindstrom 2013). However, this scholarship tends to reduce complex processes of international migration and labor market reintegration down to simple linear relationships with straightforward interpretations – e.g., international migration facilitates the accumulation of financial capital, which return migrants use to start desired businesses upon return. While important, these studies mask considerable heterogeneity among return migrants, thus limiting our ability to discern the relationship of self-employment to prior migration experiences as well as its implications for returnees’ economic mobility.

In this paper, we draw on results from the first longitudinal analysis of the labor market trajectories of Mexican return migrants to examine the micro-level processes associated with return migrants’ labor market reintegration and economic mobility. We focus on movement in and out of self-employment, which we view as simultaneously enabled by and parallel to international migration. In some cases, international migration and accumulated savings may facilitate successful business formation upon return, while in other cases migration and self-employment may be simply alternative responses to persistent economic marginalization.

To examine these alternative labor market trajectories, we rely on biographical narratives which we constructed from in-depth interviews conducted with return migrants five years apart (2010 and 2015). The narratives integrate migrants’ work histories, family and social contexts, financial capital accumulation, human capital formation, and life-course stages in the migratory circuit (before migration, in the United States, and after migration). The biographical narratives allow us to explore the heterogeneity of self-employment among return migrants and associated variations in their reasons for traveling to the United States and efforts to achieve economic mobility upon return to Mexico. Our analysis is guided by four questions: 1) What are the dimensions of self-employment among return migrants and how are they to be measured? 2) How is self-employment related to upward or downward economic mobility? 3) What mechanisms explain these different mobility pathways? 4) How can return migrants’ self-employment experiences inform economic theories of international migration?

The remainder of our paper is organized into five sections. To frame our analysis, we first engage three bodies of research that have thus far had little cross-fertilization: the literature on work and self-employment in Latin America, the scholarship on return migration and self-employment, and developments in economic theories of international migration. We then describe our site selection and research design, with an emphasis on how we constructed our interviewees’ biographical narratives. Next, in the results section, we use our respondents’ biographical narratives to identify multiple self-employment types and to explain movement into and out of these different categories. Then, drawing on our longitudinal analyses, we develop three propositions to guide future studies of return migration. We conclude with a reflection on the implications of our results for theory and research.

Work, International Migration, and Self-Employment

Work and Self-employment in Latin America

The informal economy, which is the largest economic sector in most Latin American countries, includes firms that are not formally registered, regulated, or taxed by the government. Firms in the informal economy do not provide health insurance, pensions or severance pay, or written contracts, and generally employ fewer than five workers (Arias et al. 2010; Levy 2008; Perry et al. 2007). The small size of informal sector firms is explained as a function of inefficient credit markets and onerous costs of formal registration that limit growth potential among prospective entrepreneurs (Arias et al. 2010; Tornarolli et al. 2014). In Latin America, Mexico’s economy is on the high end of the informal spectrum, with 66 percent of its workforce employed in this sector (SEDLAC 2015) which for the most part, offers extremely limited opportunities for upward mobility(Arias et al. 2010; Levy 2008).

Because of the precarious working conditions and low and stagnant wages associated with informal sector labor, self-employment represents the primary upward mobility pathway for individuals with low levels of formal education who thus face significant barriers to attaining formal sector employment (Bargain and Kwenda 2011; Perry et al. 2007; Huerta-Wong, Burak, and Grusky 2013).² In 2010, 28 percent of the economically active Mexican population was self-employed (INEGI 2011), including more than 40 percent of informal sector workers. Studies of Mexican workers who transitioned from informal wage work to self-employment find that they have higher job satisfaction than informal sector employees because of increased earnings and autonomy (Perry et al. 2007; Bargain and Kwenda 2011).

Self-employment, however, does not always lead to upward mobility (Cunningham and Maloney 2001; Perry et al. 2007). In their analysis of Latin American workers, Perry and colleagues (2007, 6) found that although self-employed workers on average fare better than their informally employed counterparts, approximately one-third of them “appear to be so largely involuntarily; they would prefer formal jobs.” For these laborers, self-employment functions as an alternative to unemployment following a failed job search (Levy 2008).³

Because of these opposing self-employment pathways scholars now distinguish between “successful” and “unsuccessful” self-employment (Gindling and Newhouse 2014). Research on self-employment in Latin America finds that voluntary entry and the presence of employees indicate successful self-employment, while involuntary entry and the absence of employees reflects unsuccessful self-employment (Cunningham and Maloney 2001; Levy 2008). Maloney (2004), for example, found that workers in Colombia and Mexico who voluntarily left wage labor to start businesses earned significantly higher wages than those who only became self-employed after a failed job search (See also Mandelman and Montes-Rojas 2009). In their analysis of self-employed workers in 74 developing countries, including Mexico, Gindling and Newhouse (2014) found that two-thirds of employers reside in the top income tercile, relative to just 40 percent of self-employed workers without employees.⁴ Similarly, Grimm, Knorringa, and Lay (2012) found that among self-employed workers in seven West African countries, only those in the top income quartile averaged at least one employee. According to Mexico’s 2010 Survey of Occupation and Employment, just one in five self-employed workers had employees (INEGI 2011), suggesting that most self-employed Mexicans, even those who may derive personal satisfaction from their work, struggle to expand their businesses over time.

Indeed, in Mexico, self-employment is often a temporary status, highlighting both the challenges faced by nascent entrepreneurs and the frequent use of self-employment as an alternative to unemployment. One study found that more than a quarter of self-employed workers in Mexico change employment statuses every year (Fajnzlber, Maloney, and Rojas 2006). Due to the rapid failure of underperforming firms, cross-sectional surveys likely overestimate the rate of successful self-employment that results in economic mobility. Moreover, cross-sectional analyses will obscure the use of self-employment for temporary income by individuals engaged in active job searches, a strategy that may be especially prevalent among return migrants re-entering local labor markets in origin countries.

