Vulnerable Women in a Thriving Country: An Analysis of Twenty-First-Century Domestic Workers in Peru and Recommendations for Future Research

Paid domestic workers represent a historically discriminated-against group comprising mostly ethnically marginalized, resource-poor migrant women. In twenty-first-century Peru, social and labor rights have not improved significantly for those in the sector despite more than a decade of sustained economic growth that took off in 2002 and lasted until 2014. Though the present national trend suggests that the absolute number of workers in this sector is dropping and that the tendency of workers to “live in” with their employers is reversing (trends that might signal improved working conditions for those in the sector), significant gaps still exist between the rights of domestic workers as compared to other workers. This article analyzes the intersectionality of gender, ethnicity and/or migration status, and class as one possible explanation for the continued vulnerability of paid domestic workers. We examine statistical information on the present situation in Peru, including a trends analysis of the National Household Survey from 2004 to 2013, and share the results of our qualitative research on the sector for the same period. We conclude with recommendations for future studies.


Introduction
In the decade between 2004 and 2013, Peru joined the league of countries that enjoy medium to high income (World Bank 2013), experiencing economic growth classified as both "pro middle class" and "pro poor" by the Inter-American Development Bank (Jaramillo and Zambrano 2013). This context included a new "emerging" middle class, improved education levels, and increasing participation by women in the workforce. Yet the same situation has been marked by the persistence of high levels of labor force informality and social inequality. Seventy-nine percent of all work continues to be informal (INEI 2014), one of the highest rates in Latin America. Additionally, the gaps between rich and poor are significant, with the richest 20 percent in the country amassing thirteen times more income than the poorest 20 percent (PNUD 2013). Given the country's dual commitment to continued growth and the implementation of policies that set the foundation for a more inclusive society, a critical challenge is to create the appropriate conditions for assuring greater socioeconomic equity.
There is work to be done on a number of fronts. In particular, Peru maintains a number of special labor regimes that diminish worker rights in order to cut costs. These include paid domestic workers. Nearly 100 percent informal, and amounting to 2.6 percent of the country's economically active population (INEI 2013), these workers merit particular attention, we argue here.
First, domestic work has confronted historic multilevel discrimination, as it has employed primarily poor, young, ethnically marginalized women. Second, the work is performed in private residences in which treatment of the worker, benefits, and discrimination are difficult to perceive and/or regulate by the appropriate authorities. Third, and related to the first two points, as compared to other labor regimes there is less knowledge about this workforce and its employers, and whether the latter comply with the basic rights afforded to these workers under the law. Fourth, while incomes have risen for all workers in Peru, there remains a comparatively large gap between the remunerations received by paid domestic workers and other workers of the economically active population (EAP). Finally, it is precisely because this job is performed by the most vulnerable women that an analysis of their collective situation may serve as a barometer to measure the country's progress toward achieving socioeconomic equity (Blofield and Martinez Franzoni 2014;Jokela 2015;Pérez and Llanos 2015a). 1 The sociopolitical context also offers a timely backdrop against which to review the status of these workers. With the passage of the International Labor Organization's (ILO's) Convention 189 (C189) and its Recommendation 201 in 2011, there exists an international framework for the constitution of decent work for this labor force. 2 While the Latin American region is now the part of world with the highest number of ratifications, 3 Peru has yet to adopt the convention, or improve its legislation. 4 The country 's Domestic Worker Law No. 27986 (2003) represents an important step toward recognizing this sector as part of the workforce, not mere "service providers" or "servants"; however, it provides only partial rights and thus needs to be improved and strengthened so that domestic workers might enjoy the same entitlements and protections as do other workers.
To date, the Ministry of Labor and Employment Promotion (MTPE) and the Ministry of Women and Vulnerable Populations (MIMP) have taken some steps to examine the viability of implementing C189. Specifically, the MTPE has led a multisector working group, including members of civil society, to monitor the development of an action plan intended to promote the enforcement of domestic workers' labor rights, though the working group appears to have little traction, if any. Moreover, two parliamentarians have called for the ratification of C189, but these efforts have not progressed. Thus, on the legislative front scant attention has been devoted to domestic worker rights.
