Migrant Women’s Economic Success in Russia: Objective Reality and Subjective Assessment

Abstract

This study contributes to the growing body of literature on the outcomes of labor migration by focusing on the effects of migrant legal status on the economic and perceptual measures of migration success. To study the effects of legal status, we use a sample of Central Asian migrant women who work in Russia and of their native counterparts who occupy the same positions on the labor market. Similar to the studies in the developed settings, we find that a temporary legal status is associated with an earnings penalty and that permanent legal status corrects this earning disparity. We also find that both temporary and permanent migrant status is positively associated with perceptions of pay inequality but that, irrespective of these perceptions, both types of migrants are more likely to be satisfied with their jobs than natives. We interpret these findings within the legal and social context of migrant economic incorporation in Russia and relate them to the findings from other migrant-receiving settings.

Introduction

Perceived economic opportunities in the receiving country, relative to those available in the country of origin, are at the core of many explanations of international labor migration (e.g., Boswell 2008; Grogger and Hanson 2011; Massey et al 1993). After migration, however, these economic opportunities may not fully materialize or may not translate into labor market success in the host society. Migrant workers in similar host environments display a wide variety of economic outcomes. Likewise, migrants’ assessments of their economic performance differ. In this study, we focus on the determinants of migrants’ economic success, as measured by earnings, and on the subjective assessments of this success, as measured by the perceptions of wage equality and by job satisfaction. Each of these three indicators represents a separate dimension of migrant incorporation into the host society: earnings are the most tangible outcome of labor market participation; the perception of earnings equality is a softer, subjective measure of labor market status or one’s position relative to other workers; and, finally, job satisfaction is an assessment of one’s work environment relative to expectations and aspirations.

For the potential economic opportunities and expectations of economic benefits to translate into actual jobs and earnings, several forces should come into play. These involve individual human capital such as education, skills, language fluency, and past labor market experience (Chiswick 1979; Chiquiar and Hanson 2005; Dustmann 1993; Sharaf 2013) as well as forces of social nature such as personal networks of relationships with other migrants and natives (Amuedo-Dorantes and Mundra 2007; Granovetter 1973; Mouw 2003). Migrants’ group-level characteristics, such as shared ethno-cultural backgrounds and collective experiences of arrival and settlement in the host society, may also be important for labor market outcomes (De Jong 2000; De Jong & Madamba 2001; Kalter and Kogan 2014). The effects of these individual and social forces may be coupled with the effects of institutional factors that shape the structure of labor market opportunities available to migrants in the host setting. Legal status is often singled out as one of such factors (Amuedo-Dorantes and Mundra 2007; Kossoudji and Cobb-Clark 2002; Rivera-Batiz 1999). Literature on the effects of legal status, however, has been constrained by limited availability and reliability of data.

In this study, we focus on legal status as a relatively less researched, yet critical axis of variation in migration success. We examine the effects of legal status on earnings, perception of pay equality, and job satisfaction for working migrant women from three Central Asian countries—Kyrgyzstan, Tajikistan, and Uzbekistan—and their native counterparts in urban Russia. The women-focused approach, while limiting the generalizability of the findings, is warranted since Central Asian men and women migrants in Russia typically concentrate in different sectors of the economy. In addition, we are mindful of the vast evidence that migrant labor force experiences in the host societies are highly gendered (e.g., Curran and Saguy 2001; Curran and Rivero-Fuentes 2003; Curran et al. 2005; Davis and Winters 2001; De Jong 2000; Donato et al. 2006; Shauman and Noonan 2007). Importantly, our study deals with a host setting that has long been neglected in migration research even though it is second only to the United States in the number of international migrants (United Nations 2015). Our primary interest is the association of migrants’ legal status with their labor market outcomes. Human capital and social capital variables, as well as migrant group-level background, complement the analysis and offer a test of the established theories in the Russian context.

The study proceeds as follows. The next section provides a brief background on international labor migration in Russia over the recent decades. The following section presents the conceptual framework and hypotheses. Next, we describe the data and models, present the results, and evaluate them in light of the hypotheses. The Discussion section highlights the substantively important findings of the analysis and relates them to the existing broader scholarship on immigrants’ economic incorporation.

Background: Labor Migration from Central Asia to Russia

After the dissolution of the Soviet Union in 1991, the Russian Federation has been a magnet for migrants from the former Soviet republics, especially from Central Asia (Abashin 2014). Migration from three impoverished Central Asian nations of Kyrgyzstan, Tajikistan, and Uzbekistan has been particularly massive, fueled by vast differences in the levels of economic prosperity between these countries and Russia.² Moscow, Russia’s capital and one of the world’s largest cities, has attracted a disproportionate share of migrants, but over the past decade-and-a-half other large Russian cities have also been drawing large numbers of migrants. Migrant flows used to be predominantly male but now include a large and growing share of women (Khusenova 2013; Laruelle 2010; Tyuryukanova 2011).