The research on Latin America has under-theorized self-employment as a mobility pathway by restricting its focus to single dimensions of self-employment, that is, successful versus unsuccessful (Gindling and Newhouse 2014) or movement in and out of self-employment (Fajnzlber, Maloney, and Rojas 2006), but rarely both. We build upon these studies by conceptualizing self-employment along two dimensions simultaneously: quality and temporality. Quality refers to personal and economic value of the work (i.e., job satisfaction, autonomy, earnings, presence of workers, and growth potential). Temporality captures workers’ propensity to move in and out of self-employment over time. Figure 1 depicts our proposed scheme. As the figure shows, self-employment classified in terms of quality and temporality can be low quality and temporary, low quality and permanent, high quality and temporary, and high quality and permanent. The dashed lines in Figure 1 illustrate the blurriness of these boundaries – both quality and temporality are fuzzy categories which necessarily include wide variations that may be difficult to assess in large, survey-based studies. We believe however that assessing individual return migrants’ self-employment ventures along the lines of quality and temporality throughout the life course can make an essential contribution to research on return migrants’ labor market reintegration and social mobility.

Figure 1.

Classifying the Dimensions of Self-Employment

Return Migration and Self-employment

Numerous studies conducted in Mexico and other countries with substantial return migration also identify high rates of self-employment among return migrants relative to non-migrants (Lindstrom 2013; Massey and Parrado 1998; Mezger Kveder and Flahaux 2012; McCormick and Wahba 2001; Hagan and Wassink 2016; Cassarino 2008). Some scholars explain return migrants’ concentration in self-employment as a function of foreign savings and remittances or “migradollars” that returnees use to fund new businesses (Massey and Parrado 1998; Mesnard and Ravallion 2006). This explanation fits within the New Economics of Labor Migration (NELM), which argues that foreign savings are especially important for migrants returning to economically dynamic communities with ample opportunities for investment (Lindstrom 1996; Stark and Bloom 1985). Lindstrom and Lauster (2001, 1232), for example, argue that “in communities where credit is absent or prohibitively expensive, the presence of opportunities for small-scale investment encourages temporary migration as a means to acquire funds for investment purposes.” In this way, international migration allows prospective entrepreneurs to overcome a challenge inherent to informal markets – access to start-up funds.

Migrants also acquire non-financial capital while working abroad that they can mobilize upon return to launch new businesses. Several studies, for example, point to “language capital” acquired in destination countries that facilitates returnees’ entry into new businesses catering to international tourists, business people, and foreign retirees (Hagan and Wassink 2016; Williams and Baláž 2005; Dustmann 1999). Recent research in Mexico highlights the new technical skills and social competences that migrants acquire in foreign workplaces, (Hagan, Hernández-León, and Demonsant 2015; Sanderson and Painter 2011), which are associated with greater odds of self-employment upon return (Hagan and Wassink 2016). Collectively, this scholarship, which is informed by the NELM, shows that through the accumulation and transfer of financial and human capital, international migration can create an upward mobility pathway from informal wage labor into self-employment.

In contrast, neoclassical approaches to migration, which treat international migration as a response to wage differentials between origin and destination communities (Harris and Todaro 1970), view return as indicative of a failed migration. From the neoclassical perspective, individuals migrate when expected wages in the destination community exceed their current wages. Individuals return home when they fail to secure work or earn lower than expected wages in the destination (Massey and Espinosa 1997). Thus the NELM views return and entrepreneurship as the logical follow-up to a strategic migration trip while the neoclassical model posits that return indicates a failed migration and overall downward mobility.

Up to now, self-employment among return migrants has been studied primarily through the lens of the NELM. Yet, given the high rate of unsuccessful or marginal self-employment in migrant sending countries (Perry et al. 2007; Gindling and Newhouse 2014), it is entirely possible as suggested by neoclassical theory that returnees, having failed to achieve desired wages (both at home prior to migration and in the United States), turn to self-employment out of economic desperation. Moreover, given the high rate of turnover among self-employed workers in Mexico and elsewhere in Latin America, it is likely that many return migrants exit self-employment, either upon securing salaried work or following an inability to generate profit.

We identify three specific limitations in prior research on the relationship between international migration and self-employment, which inform our analysis. First, we find that there are shortcomings in the ways that most scholars have conceptualized and measured the migration experience, by merely categorizing individuals as either return migrants or non-migrants (e.g., Sheehan and Riosmena 2013), or by linearly measuring years abroad or cumulative remittances (e.g., Massey and Parrado 1998). These measurements obscure variations in the ways in which different migration experiences create different opportunities to accumulate financial and human capital, which have implications for migrants’ self-employment and social mobility trajectories upon return (Hagan and Wassink 2016). Moreover, although accumulated capital can enable self-employment among return migrants (Hagan and Wassink 2016; Massey and Parrado 1998), not all migrants accumulate new capital while abroad. Migrants who remain in low-wage, entry-level jobs in the U.S. labor market, even for many years, may lack the skills and funds necessary to initiate successful businesses or secure steady wage work, which could push them into survivalist self-employment upon return.

A second factor limiting our understanding of the place of self-employment in economic reintegration is that research on self-employment among return migrants does not distinguish between successful and unsuccessful self-employment. As already noted, self-employment in Latin America is a complex and multi-dimensional category (Fajnzlber, Maloney, and Rojas 2006; Perry et al. 2007). For some return migrants, self-employment may represent a temporary economic status while they search for salaried work. For others, it may indicate a last resort after a failed labor market reintegration. Yet, studies of return migration almost exclusively treat self-employment as a static, homogenous category expected to promote economic mobility (e.g., Hagan and Wassink 2016; Lindstrom 2013; Massey and Parrado 1998).