As to the work of the labor unions on this issue, two large domestic workers unions exist-the Domestic Workers Union for the Lima Region (Sindicato de Trabajadoras y Trabajadores del Hogar de la Región Lima, SINTTRAHOL), affiliated with the General Confederation of Workers of Peru (Confederación General de Trabajadores del Perú, CGTP); and the National Union of Domestic Workers (Sindicato Nacional de Trabajadoras del Hogar, SINTRAHOGARP), associated with the United Confederation of Workers of Peru (Central Unitaria de Trabajadores del Perú, CUT). Although unions participated in the drafting of C189 in Geneva prior to its international adoption by the ILO, interviews with union members and other civil society activists suggest that a key challenge for the syndicates-and, thus, for improving domestic worker rights-is that they do not have an appropriate counterpart, an employers' union, with which they can collectively bargain. Moreover, relations between the two existing unions are strained. This, combined with the present impossibility of negotiating with an appropriate counterpart and the comparatively more powerful national 1 Based on a macroeconomic cross-country statistical analysis of seventy-four countries, Jokela (2015) suggests that income inequality is a key factor in determining the proportion of paid domestic workers in the labor force of a country. 2  business guild that opposes any increases in wages for these workers, is an impediment to their success as compared to other unions of the region like those in Chile or Brazil, for example (Pérez 2015). 5 In sharp contrast to Peru, Uruguay was the first country of the world to ratify C189 and has made great strides in creating parity between the social and labor rights of paid domestic workers and those of other members of the labor force. Most recently, it has inaugurated a National Care System in which caregiversamong them, paid domestic workers-are considered under the country's social protection policies. 6 Additionally, one of the most recent ratifications in Latin America took place in neighboring Chile in 2015, following a national legislative process that sought to improve the rights of domestic workers by increasing the minimum wage, reducing the number of hours worked, sanctioning and requiring vacation time and weekly rest, and written contracts, among other rights (Pérez 2015). Unlike Chile and Uruguay, for example, Brazil has not ratified C189, but the country's legislature passed a constitutional amendment in 2013 that sought to establish parity between the rights of the domestic worker and others and ensure compliance with these rights among employers. This includes employers' public registration of their domestic workers; a mandated eight-hour workday, minimum wage, social security and health benefits, and severance pay. 7 Even though employer compliance with worker rights continues to be a challenge for Brazil and other countries, the political will and active efforts to expand rights through national legislation are nonetheless notable. 8 Not least important has been the role of civil society organizations, unions, and the empowerment of domestic workers through their association with these organizations (Bernardino-Costa 2014).
The present situation of the sector in Peru reveals that the economic boom in the first decade of the twenty-first century was not enjoyed by paid domestic workers because of the undervaluation of their work and because the social protections that would have limited their exploitation were only minimally present. Instead, this sector subsidizes the state by providing cheap labor, thereby filling gaps in services that the state does not provide (Blofield 2012). Furthermore, we suggest that economic growth has had a positive effect for many in Peru precisely because, among other factors, these highly vulnerable and easily exploitable workers have helped to subsidize economic and social mobility for other social classes. As Schellekens and van der Schoot (1989, 303) suggested in 1989: "The existence of a cheap labor force frees the upper and middle classes from domestic work; they do not need to make costly collective provisions and can use their savings in other ways." As we consider the status of this worker in a context of economic growth, legislative changes, and the country's longest period of uninterrupted democracy, we observe that the situation has not changed significantly for this sector in the past nearly three decades. While incomes have risen for all, why is there such a comparatively large gap between the remuneration and social rights of these workers and others? While we know from the international and national literature that their profile is that of low-income, racially and/or ethnic minority migrant women with limited education (Anderson 2004;Blofield 2012;Chaney and Castro 1989;Pérez and Llanos 2015a;Rutté García 1973), is there one condition that is more defining than the others in determining domestic workers' lack of access to decent wages, quality of employment, and overall social protections? Or is there a confluence of conditions at play, as suggested by research on intersectionality? Can an analysis of the Peruvian case, viewed through this perspective, offer us new insights?