The citizens of Kyrgyzstan, Tajikistan, and Uzbekistan do not need a visa to enter Russia and can stay in the country for three months without a work or residence permit (FMS, 2014). They can adjust their legal status after arriving in Russia. However, it is common for Central Asian migrants, men and women alike, to remain to some degree undocumented due to administrative hurdles and high financial costs of obtaining and maintaining a regular legal status (Reeves 2013a). Frequent changes in the Russian immigration legislation and labor market regulations further complicate migrants’ legal and work trajectories. In theory, the legalization path involves three main steps—temporary residence, permanent residence, and naturalization. A temporary migrant may stay in Russia for two years and can apply for permanent residence after a year of being a legal temporary migrant. After a five-year residence in Russia, a migrant may apply for Russian citizenship. Certain groups of migrants may obtain Russian citizenship in an expedited fashion due to the special bilateral treaties between Russia and their country of origin. Of the three migrant groups included in our study, at the time of the survey the expedited process applied to the nationals of Kyrgyzstan.³

Lacking in some aspects of legalization, many Central Asians are at a disadvantage on the labor market compared to their Russian counterparts. Legal vulnerability may dampen not only their employment opportunities but also their expectations for wages, working conditions, and job advancement. Even a fully legalized status and acquisition of citizenship may not shield a migrant from a differential treatment on the labor market. Central Asian migrants, who are both racially and culturally distinct from the Russian ethnic majority, face widespread formal and informal discrimination by employers, law enforcing officials, and ordinary ethnic Russians (e.g., Agadjanian et al. 2017; Galliamov 2005; Mukomel 2013; Reeves 2013a, 2013b; Regamey 2010).

Migrants from Kyrgyzstan, Tajikistan, and Uzbekistan have considerable commonalities, especially when compared to the host population, but there are important ethno-cultural distinctions among them. Kyrgyz and Uzbeks speak similar Turkic languages whereas the Tajik language has Iranian roots. Uzbeks and Tajiks represent traditionally sedentary populations of Central Asia whereas Kyrgyz are a nomadic group with historically recent sedentarization at the turn of the 20^th century (Zotova and Agadjanian, 2014). Although all three groups are Muslim, the influence of Islam is generally stronger among Uzbeks and, especially, Tajiks, with corresponding implications for the position and autonomy of women in society (Agadjanian et al. 2014).

Declining economies and political instability in Central Asian countries in the wake of the Soviet collapse were initial triggers of migration to the Russian Federation. Of the three ethnic groups, Tajiks have the longest history of migration to Russia. Starting as a migration of refugees fleeing the Tajik civil war in 1993–1997, the Tajik migration flow began to include large numbers of labor migrants in years following the war, and became increasingly feminized in the late 2000s. Kyrgyz migration followed Tajik migration chronologically and has had a larger share of women. Mass migration from Uzbekistan started in earnest since the middle of the 2000s, but Uzbek migrants have quickly outnumbered the other two migrant groups because the population of Uzbekistan by far exceeds the populations of Kyrgyzstan and Tajikistan combined. According to the official statistics on foreign nationals residing in the Russian Federation (which are likely to underestimate the actual migrant population), as of February 2016, a 1.8 million Uzbek, 861 thousand Tajik, and close to 563 thousand Kyrgyz adult citizens resided in Russia with 19%, 19%, and 39% of women migrants among them, respectively (FMS 2016). Considering Uzbekistan’s population of 30.8 million, Tajikistan’s population of 8.3 million, and Kyrgyzstan’s population of 5.8 million in 2014 (World Bank 2017), these data suggest that at least 5.8 percent of the Uzbek, 10.8 percent of the Tajik, and 9.7 percent of the Kyrgyz population live in Russia.

Theoretical Framework and Hypotheses

Determinants of migrant incorporation in the host society have received substantial attention from scholars, with the theory and empirical evidence suggesting that individual, economic, social, and institutional factors may all affect migrant economic success and experiences (for an overview of this scholarship, see Portes and Böröcz, J. 1989; Massey et al. 1993, De Haas 2010).

Individual human capital, such as education, skills, language fluency, and past labor market experience come into play before and after migration and affect labor market outcomes of migrants, and the effects of these individual-level factors on migrant aspirations and labor market outcomes may vary across migrant receiving settings (Chiswick 1979; Chiquiar and Hanson 2005; Dustmann 1993; Sharaf 2013; Zhou 2004, 2005). Wage inequalities between the sending and the receiving countries and within the sending and receiving countries are another powerful determinant of migration and composition of the migrant pool (Borjas 1987, Massey et al. 1993, Munshi 2016). The structure of the host economy commonly sets the stage for migrant journeys and trajectories: migrant incorporation and labor market success are more rapid when the host economy is based on a dual labor market where migrant workers are more attractive to employers, who would otherwise have to hire native workers at higher wages and adjust wage scales across the occupational hierarchy to maintain fairness (Piore 1979). Though the Russian labor market as a whole is less polarized than the labor markets in Western economies, which have seen a strong divergence between the skilled and the unskilled labor wages over the past decades,⁴ it does exhibit a strong duality, with low-skill low-pay jobs increasingly reserved for migrant workers and with positions that confer higher earnings and social prestige taken by natives. The notion of “bad jobs” (Kalleberg et al. 2000) as jobs performed by migrant workers is well articulated in urban Russia (Reeves 2013a, 2013b; Schenk 2013), and natives who find themselves working in the same economic niches as migrants are expected to be less satisfied with their employment.