Finally, absent from studies of return migration and self-employment is a consideration of how life-course stage affects return migrants’ self-employment trajectories. The life-course framework, which emphasizes the intersection of individual and historical context (Elder 1985), has been instrumental in conceptualizing and identifying trajectories of family formation, health, and work, as well as immigrant incorporation in the United States (Gonzales 2011). We borrow the concept of timing from the life-course perspective to consider the role of age and structural opportunities in return migrants’ labor market reintegration. At a cohort level, timing reflects macro-structural changes that occur across generations. At an individual level, timing reflects the age at which a major event occurs.

At the cohort level, timing is important because structural opportunities change across generations. Improvements in the quality and availability of education in Mexico over the last several decades have created significant educational disparities between successive Mexican cohorts (OECD 2012). The Mexican government implemented major educational reforms in the late 1980s to address educational deficiencies at the local level. Then, in 1992, Mexico passed the National Agreement to Modernize Basic Education, which made lower secondary school (seventh through ninth grade) mandatory, with the goal of reducing educational inequalities. Migrants returning from the United States in their forties and fifties, who were not beneficiaries of educational reforms, may encounter limited opportunities for salaried employment, potentially directing them toward self-employment as an alternative to joblessness. As a result of their greater formal education, we expect members of younger immigrant cohorts to achieve greater occupational mobility while in the United States, increasing their opportunities to acquire and transfer new skills to the Mexican labor market.

We also expect age to be associated with labor market marginalization at the individual level. Workers of different ages exhibit distinct entrepreneurial preferences, owing to divergent familial and economic pressures (Cunningham and Maloney 2001; Heinz 2003). Younger workers are more likely to be unmarried, affording them the opportunity to take greater financial risks. Cunningham and Maloney (2001) have shown that successful Mexican entrepreneurs tend to launch their businesses at younger ages. By contrast, entry into self-employment later in life is associated with a higher likelihood of prior job loss and the economic necessity to support one’s family members. Not surprisingly, “being older appears to make finding alternate salaried employment more difficult” (Cunningham and Maloney 2001, 143). Hence, migrants returning later in their working lives may encounter constrained employment opportunities, forcing them to consider survivalist self-employment as a last resort.

Site Selection, Research Design, and Sample Profile

Research Site

To capture the increasing return of Mexican migrants to the country’s urban areas, we conducted our longitudinal study in Leon, a large city (1.3 million inhabitants) with a diverse industrial base and growing economy in the historic migrant sending state of Guanajuato. With a dynamic manufacturing base and a thriving service sector, Leon is an ideal site to study economic mobility and self-employment. Figure 2 compares the industrial location of Leon’s labor force with that of Mexico. As the figure shows, less than one percent of the labor force is in agriculture – the primary industry in many well-established rural migrant-sending communities. By contrast, 35 percent of Leon’s workforce is in manufacturing, twice the national average. Moroever, a recent report by Banamex-Citigroup, an international financial institution, documented annual economic growth of 7.1 percent in Guanajuato in 2016, more than three times the national average (Rodriguez 2016).

Figure 2. Industrial composition in Leon compared to the industrial composition in Mexico.

Source: Authors’ calculations based on data from the 2010 Mexican Census of Population and Housing. Estimates were weighted to adjust for the Census’s complex design. All differences are statistically significant at the 0.01 level.

^a Includes hotel and restaurant, transportation services, private household services, and other services.

^b Includes financial services, public administration, real estate and business, education, health, and social work.

^c Less than 1% of Mexicans work in mining, electricity, gas, and other utilities.

Leather, shoemaking, and textiles manufacturing are the largest employers in Leon, comprising roughly 20 percent of the city’s labor force (over half of those employed in manufacturing). In addition to the shoemaking and leather sector, Leon houses a cluster of automotive, chemical, and transportation industries, as a well as a fast growing retail, service, and hospitality sector, fueled by domestic tourism and international commerce. Leon’s diverse industrial base and rapid economic growth provides ample opportunities for return migrants to invest remitted financial and human capital in new businesses.

Research Design

The data we present here is based on a longitudinal and multi-method study, which included two rounds of semi-structured interviews (5 years apart) with a sample of return migrants in Leon, along with worksite observations of large and medium-sized factories and small family enterprises. In 2010, we conducted a survey with a sample of 200 return migrants.⁵ The interviews took place in the return migrants’ homes and worksites. The 2010 interviews, which were conducted in Spanish, averaged about one-and-a-half hours in length. The surveys included a combination of close- and open-ended questions, which captured migration histories, complete job histories, and well as migration motivations.

In 2015, we returned to Leon and, with a team of graduate students from Mexico and the United States, conducted follow-up interviews with the original sample of return migrants. To our knowledge this represents the first follow-up study of return migration and it was not without logistical challenges. Because of rapid growth and urban reconfiguration in Leon, we were unable to locate one quarter of our respondents’ addresses. Among those whose addresses we did locate, not all respondents were available to interview.⁶ From family members, new residents, and neighbors, we learned that of the 150 return migrants’, whose homes we relocated, two had died, eleven had moved away, and 37 were unreachable (neighbors and residents could not provide any information). We recognize that our sample does not capture those migrants who return home, but re-migrate to the United States. Yet, a benefit of this omission is that it restricts our analysis of labor market reintegration to migrants who have completed their U.S. migratory careers and are settling permanently in Mexico.

In the end, we successfully located and re-interviewed half (100) of the original sample.⁷ The follow-up interviews conducted in 2015 included questions about the labor market and migration histories of the return migrants since the last survey in 2010. Because our interest in this paper is in transitions in and out of self-employment, the analyses presented here are based upon 52 of the 100 respondents interviewed in 2015 who reported being self-employed at some point following their most recent return to Mexico. The fact that 52/100 return migrants that we interviewed were currently or had previously been self-employed highlights the centrality of self-employment to the labor market reintegration of return migrants. Our self-employment figure is higher than the 35 percent reported by studies using the 2010 Mexican Census because our longitudinal approach that captures both current and former self-employed returnees. When accessible, we also visited the worksites of these respondents, some of which were family-based enterprises located in or adjacent to their homes.