Based on our examination of the extant literature, statistical data available, and the results of nascent qualitative research (Pérez and Llanos 2015b), we hypothesize that at least one of the reasons that domestic workers in Peru remain vulnerable and easily exploitable by others is because they are discriminated against and disempowered on the basis of the intersectionality of at least three conditions: gender, ethnic origin and/or migration status, and class. While their generally low educational attainment also plays an important role, as we will show below, in the very particular case of paid domestic workers in Peru it may not be the main reason for why they remain vulnerable in this employ. Rather, they maintain low educational attainment and enjoy little vertical labor mobility precisely because of the confluence of the aforementioned conditions which, in turn, serve to position the women of this sector for a life of service as domestic workers with 5 Situación organizativa de las trabajadoras remuneradas del hogar en la región andina y procesos de incidencia política para ratificación del Convenio (

A Brief History of the Sector
Domestic service in Latin America has existed since colonial times. The "patriarchal household was the central unit of social control" in this period and domestic service was one of the few areas permitted to women (Kuznesof 1989, 18). At first, depending on the place and circumstance, women of all races and ethnicities participated in the work; however, writing on sixteenth-century Peru, Elinor Burkett notes that the most common and least-well-paid servants were indigenous women. As men went to work in the mines and women were left alone, the option of traveling to the cities from their traditional communities to work in Spanish homes represented a more economically secure option, as they were still responsible for paying tribute to the Spanish crown. Once installed in these homes, they were treated like slaves and "prevented from leaving or marrying" (Burkett 1978, 111, quoted in Kuznesof 1989. The colonial history is relevant for understanding the composition of today's paid domestic service sector. First, the colonial system-different from that experienced in Europe of the same period-established domestic service as primarily a woman's job. Though in the early period of colonization there were men performing some domestic tasks, they were primarily former slaves or indigenous. Second, given the colonialists' racial/ethnic stratification, by the eighteenth century the work was predominantly feminized and indigenized. 9 By the nineteenth and twentieth centuries, domestic service became "an almost entirely female and lower-class occupation" (Kuznesof 1989, 31).
Chaney and García Castro's 1989 compilation of research on domestic work in Latin America examined the status of worker rights and their organization at that time, noting organization efforts, particularly in Brazil, Chile, and Uruguay. Given that the legislation giving way to equal rights was not in effect in most of the region at the time of Chaney and García Castro's publication, there was some hope that through workers' affiliations with unions and enhanced class consciousness, their labor and social conditions might improve. In the case of Peru, Smith (1989) expressed some optimism on the possibility of vertical labor mobility for former domestic workers, though her sample was small and the full cohort was not followed over time.
Today, though there have been some important victories in the struggle to "visibilize" domestic work internationally through C189 as well as through advances in national legislation, for those who remain in the employ there are varying levels of vulnerability related to the enforcement of the laws (Bernardino-Costa 2014; Blofield 2012). In Peru in particular, the basic precariousness of domestic work continues on a number of levels. It remains largely unorganized, unregulated, and discriminated, as Kuznesof (1989, 32) described domestic work in Latin America nearly thirty years ago: a "dead end, low status, nonregulated and often hostile condition of exploitation." Considering Kuznesof's original observation, the demand for domestic service today continues to be especially strengthened by the extraordinary disparity between income levels (Kuznesof 1989, 32;Jokela 2015). Moreover, it is still stratified by gender, race or ethnicity, and class.

The Theory
Contrary to human capital theory, which argues that there exists a correlation between years of formal education of individuals and their incomes (Checchi 2006;Mincer 1958), we find that in Peru this relationship seems to be mediated by occupational group membership. As Figure 1 shows, domestic workers with completed secondary studies still receive an average monthly income that is less than all other occupations of the EAP. 10 These descriptive findings are consistent with research on the existence of occupational segregation patterns in domestic workers in general. For example, Lautier (2003) and Tizziani (2011) point out how occupational mobility tends to be horizontal and occurs between precarious jobs with similar characteristics as a result of the overlay of multiple exclusions that limit possibilities for upward social mobility for this sector. Browne and Misra (2003) argue that, because most of these workers are women, they are not perceived as the primary family breadwinner, and therefore employers may justify paying them less. Moreover, they find that these wages are also justified by the limited opportunities to which employers assume these women have access. Cutuli and Pérez (2011) find that educational credentials go unnoticed by their employers, particularly where the intersection of at least two conditions is present: being a woman and a migrant. Likewise, according to Pratt (1999), stigmatized discourse based on race and/or ethnicity contributes to employers' unwillingness to recognize workers' educational credentials.
Considering the conclusions of the aforementioned research together with work that has utilized the concept of intersectionality (Bernardino-Costa 2015; Crenshaw 1989Crenshaw , 1991Gorbán and Tizziani 2014;Romero 1992;Young 2001) may serve us here in two aspects. First, it can provide a potential explanation for why it is that regardless of their educational status and the comparatively improved economic situation of other workers of the EAP, those employed in domestic work continue to be a highly vulnerable group. And, second, considering the intersectionality of gender, race/ethnicity, and class in this case can provide a framework through which to understand the conditions that serve to marginalize this particular worker in Peru. Crenshaw first coined the term "intersectionality" in 1989 to distinguish the uniqueness and multidimensionality of the experience of black women in the United States. She has since argued how the intersectionality of race, class, sexuality, and gender are responsible for oppression and disempowerment (Crenshaw 2006), noting that "racialized women frequently find themselves in a position where racism or xenophobia, class and gender meet each other. Consequently, they are liable to be hurt by the intense flux on these roads" (Crenshaw 2002, 177).
Some studies on paid domestic work have employed intersectionality as an analytical instrument. Mary Romero's Maid in the USA (1992), for example, documents, among other issues, the discriminatory treatment and/or stereotyping experienced by Chicana domestic workers in the United States due to their Mexican origin. At the same time, she describes how they also experienced class-based discrimination based on their low-income status in contrast to other women, namely their employers. The more recent work of Bernardino-Costa (2014) showcases the intersectionality of race, class, and gender, examining these in the context of two different power relationships in Brazil: one that highlights how the intersectionality of race, class, and gender has played out in the households where domestic workers are employed, serving to disempower and oppress them; the other in the empowering and mobilizing role of the same intersection of conditions in the context of their political organization. Regarding the intersectionality of these conditions in the context of their place of employment, Bernardino-Costa argues that class does not operate in isolation but rather in combination with race and gender: "If it is true that child domestic labor is an option for poor girls in Brazilian society, it is more so for girls who are poor and black" (2014,75). Their gender, combined with race and class, also makes them highly susceptible to sexual harassment. Also noteworthy is that the physical space and context of the intersection of these conditions can either create a sense of disempowerment-as is the case of the work conducted in private homes-or a sense of empowerment through the camaraderie of the collective struggle that union membership provides. As noted elsewhere, most of the research agrees that the historical origins of domestic work in Latin America and the Caribbean and the fact that it is performed mostly by low-income women of color, in many cases migrants, serves to diminish the value ascribed to the profession. Likewise, we argue that in Peru it is the intersection of gender, class, and ethnicity and/or migrant status of domestic workers (i.e., being a woman who is materially poor and of indigenous and/or rural or periurban origin), juxtaposed against their employers and the particular context in which they work (mostly alone in private homes without regular state monitoring) that assures their continued vulnerability. In this light, we agree with Brigitte Young's assertion that as domestic services continue to be privately purchased as a solution to household reproduction, it "creates at the household level a new ethnically defined female underclass that lacks political rights and legal rights" (Young 2001, 322-23). Furthermore, as she has noted, "We witness a new international division of labour between women of different ethnicity, class, generation, and citizenship" (Young 2001, 324). Along similar lines, through their research on domestic workers in Buenos Aires, Gorbán and Tizziani (2014, 54-55) have made the case that this work is "inferiorized" suggesting that through the intersectionality of the diverse conditions that characterize this sector, employers "construct a stereotype of social inferiority for domestic workers through which they legitimize their dominant position in the labor relationship." This is not unlike what occurs in Peru, as the intersectionality of ethnicity and/or migration status, gender, and class in the context of domestic work in the homes of their employers continue to serve as a disempowering force for the women in this employ. Despite that the majority has completed some level of secondary education, this condition has neither improved wages nor led to more rights. Rather, as the national statistical data reveals-and our interviews confirm-those who remain in paid domestic work remain at "the bottom of the scale" (Pérez andLlanos 2015a, 2015b).