Although our focus is not on gendered experiences per se, we report differences in the self-employment experiences of men and women because we recognize that gender permeates the migration experiences (Hagan 1998; Hondagneu-Sotelo 1994). Female Mexican migrants often undergo role changes in the United States and can experience shifts in their household and economic roles, in many cases entering the workforce for the first time when they migrate abroad (Hondagneu-Sotelo 1994). Thus, it is probable that the U.S. migration experience may have distinct effects on the self-employment experiences of women who return to Mexico, where women’s labor market opportunities are more limited relative to men.⁸

Sample Profile

Table 1 provides a summary profile of our respondents, including their demographic backgrounds, work histories, and migration experiences. As Table 1 shows, on average they possess fairly low levels of formal human capital as measured by an average of seven years of formal education, a finding consistent with the other studies that report that Mexican migrants and return migrants have fewer years of schooling than non-migrants (Rendall and Parker 2014; Campos-Vazquez and Lara 2012). Despite their limited schooling, our respondents possess considerable human capital as measured by work experience. All but four respondents worked in Mexico prior to migrating, averaging six years in the Leon labor market before traveling to the United States. On average, our respondents returned to Mexico with 10 years of combined work experience, ample opportunity to develop new skills and abilities.

Table 1.

Respondents’ demographic backgrounds, work histories, and migration experiences (n=52)

	Median/Percent	(Min/Max)
Demographic background & pre-migration work history and educational attainment
% male	81%	--	--
Mean age at 2015 interview (years)	42	(27	83)
Marital status at 2015 interview
% Single (Never married)	10%	--	--
% Married	79%	--	--
% Divorced/widowed	12%	--	--
% Had children at 2015 interview	86%	--	--
Mean years of education	7	(0	17)
% in labor force prior to migration	92%	--	--
Mean years of work experience before first migrationa	6.0	(0	31)
Context of departure, US work history, and migration history
Median age at first migration	24	(2	54)
% in US labor force	98%	--	--
% Migrated multiple times	42%	--	--
Mean years of cumulative US work experience	4	(0	24)
Context of return & post-migration work history
Mean age at most recent return	33	(15	66)
Reason for returnd
% Family reunification	56%	--	--
% deported	15%	--	--
% Completed migration goal	13%	--	--
% Other reasonse	29%	--	--
% in labor force at 2015 interview	100%	--	--
Mean years in labor force since returning from US	14	(5	50)

^aWe suspect that many respondents actually attained significantly greater work experience by helping in their parents’ work as children and teenagers.

^bPercentages do not sum to 100 because respondents could list multiple reasons.

^cIncludes: Arizona, Colorado, Florida, Georgia, Illinois, Louisiana, Nevada, North Carolina, Ohio, South Carolina, Wisconsin.

^dRefers to most recent trip among migrants who made multiple trips. Numbers do not sum to 100% as respondents could cite multiple reasons.

^eOther reasons for return included: boredom (3), end of work contract (2), loneliness(1), fear of deportation and racism (2), difficulty finding work (1), a medical operation (1), higher education (1), personal problems (1), and to start a business (1).

Our respondents’ migration experiences varied considerably. While, on average, they traveled to and returned the United States as unmarried young adults, some migrated late in their working careers, after they had married and established families in Mexico. Once in the U.S. labor market, men and women entered distinct occupational niches. Most men found work in agriculture, construction, and manufacturing (63 percent). In contrast, five of the ten women initially entered live-in domestic work. Women respondents were much less likely than the men to successfully exit low-skilled “gateway” occupations. Isolated in their jobs as live-in domestic workers they were unable to transition out of this industry. These gendered jobs condition men and women’s opportunities for occupational mobility, higher wages, and potential skills learning (Hagan 1998).

Despite our respondents’ high rate of self-employment, business formation was not a commonly cited reason for returning to Mexico. When we asked individuals why they returned to Mexico, 56 percent – men and women alike – cited family reunification as the primary motivation, a finding consistent with other recent return migration studies (MATT 2013). Another 15 percent reported being deported by the U.S. government and 29 percent returned for other reasons, including discrimination, nostalgia, illness, or job loss. Indeed, despite the fact that all of our respondents became self-employed upon return, just six (13 percent) cited successful completion of their migration goals as the reason for their return.⁹ Despite recent evidence that the decline in Mexico-U.S. migration and increase in return migration was driven in large part by the Great Recession (Gonzalez-Barrera 2015; Villarreal 2014), only three of our respondents attributed their returns to job loss or the termination of a contract in the United States.

Results

Identifying Types of Self-Employment

In contrast to the literature on return migration that generally treats self-employment as a single employment category, we propose a multi-dimensional typology that reflects variations in self-employment along two dimensions: quality (i.e., job satisfaction, autonomy, earnings, presence of workers, and growth potential) and temporality (i.e., the relative permanence of self-employment ventures). Figure 1 in the background section illustrates these two dimensions. Although these two dimensions suggest four categories of self-employment, we only identified one respondent whose business could be classified as high quality but temporary.¹⁰ In other words, our respondents who started prosperous businesses tended to remain self-employed. Thus, we propose three types of self-employment: 1) survivalist self-employment (low level of job quality and permanent self-employment); 2) stepping-stone self-employment (low level of job quality and temporary self-employment); and 3) prosperous self-employment (high level of job quality and permanent self-employment).¹¹

Table 2 presents business profiles for each respondent by self-employment type. In terms of quality, survivalist self-employment appears to resemble the unsuccessful self-employment that development scholars have documented among informal sector workers throughout the developing world (Gindling and Newhouse 2014). As the table shows, the activities of the survivalist self-employed ranged from unreliable contract work to ambulatory food and beverage stands to appliance repair and reselling used clothing and other items in local flea markets. Most of the survivalist self-employed commented on the instability and low earnings associated with their work. Several of our older self-employed respondents in this category reported that they were unable to exit these marginal ventures because they lacked adequate savings and/or pension payments to retire. At our follow-up interview in 2015, none of those in the survivalist category had exited or expanded their businesses, reflecting the permanency of survivalist self-employment. We argue that survivalist self-employment is a last resort used by migrants unable to find wage work after returning to Mexico.