A Review of the Statistical Data
Based on a 2012 ILO study on the rights of domestic workers in Peru, Table 1 illustrates the overall undervaluation of this type of work. Our analysis of data from the Peru National Household Survey, or Encuesta Nacional de Hogares (ENAHO), for the years 2004-2013, 11 which supports these findings, additionally compares domestic work against other occupations 12 and provides trends data that contrast distributions for the total of the EAP against domestic workers for the year 2013. 13 Under "Discussion" below we analyze these statistical trends together with the findings resulting from semistructured interviews conducted in early 2015 with twenty women employed in paid domestic work at some point in the years between 2004 to 2015 (Pérez and Llanos 2015b, 11-17). 14

Fewer and older, but still women and adolescents
As illustrated by Figure 2, in absolute terms the growth of the EAP has been accompanied by a gradual decline in paid domestic work.
As Figure 3 shows, the decline in domestic employment has also occurred in relative terms, descending from 3.6 percent of the EAP in 2004 to 2.6 percent in 2013.
In terms of age distribution (Table 2, Figure 4), we can see that domestic workers include an important proportion of adolescents between fourteen and eighteen years of age (13.7 percent). However, much like the total of the EAP, the majority of workers are between the ages of thirty-six and sixty-five years.
Despite the above, we can see in Figure 5 that the average age of domestic workers has increased rapidly over the past ten years, coming close to that of other occupations. These data raise a number of questions.   Is the aging of this sector a result of a decreased supply or a decreased demand? What are the underlying causes of this change? Who are the people that tend to remain in this kind of employment? Are they more or less vulnerable? Why, or why not?
There are a few possible explanations for this process of relative aging. It may indicate that over time it is less common to recruit younger domestic workers, or that an older population is joining the sector. It is also possible that younger workers are experiencing some kind of occupational mobility, leaving only those that have been in this employment for many years.   Analyzing the statistical data together with our interviews (Pérez and Llanos 2015b), we suggest that another possibility is that workers cycle in and out of the sector. While many women enter domestic work at an early age (Figures 5 and 6), many of the life stories that we have gathered suggest that their experience with this work is not linear. We found in a number of cases that workers may have been employed in this work for a period of time and then later abandoned it, only to return to it years later. We explain this in more detail below in the discussion section.

Mostly migrant, poorly educated, separated or single women and adolescents
A comparison between the gender distribution of the EAP and domestic workers (Table 3, Figure 7) confirms the highly feminized nature of paid domestic work in Peru. This is consistent with the regional trend reported by the Economic Commission for Latin America and the Caribbean (Comisión Económica para América Latina y el Caribe, CEPAL 2009CEPAL , 2012CEPAL , 2013.   Another aspect that deserves attention as we consider the profile of domestic workers in Peru is migration. Table 4 and Figure 8 estimate internal migration via a comparison of the province of birth against the province in which the worker resides. The comparison between the total EAP and domestic workers reveals that the latter group is composed in its majority (51.4 percent) of migrants. By contrast, migrant members of the EAP constitute 34.9 percent of that group.
Comparing the educational attainment of the EAP to that of domestic workers by level of education (Table 5, Figure 9) shows that the majority of these workers lack higher education (university, technical institute, or trade school). More than 80 percent of this group is composed of people who have only attained primary or secondary education as compared to 66 percent of the total occupied population.   As shown in Figure 10, the average number of years of formal education experienced by other occupations is superior to the average for domestic workers. The gap appears to grow slightly with the passage of years due to the stagnation of domestic workers' education levels and the comparatively improved education of the labor force in general.
Finally, as illustrated in Table 6 and Figure 11, as compared to the EAP, domestic workers tend to be single or separated in greater proportion. This group also presents a lower proportion of married people (15.4 percent).