Table 2.

Business activity and size in 2015 by self-employment type

Survivalist Self-employed (13)		Stepping-Stone Self-employed (16)		Prosperous Self-employed (23)
Business activity	Business Has employees	Business activity	Business Has employees	Business activity	Business Has employees
Construction contractor	--	Small grocery store	--	Auto repair	Yes
Domestic	--	Small grocery store	--	Auto repair	--
Leather worker	--	Small grocery store	--	Auto repair	Yes
Repair business	--	Small grocery store	Yes	Carpentry	Yes
Second-hand shop	--	Auto mechanic	--	Carpentry	Yes
Sells clothes/shoes	--	Bar & restaurant	Yes	Carpentry	Yes
Sells clothing	--	Glasswork business	--	Carpentry	Yes
Street food cart	--	Juice stand	--	Construction	Yes
Sells shoes	--	Manufactured knapsacks	--	Construction	Yes
Sells used appliances	--	Shoemaker	--	Construction	Yes
Sells used clothes	--	Sold used clothes	--	Cyber café	Yes
Sells used clothes	--	Sold used clothes	--	Grocery/restaurant	Yes
Strugling artist	--	Sold used clothes		House cleaning	--
		Sold used clothes	--	Manufactures boots	Yes
		Taco vendor	Yes	Manufactures drum covers	--
		Welder	--	Manufactures purses	Yes
				Nutritionist	Yes
				Restaurant	Yes
				Restaurant	--
				Taxi driver	--
				Taxi driver	--
				Video arcade	Yes
				Water purification	Yes

Turning to the middle column of Table 2, roughly one third of our respondents were stepping-stone self-employed. Stepping-stone self-employment includes respondents who were self-employed in 2010, but transitioned to wage labor or re-migrated to the United States between 2010 and 2015. In terms of quality, with one exception, the business ventures of the stepping-stone self-employed resembled those of the survivalist self-employed. Ten of the 16 operated small in-home grocery stores (abarrotes), sold used clothes in a local flea market, or operated ambulatory food and beverage stands. Only three of the stepping-stone self-employed migrants reported hiring workers, and in two of those cases, the employees were family members. Yet, unlike the survivalists, these stepping-stone self-employed respondents later exited self-employment for wage work. The most important difference between survivalist and stepping-stone self-employed workers is not the quality of their self-employment, but whether or not they remain self-employed across time. Given the high rate of turnover among the self-employed in Mexico (Fajnzlber, Maloney, and Rojas 2006), stepping-stone self-employment likely includes a large proportion of self-employed return and non-migrants alike. However, cross-sectional assessments of self-employment conflate survivalist and stepping-stone self-employment (e.g., Gindling and Newhouse 2014).

Turning to the third column of Table 2, we see that nearly half (44 percent) of our respondents launched self-employment ventures that facilitated economic stability, employed workers, and provided high levels of job satisfaction, reflecting high job quality and temporal permanence. Several features distinguish the self-employment ventures of these respondents, whom we term prosperous self-employed. First, they launched a far more diverse set of businesses compared to the survivalist and stepping-stone self-employed. Their enterprises ranged from auto repair shops, restaurants, and taxi drivers in the service economy, to construction, to manufacturing leather goods. Second, their businesses were located in establishments separate from their homes and often created new niches in the Leon economy. For example, several used advanced machinery acquired in the United States to improve and expand their businesses in construction, manufacturing, and auto repair. Finally, unlike the survivalist self-employed, their businesses expanded over time. When we interviewed these respondents in 2010, only seven of the 23 had employees. But, by 2015, 17 had hired workers.

Explaining Self-Employment Types

Through our respondents’ biographical narratives, we consider two sets of mechanisms that explain different self-employment experiences among return migrants: life course stage and human capital formation. Respondents who became survivalist self-employed tended to be older and possessed little formal schooling, indicative of an earlier cohort. Multiple survivalist self-employed respondents cited a lack of education as a limiting factor when searching for work upon return. Returning to Mexico in their forties and fifties, multiple survivalist self-employed respondents attributed their inability to find wage work to age-based discrimination.

Moreover, the survivalist self-employed acquired limited human capital across the migratory circuit. This group experienced little occupational mobility in the United States, with most remaining in low-skill gateway jobs, such as farmworkers, low-wage service workers, and live-in domestics. Because they were stuck in these gateway jobs, they encountered few opportunities to learn new skills, which have been shown to enable mobility among return migrants (Hagan and Wassink 2016). Indeed, only one of our 13 survivalist self-employed respondents described applying skills learned in the United States to their work in Mexico.¹² With limited skills to invest in productive enterprises, these returnees turned to the survival ventures featured in Table 2 out of financial desperation. As of 2015, these respondents had been back in Mexico for an average of 17 years and none had managed to exit their marginal businesses. For many like Lorenzo, age and limited education precluded securing alternative work:

Lorenzo returned to Mexico in 2004 in his late 40s after working on a farm in Arizona for three years. Since returning, he has struggled. After holding low-paying jobs as a maintenance worker and an assistant to a plumber in Leon, he started selling tacos out of his home. Age limited Lorenzo’s employment prospects upon return. As he put it, “nadie le dio trabajo con la edad” (No one gives you work because of your age). With few marketable skills and little formal education, Lorenzo struggles to make ends meet as an informal food vendor. Lorenzos’ description of his motivations for remaining self-employed in 2015 accurately portrays the experiences of many of our older respondents who remain trapped in survivalist businesses, “cualquier trabajo es para dar el pan de cada día a la familia” (Whatever work to provide bread each day for my family).