Poor remuneration and social protections
As Figure 12 illustrates, real monthly income has increased substantially for the whole of the labor force. 15 However, there is a statistically significant difference in the average income of domestic workers compared to the rest of the EAP for every year reviewed. Table 7 contrasts the modes of contracting for the EAP against those for domestic workers. As with the high incidence of their presence in the informal sector, we can observe that an overwhelming majority (99.3 percent) of domestic workers are employed without a contract. 15 In order to calculate average real employment income we consider income that is derived from primary and secondary occupations.  Furthermore, Table 8 contrasts how the EAP and domestic workers are paid. While in the labor force in general there exists great heterogeneity, and payment in kind is rather low (19 percent), for domestic workers, payment via wages (97 percent) and payment in kind (80 percent) are the predominant forms of remuneration. We should note that these percentages do not total 100 percent as these categories are not mutually exclusive: a worker may receive employment income through different forms of payment.   As shown in Figure 13, there appears to be a general improvement in the rates of workers that are affiliated to pensions, both at the level of domestic employment as with other occupations. However, we should point out that the proportion of people not affiliated to the pension system is much higher for domestic workers.

Discussion
As we examine the aging of the work force together with our interview data we also find that a number of women who left the work actually go back to it. The most common reason we found is based on motherhood and the tendency in the period of pregnancy and early maternity to be economically inactive, sometimes for years after having children. Sarita (b. 1989, Cajamarca), for example, returned to work as a live-in worker shortly after her husband died in a work-related accident. At the time she had a three-year-old toddler and an infant whom she later left with her mother in order to seek work in an urban center. Margarita (b. 1968, Puno) thought that, once married, her husband would take care of her. Indeed, he did for some time until he lost his job and eventually abandoned the family. As Margarita shared, domestic work is her only skill and, with an absent husband, it is up to her to ensure a university education for her daughter. There is also Susana (b. 1976, Lima) who began this work at the age of ten and has cycled in and out of domestic work for most of her life, with a number of microenterprises on the side, but now seeks to ensure a steady income for her daughter's future studies in economics and international business. Additionally, younger women whose origins are in the rural world tend to see this kind of job as the first step in a social-mobility ladder and, in some cases, as a way to escape from unstable family lives and/or domestic violence. This was a clear pattern in almost all of the interviews. The lack of educational credentials, combined with the fact that the social and ethnic origins of domestic workers can make them subject to many forms of discrimination, make it rational for them to seek to migrate to Lima or other big cities and, while making a living as a domestic worker, to try to attain technical credentials in other gendered occupations such as nursing, preschool teaching, pharmacy, and/or cosmetology.
What we found, however, with the exception of one person interviewed who was studying pharmacology, was that plans for future study or other jobs did not materialize for the cohort. Instead, most common was that at some point the women interviewed had either engaged in a marriage or partnership and/or had children. Thus, the decisions to leave and later return to domestic work are based on at least two important factors. First, maternity leave policies for the workers we interviewed were nonexistent and/or unenforceable, so in a number of cases it was suggested that they left essentially so that they could care for their children. Second, gender ideology, particularly in the Peruvian working class, emphasizes that the father of one's child will take care of the family as well as the primacy of the woman's role in child-rearing during the early phases of development. This was reflected in phrases like "No, no trabajaba . . . mi hijo estaba chiquito." (No, I didn't work because my son was little.) However, as their partners abandoned them and their children began to grow, the imperative for income to support the family and, hopefully, to pay for higher education for their offspring made it necessary for the full-time mothers to return to domestic work as a means to secure steady income.
We also find that nearly 100 percent of paid domestic workers in Peru are women and more than half are internal migrants. Examined together with ethnicity and class, the simultaneous existence of these conditions paints an interesting portrait. According to our informants, there are two primary drivers for migration: economic mobility and/or education, and escape from domestic violence. As seen through the lens of intersectionality, poor, ethnically marginalized, and/or rural women migrate to larger cities (mostly Lima) in order to gain the education that was not available to them either for institutional reasons (the absence of a school nearby) or family structure or cultural reasons (their families did not permit them to go to school, and/or family life was abusive). In both cases, their discrimination and ill treatment started in their homes, based on their gender and their situation of poverty.
The interviews reveal a complex reality in which a clear pattern indicates that in order to accomplish even a low level of educational attainment great barriers must be confronted, including the poor quality-and sometimes nonexistence-of basic education in rural and marginal areas. For young women, this situation is compounded by a strong gender ideology that still favors the educational advancement of boys over girls. Modesta (b. 1989, Puno), for example, shared with us that her father told her, "Las niñas no van al colegio." (Girls don't go to school.) Moreover, the explicit resistance of some employers to allowing domestic workers to continue studying is still a barrier to higher education and social and labor mobility. Elizabeth (b. 1996, Huaraz) arrived in Lima in order to send money back home to support her two-year-old son. She also hoped to study and managed to finish high school while working for a family of four as a live-in worker. However, when she decided to pursue higher education at an institute for design, her employers refused to provide her with free evenings or a day off to accomplish this goal. As she told us, "A veces cuando quiero salir, no me dejan o me llevan a la casa de playa . . . siempre hay algo que hacer." (Sometimes, when I want to go out, they don't let me, or they take me to their beach house . . . there is always something that needs to be done.) Regardless of all these structural constraints some workers still manage to attain a technical degree and/or to find other employment outside of domestic work. However, in practical terms this has meant that they have moved on to undervalued, underpaid, and gendered occupations. As a result, some pointed out that after experiences in working-class jobs or as having been self-employed, they thought that paid domestic work was the best option they had. In particular, Modesta (b. 1989, Puno) shared that after completing secondary education and becoming a chef, "No vale la pena por la plata." ("It's not worth it, for the money they pay.") Analyzing this together with the research of Bastidas (2012), which shows that 60 percent of domestic workers are responsible for dependents, this also supports the idea that a single head of household and/or young migrant, with family in another part of the country that is dependent on her, may take up this work-even if only with the idea that it is temporary-for lack of better options and in order to gain access to regular wages.
Finally, as illustrated in Figure 12, there is a statistically significant difference between the average income of domestic workers and the rest of EAP. Even when we compare domestic workers with workers in other unskilled occupations of the EAP who have less educational attainment (Figure 1), domestic workers still earn less and are more vulnerable.
In summary, we find that the historical and multilevel discrimination suffered by those that typically swell the ranks of domestic employment characterizes domestic work as unattractive and at the bottom of the scale of social prestige. This may act as a chilling effect on some young women seeking to enter this employ. However, as we have noted previously, the profile of the women who enter domestic work may equally serve to block their labor mobility in other jobs. Ultimately, the intersectionality of gender, ethnicity and/or migration, and social class have served to discriminate against the domestic worker, ensuring vulnerability through limited labor and social rights for those who remain in this sector.

Conclusions and Recommendations
In present-day Peru, paid domestic work continues to be undervalued and discriminated against due to the intersectionality of the above-mentioned conditions. As a result, notable disparities exist between these and other workers of the EAP in terms of remuneration and social and labor rights (Anderson 2007(Anderson , 2009Blofield 2009Blofield , 2012Pérez and Llanos 2015b;CEPAL 2009CEPAL , 2012CEPAL , 2013Stefoni 2009). As it is work that is performed in the informal sector, within a private residence and without any effective mechanism for state regulation, it becomes extraordinarily difficult for these workers to enjoy even the limited rights that the law presently affords them.
As we stated at the outset of this article, we contend that the condition of vulnerability results from the discrimination produced as a result of the intersectionality of gender, ethnicity and/or migration status, and class. Put another way, a recent migrant to Lima who is a young woman of rural and indigenous origin is an easy target for those who would exploit her, having little, if any, knowledge of her legal rights, poor knowledge of local customs, possible lack of proper documentation, working alone in a household where no one is monitoring, and exposed to discrimination as well as the possibility of physical and sexual aggression. Taking into account the history of this employ and the labor segmentation that it established, in Peru the quality of the working conditions for many in this sector depends nearly entirely on the benevolence of the employer.