Even those like Valentin, who accumulated savings while abroad, tended to invest in survivalist enterprises that provided little economic mobility:

Valentin was born in Leon in 1960. He left school at 11 to work for his father, a skilled mason. From his father he learned to lay tile and mold and lay brick. In 2004, at 44-years-old, Valentin migrated to the United States in search of higher wages. He found work as a farmhand in California where he worked for six years. In 2010, Valentin returned to Leon to rejoin his family. Although Valentin told us that he did not develop any new skills in the United States that he could apply to the Leon labor market, he did manage to save some of his U.S. earnings, which he invested in a small business where he repairs old washing machines. Since 2010, his business has not grown and provides little profit. Now in his mid-50s, with only a primary education, Valentin has limited employment prospects. With his wife, father, and seven of his eleven children to feed, Valentin has little choice but to maintain his marginal business.

Lorenzo and Valentin’s experiences illustrate the constraints that family obligations impose on many middle-aged and older male return migrants. Confronted by the need to be a good provider and support spouses, children, and other relatives, these men have little opportunity to reskill or save money to expand their businesses.

Turning to the stepping-stone self-employed, we found that returning at significantly younger ages and with some schooling, these respondents were able to transition from self-employment to wage labor. The stepping-stone self-employed completed their migrations during their twenties, early in the life course, when age-based employment discrimination is not a concern. Moreover, as members of a younger cohort relative to the survivalist self-employed, these individuals benefitted from Mexico’s enhanced educational opportunities. They achieved four times the secondary school completion rate as the survivalist self-employed. However, despite their youth and relatively high educational attainment, these respondents did not launch businesses that provided economic mobility or job satisfaction. Indeed, in most cases, they engaged in relatively marginal self-employment as a source of temporary income while searching for stable work in the formal economy. We found the quality of their businesses to correspond to limited transfer of new skills that could be invested in innovative, prosperous ventures.

These respondents’ post-self-employment occupational trajectories were quite heterogeneous. While half successfully landed formal sector jobs with steady pay and benefits, others returned to informal sector employment, in several cases resuming the same low-wage work that they did prior to migration. Take the case of Veronica:

Veronica worked in Leon for more than a decade, until in 1998, hoping for economic mobility, she migrated to North Carolina. Veronica initially worked as a janitor through an area firm. After several years, she took a slightly better paying job at a Taco Bell in Raleigh, where she worked as a cook. After six years, she returned to Leon to reunite with her family. When we interviewed her in 2010, Veronica had failed to secure steady employment, and was operating a small juice stand to support herself. She did not gain any new skills while in the United States, and failed to achieve mobility upon return. In 2015, she had closed the juice stand, and was again cleaning houses, the same position she held prior to migration.

Of the eight returnees who secured stable formal sector jobs, five relied on skills that were either learned or improved while in the United States. Take the case of Jose.

After completing lower secondary school in 1975, Jose worked as an auto mechanic at a large BMW plant in Leon, where he achieved substantial mobility, recording and diagnosing major clients’ automotive issues. In 2003, Jose received an intracompany one-year visa to work at a BMW plant in California. While there, he was enrolled in a training program to work with electric motors. Returning after the expiration of his visa, Jose initially opened his own mechanic shop. But, unable to secure a loan, he closed the business. Drawing on his experience and formal training received in the United States, Jose found a job at one of several major automotive plants in Leon. When we interviewed him in 2015, he had been promoted to director of the electric motors line, a well-paying position where he receives health and retirement benefits.¹³

For migrants like Jose and Veronica, labor market trajectories upon return were shaped by human capital accumulated before and during migration. In the United States, Jose improved upon his already impressive automotive repair and assembly skills, which enabled him to secure a well-paying formal sector job upon return. Veronica, by contrast, did not acquire any new skills in the United States, and after failing to find formal sector employment in Leon, she eventually returned to what she knew, low-wage domestic work in the informal sector.¹⁴

Turning to the prosperous self-employed, we found that this group of return migrants benefitted directly from their labor market experiences in the United States, gaining new skills, ideas, and business strategies, which enabled and inspired their diverse array of entrepreneurial ventures upon return. About 40 percent of the prosperous self-employed returnees achieved occupational mobility while in the United States,¹⁵ and 19/23 applied skills learned abroad to their business ventures in Mexico. Perhaps more importantly, while the stepping-stone self-employed acquired technical and English language skills, which could be directly applied to salaried jobs, many of the prosperous self-employed returnees reported developing a set of social and organizational skills that motivated their entry into self-employment upon return.¹⁶ Indeed, in many cases, skills and experiences gained in the United States led return migrants, like Jesus, to launch new businesses in industries with which they had no familiarity prior to migration:

Jesus began working in construction when he was 11 years old, after just six years of school. When he was 18 years old, Jesus migrated to California, in search of better wages. As he put it, “aquí [en Mexico] es muy dificil” (here in Mexico it is very difficult). Upon arrival, Jesus found work in a local restaurant. In 1988, Jose moved to Illinois, where he found a more skill intensive position in an auto manufacturing plant. Through his new job, Jesus enrolled in computer classes and studied automotive mechanics. In 2006, Jesus decided to return to Leon. On arrival, he invested his U.S. savings in his own auto repair shop. He relies entirely on technical and organizational skills that he acquired working in the United States and uses a computer to track his accounts and diagnose automotive problems. When we interviewed Jesus in 2015, Jesus had two employees, earned a steady income, and took pride in his work.

Jesus’s experiences exemplify how some labor migrants who travel abroad in search of better wages and new opportunities, especially those like Jesus who migrate early in their careers, can accumulate unexpected human capital skills that create new opportunities for entrepreneurship upon return. For some of the women in our sample, migrating to the United States provided more than just new technical and organizational skills, it also emancipated them from the patriarchal society in which they originated (Hondagneu-Sotelo 1992). For Joanna, migrating to the United States demonstrated that she could rely on herself:

When she was just fifteen years old, having never worked in Mexico, Joanna left Leon to join family members in San Diego, California. She told us that she was fleeing an abusive boyfriend. Joanna explained to us how her experience of migrating by herself at such a young age created a powerful sense of self-trust and a desire for independence, which contributed to her decision to become self-employed upon return. In the United States, Joanna found work as an in-home domestic. After just six months, she was apprehended by Immigration and Customs Enforcement and deported back to Mexico. Since returning, she has actively invested in herself and her community. Drawing on her time spent in the United States, Joanna opened a clothing boutique that sells the latest U.S. styles. She also manages her husband’s carpentry business, which designs and installs U.S.-style cabinetry – Joanna researches designs, drawing from her work in American kitchens in California. Joanna and her husband have seven employees.

Consistent with the migration literature (Lindstrom and Ramirez 2010), we found many of the prosperous self-employed to be selective individuals, ambitious and driven to succeed. But our interviews also revealed how their business ventures were enabled and/or enhanced by their work experiences abroad. Some migrated to the United States with the goal of accumulating financial and human capital to invest in a desired business venture, but later incorporated skills and experiences gained in the United States into their organizational strategies. For others, like Jesus, exposure to new technologies resulted in the creation of new and unexpected labor market niches. For some, like Joanna, migration provided the independence and self-confidence necessary to launch a prosperous business. Our finding that the mobilization of new skills is directly associated with prosperous but not survivalist or stepping-stone self-employment extends existing recent research that has documented the contributions of skill transfers to return migrants’ self-employment (Hagan and Wassink 2016). We move beyond Hagan and Wassink’s assessment of the association between skill transfers and business formation, finding that the formation of new human capital while abroad can provide an essential ingredient for return migrants who have risk-takers with the ambition to achieve entrepreneurial success upon return.

In contrast to the NELM, which views international migration as a source of financial capital (money) that migrants can invest in businesses (Lindstrom and Lauster 2001), we found that remittances played a relatively minor role in our respondents’ self-employment trajectories. Although the majority of our respondents remitted money to household members while abroad, only a small percentage of those remittances were earmarked for investment, with the majority contributing to household consumption.¹⁷ This finding is consistent with other studies that find that the vast majority of remittances are directed toward household consumption, debt repayment, and investments in housing and education (e.g., Massey, Durand, and Pren 2013). Indeed, a recent investigation using data from the MMP found that only 0.9 percent of remittances and 7.1 percent of U.S. savings were invested in new businesses (Massey et al. 2013:Table 3). Most of the migrants that we interviewed did not migrate to the United States with the intention of saving and remitting financial capital to invest in a business. Our findings question the assumption that Mexicans travel to the United States with the express purpose of saving money for investment in a business upon return (Lindstrom 1996; Lindstrom and Lauster 2001).¹⁸ Although many of our respondents did invest U.S. savings into their new businesses, our interviews suggest that in most cases, urban migrants enter self-employment in response to labor market constraints encountered upon return and new opportunities created by their experiences abroad. Moreover, while accumulated savings and remittances often enable the initiation of a new business venture (Massey and Parrado 1998), we found that skills learned abroad and life course stage upon return were the primary determinants of the type of self-employment –survivalist, stepping-stone, or prosperous.

Discussion

From our identification of three divergent self-employment types among return migrants, we extend three propositions, which we believe can inform and perhaps even direct future research on labor market mobility among return migrants. First, we propose that the quality of self-employment among return migrants, which includes job satisfaction, wages, innovation, and the presence of employees, depends primarily on the intersection of new skills acquired or enhanced while laboring abroad and talent and ambition developed across their lives, both at home and abroad. Scholars have long recognized that the acquisition of new skills can enable prospective entrepreneurs to identify investment opportunities, which can lead to the formation of new economic niches in their local economies (Jütting, Parlevliet, and Xenogiani 2008; Hagan, Hernández-León, and Demonsant 2015; Hagan and Wassink 2016). We found that for migrants who traveled abroad in search of higher wages and new opportunities, skills learned abroad expand their labor market repertoire, creating new opportunities for entrepreneurship upon return. Our results suggest that these new skills specifically relate to the quality of self-employment because they enable returnees to improve upon existing practices or identify new opportunities, offering an advantage in saturated labor markets. The experiences of the prosperous self-employed, especialy when compared to those of the temporary self-employed who treated self-employment as a stepping-stone to wage work, suggest an interaction between pre-migration attributes such as ambition and risk-taking and human capital formation abroad. By migrating from Mexico to the United States, these risk-taking risk-taking individuals were able to acquire and enhance skills that provided them with the confidence and knowledge to create their own businesses.

Second, we also found the quality of self-employment is influenced by the timing of migration and return. Labor migration late in individuals’ working careers is often associated with labor market marginalization; our interviews suggested that those respondents who migrated in their forties and fifties were responding to decades of low wages and blocked mobility. Others have found that employers’ preference for younger workers motivates the migration of older Mexicans to the United States (Hernández-León 2008). In other words, those who migrate later experienced labor market marginalization prior to migration, suggesting a greater likelihood of marginalization upon return. Indeed, it is reasonable to expect that individuals at later stages of their labor market careers who possess the skills to start prosperous businesses would do so rather than migrating to the United States.¹⁹ Upon return, limited levels of formal education and informal skills are compounded by employers’ preference for younger more physically fit and attractive workers.²⁰ Thus, old age pushes return migrants with limited skills into marginal self-employment as a survival strategy. Because their entry into self-employment is motivated by economic necessity rather than opportunity or ability, the quality of their investments tends to be low in that they do not facilitate social mobility or create jobs in the local labor market.

Third, our results suggest that the temporality of survivalist self-employment depends primarily upon age at return and skill formation while abroad. Employers’ preferences for younger workers made it easier for migrants who returned to Leon in their 20s and 30s to find salaried jobs, even if the work was low-paying in the informal sector. The challenges of older returnees were compounded by low levels of education due to cohort differences resulting from Mexico’s educational expansion in the 1980s and 1990s, which all but prevented older returnees from obtaining work in the formal sector of Leon’s economy.