The situation cannot improve unless the working conditions are corrected. Given that services are provided in private homes, a state-enabled monitoring system would provide some protection of worker rights. Furthermore, as paid domestic workers establish relationships with other workers, union organizations, and civil society efforts, a challenge will be to ascribe a higher value to the work itself. Perhaps, as in the case of Brazil, employing the very intersectionality of the conditions that has served to marginalize them historically would be one important step. This means mobilizing in the context of union membership and along the lines of gender, ethnicity, and class in order to pave the way for a heightened sense of collective agency and power (Bernardino-Costa 2014;Blofield 2009Blofield , 2012. To some degree this is already occurring, but there is still a long road to travel in ensuring that the larger unions with which each of the worker syndicates described here are associated recognize that the struggle must also include gender and ethnic concerns. We must also note that, despite the economic growth experienced in Peru for the period analyzed here, the precariousness of this employment has persisted. As we have seen, the irony of the situation is that, for many, this work may represent the only viable option for a steady income. As the economy has slowed down, what implications will this have on those who remain in this line of work? The reality for most workers across the country, and even in Lima, is that their well-being in this employ is a matter of luck. We suggest here a few possible directions for future research on the subject. First, we concur with the valuable observation made by the ILO (Fuertes, Rodríguez, and Casali 2013) regarding the need for more information about employers in this sector. As domestic workers and their opportunities and/or preferences change, it would be important to know how the role of present employers has also changed or remained the same. Moreover, given that the defining factor of respect for workers' rights is nearly completely dependent on the employer, it would be important to understand to what degree improved economic conditions such as those experienced in the first part of the present century in Peru might translate-or not-into improved working conditions for domestic workers. Our analysis suggests that, at least for the period under study, economic growth has meant quite little for improved labor and social rights. However, it would be important to pursue this line of research in more detail, asking: Who continues to employ domestic workers in twenty-first-century Peru, and why? Under what circumstances and in what conditions do they work? What are the perspectives of employers regarding the rights of workers? Are there possible changes in public policy that would incentivize the formalization of this work and better compliance with the laws? Or are there more realistic possibilities for changes outside of state-centric efforts to improve working conditions? For example, would universal family leave policies open to all workers, whether informal or not, make a difference? Are there incentives that can be offered to employers to ensure that workers be encouraged and allowed to pursue higher education?
Second, we have also noted here the trend toward "independent" employment, performed by those who do not "live in" with their employers. What does this trend imply? Our findings may be a sign of positive socioeconomic and cultural changes that have made occupational mobility possible for domestic workers. However, as we have noted, it may also mean a more precarious situation for those who remain in the sector or for those who work independently. It is possible that this trend implies that rights are even further diminished.
Likewise, if growing numbers of domestic workers are transitioning to other jobs, where are they going, and why? As our qualitative research has suggested, it appears that workers may return to the sector following previous horizontal moves to poorly paid and/or equally exploitative positions. Are these moves akin to the "search for respect" that Bourgois (1995) outlined in his work devoted to understanding the lives and rationale behind the actions of street-level drug traders in East Harlem? As suggested previously by some of the literature on segmented labor markets (Browne and Misra 2003;Cutuli and Pérez 2011;Lautier 2003;Pratt 1999;Tizziani 2011), it is possible that additional years of education for domestic workers do not much improve the probability of obtaining better remuneration or social inclusion. It would be therefore important to study to what degree transitions to other jobs are possible and whether improvements in legislation and specialized education and/or capacity building might help to make vertical labor changes viable.
Third, according to Anderson (2007Anderson ( , 2009Anderson ( , 2012, a complex issue related to the study of this sector is the work's role in improving the economic status of entire families. It is important to document whether, and how, present employment in this sector effectively helps families make their way out of poverty. The corollary question to this is to what degree such a model only continues to reinforce exploitation of this worker by both her family and her employer. In the final analysis, does this worker subsidize both her employer and her family as a result of her services? Additionally, given that many families send their daughters to work in the homes of relatives or acquaintances, there exists a hidden component of child labor that requires visibilization regardless of how these children self-identify or are characterized by their families. Statistical data can only tell us who self-classifies as a domestic worker, but some of the subtleties of what actually occurs might be drawn out more precisely via qualitative methods.
Finally, should Peru decide to ratify C189, we expect that it would have implications for national legislation, labor relations, and social relations. Research that tracks the implementation of changes in national policies due to the ratification of C189 would be important in terms of analyzing what bearing, if any, these shifts imply for the improved situation of paid domestic workers as well as other special labor regimes in Peru.
In summary, given the intersection of gender, ethnicity and/or migration status, and class as key conditions in the profile of the paid domestic worker, an understanding of how labor conditions are or are not changing may help us to understand where some of the most vulnerable women in Peru stand today. We suggest that this research is relevant not only for one country but for others on similar development paths, including countries with high growth rates coupled with persistently elevated degrees of gender, ethnic and/or racial, and class inequality.