Conclusion

As return migration has emerged as a major demographic phenomenon in the twenty-first century, a rapidly expanding literature considers the implications of this reverse flow for individual reintegration and local economic development in sending communities (Cobo, Giorguli, and Alba 2010; Hagan, Hernández-León, and Demonsant 2015; Hagan and Wassink 2016, 2016; Lindstrom 2013; Mezger Kveder and Flahaux 2012; Ruben, Van Houte, and Davids 2009; Wassink 2016; Sheehan and Riosmena 2013). In general, these studies find that migration experience is associated with greater odds of upward or downward occupational mobility and entry into self-employment (Cobo et al. 2010; Lindstrom 2013; Sheehan and Riosmena 2013). In other words, following return, most migrants experience some change in their labor market position. However, because research to date has relied on large cross-sectional surveys with limited measures of migration experience and self-employment, scholars have been limited in their ability to parse out the implications of these various mobility trajectories.

Drawing on the biographical narratives of current and former self-employed return migrants, and guided by the extant scholarship on work and self-employment in Latin America, we have presented a theoretical framework to understand return migrants’ transitions in and out of self-employment. Our account challenges existing models of return migration. In contrast to research that views migration as one piece of a larger economic strategy (Massey and Parrado 1998; Lindstrom and Lauster 2001), our results indicate that upward mobility among self-employed return migrants is driven by cumulative human capital skills learned on- and off-the-job at all stages of the migratory circuit. New skills create new and often unexpected opportunities. Although informal skill learning is not unique to the migration experience, our analysis suggests that work in foreign countries exposes international migrants to new social and technical approaches to work, which can lead to the identification of new niches upon return.

Our findings also challenged the view that self-employment among return migrants can be measured as a single category indicative of upward or downward mobility. We identified three distinct types of self-employment – survivalist, stepping-stone, and prosperous – which capture both the quality and temporality of returnees’ self-employment ventures. This typology, which draws on and extends the scholarship on work and self-employment in Latin America (Gindling and Newhouse 2014; Perry et al. 2007; Fajnzlber, Maloney, and Rojas 2006; Mandelman and Montes-Rojas 2009), can inform future studies of self-employment and mobility among return migrants in Mexico, Latin America, and beyond.

Our identification of alternative self-employment trajectories also yields insights into canonical economic theories of international migration. In contrast to scholarship informed by the NELM, which views migration as a strategy for accumulating capital to invest in new businesses (Lindstrom 1996; Lindstrom and Lauster 2001), the survivalist self-employed tended to migrate based upon a neoclassical rationale – seeking higher wages in response to stagnant economic mobility in Mexico’s informal sector. Their transitions – both into migration and subsequently into self-employment – reveal a pattern of labor market marginalization that pushes informal sector workers, especially those later in their careers with few marketable skills, to attempt multiple mobility strategies in response to persistently blocked mobility. For the survivalist self-employed and some of the stepping-stone self-employed, deciding to migrate and deciding to become self-employed is largely motivated by the low cost associated with relinquishing their marginal employment prospects in the informal sector.

By contrast, the prosperous self-employed more often traveled abroad early in their working careers in search of self-improvement, new experiences, or financial capital, motivations more consistent with the NELM framework. At the same time, their businesses were rarely the culmination of long-term strategies conceived of prior to migration. Instead, our interviews suggest that the prosperous businesses initiated upon return were more often motivated by the unanticipated accumulation of new social and technical skills on the job in the United States.

To the extent that migration to the United States is motivated by blocked mobility in Mexico, self-employment trajectories among return migrants hinge upon the accumulation of new skills in the United States which open doors upon return. Overall, our investigation of self-employment types suggests a complex relationship between international migration experience and the labor market mobility of return migrants that cannot be understood without taking into consideration migrants’ social and economic circumstances before, during, and after migration.

Finally, we recognize that this study, which focused on self-employed return migrants in a large industrial city, likely does not represent the experiences of all return migrants. We hope that our efforts will inspire similar considerations of self-employment in rural communities, which have historically been the primary recipients of return migrants (Garip 2016). Rural communities with high rates of poverty and correspondingly small service and retail sectors offer more limited opportunities for skill transfer and investment (Hagan, Hernández-León, and Demonsant 2015). Thus, migrants returning to these areas may struggle to benefit from remitted human capital. Additionally, although we identified mechanisms that affect self-employed workers’ access to salaried jobs, we did not consider the experiences of return migrants who never enter self-employment upon return. Future studies might consider whether the mechanisms that we identified – i.e., formal and informal human capital formation and life- course stage – influence the mobility trajectories of non-self-employed returnees in ways similar to those described here.

Biography

Joshua Wassink is a PhD candidate in the Department of Sociology at the University of North Carolina at Chapel Hill. His research interests include international migration, economic mobility, social stratification, and health. His work has appeared in Social Problems and the American Journal of Public Health.

Jacqueline Maria Hagan is a Kenan Professor of Sociology at the University of North Carolina at Chapel Hill. An international migration specialist, she is author of Deciding to be Legal (1994 Temple University Press), and award-winning books, Migration Miracle (Harvard University Press 2008), and (co-authored with Ruben Hernandez Leon and Jean Luc Demonsant) Skills of the Unskilled: Work and Mobility among Mexican Migrants (2015 University of California Press).

APPENDIX

Table A1.

Assessment of bias due to sample attrition

	Original sample	Longitudinal sample
Demographic characteristics
Male (%)	89%	81%
Age in years (median)	35	39
Married/civil union (%)	82%	83%
Human capital characteristics
Years of schooling (median)	9	6
Years of work experience (median)	20	21
Total number of jobs (median)	5	5
Ever directed/supervised others (%)	36%	40%
Migration characteristics
Age at first migration (median)	17	17
Transferred skills from U.S. to Mexico (%)	50%	54%
Number of jobs in the U.S. (median)	2	2
U.S. duration in months (median)	30	24
Remitted money for investment (%)	17%	22%
Undocumented (%)	90%	90%
Ever deported (%)	9%	15%