The consequences of the dual labor market for earnings and wage fairness perceptions, however, are less clear. On the one hand, natives employed in similar positions as migrants may work there out of necessity rather than choice, and may lack the bargaining power to command higher wages. On the other hand, migrants may have lower reservation wages⁵ than natives (Holzer 1986) and may be seen by the employers as willing to work more for less pay and treated accordingly (Waldinger 1997). Research by Berhnardt et al.(2013) in large US cities shows that immigrants are more likely to experience workplace violations, such as being paid less than the minimum wage and receiving less compensation for overtime than natives working in similar low-wage positions. Whether systematic differentials in the treatment of migrants and natives exist in Russia and whether migrants see their wages as inferior to natives in similar jobs is a question that needs investigation.

The effects of social networks on migration decisions and outcomes have also received considerable attention from scholars. Diasporas have been shown to attract new migrants, reduce migration costs, and increase the likelihood of finding a job, with both strong and weak social ties playing a role in migrant incorporation (Aguilera and Massey 2003; Beine et al. 2015; Haug 2008; Joassart-Marcelli 2014; Massey and Espagna 1987; Zhou 2005). The effects of social networks on earnings, however, are difficult to predict with certainty. Some studies show their positive effects on earnings and on the likelihood of employment in higher-paying positions (Munshi 2003, Kerr and Mandorff 2015, Patel and Vella 2013); others offer similarly compelling evidence of their negative effects (Hagan 1998; for recent reviews, see Joassart-Marcelli 2014; Sanders, Nee, Sernau 2002). Still other analyses suggest that network effects may be positive for undocumented migrants and insignificant for migrants with regular legal status (Aguilera and Massey 2003). Social networks may boost earnings when migrants enter the labor market via strong entrepreneurial and thriving ethnic enclaves and economies (Kerr and Mandorff 2015) and depress earnings when ethnic enclaves and economies are weak and anchored in blue-collar low-paying occupations (Joassart-Marcelli 2014, Munshi 2016).

The scholarship on the effects of social capital on migrant perceptions of wage fairness and job satisfaction is limited yet generally agrees that migrant group-level characteristics, such as shared ethno-cultural backgrounds and collective experiences of arrival and settlement in the host society, may affect expectations and, consequently, ways in which migrants perceive hardships and opportunities of the workplace, negotiate them with the employer, discuss them within the network of friends and colleagues, and assess their labor market achievements (De Jong 2000; De Jong & Madamba 2001; Kalter and Kogan 2014). Importantly, social networks are embedded in gender and class relations, which may put migrant women at a particular disadvantage (for reviews of the literature on the topic, see Joassart-Marcelli 2014).

The effects of the commonly studied individual, economic, and social factors are complemented by the institutional factors that shape the structure of labor market opportunities for migrants in the host setting. Legal status is often singled out as one of such powerful institutional factors that shape migrants’ economic opportunities in the host society. Studies on migrant legal status suggest that it may significantly affect various aspects of migrant incorporation into the host society including access to and positions on the labor market (Gentsch and Massey 2011, Gleeson 2010; Morris 2003; Snel et al. 2015; Takei et al. 2009), as well as access to housing, education, and social services (for review, see Hagan et al. 1998). Expanding on this literature, we propose that differences in legal status have significant implications for labor market objective and subjective outcomes of migrants in Russia.

Legal Status and Earnings

Research on the effects of legal status typically shows that legalization improves wages and occupational mobility (Amuedo-Dorantes et al 2007; Kossoudji and Cobb-Clark 2002; Powers et al. 1998, Rivera-Batiz 1999, Gleeson 2010; Takei et al. 2009). Kossoudji and Cobb-Clark’s (2002) estimate that migrants with unauthorized status in the U.S. incur an average wage penalty of 14–24 percent. Rivera-Batiz (1999) finds that hourly wages of undocumented migrants in the U.S. are forty percent lower than the wages of legal migrants. Pena (2010) shows are 5–6 percent penalty on the wages of undocumented agricultural workers in the U.S. Recent work by Massey and Gentsch (2014) shows that undocumented Mexican migrants in the U.S. earn less than documented migrants and find themselves in particularly exploitable positions in the context of rising enforcement. While showing positive effects of legal status on earnings, Amuendo-Dorantes et al. (2007) also point out that legalization may also reduce migrants’ labor market participation when it is coupled with an expanded access to social welfare services. Based on this research, we expect that as migrants integrate into Russian society legally and obtain permanent residency or citizenship, their earnings will become similar to the earnings of natives. At the same time, we acknowledge that even a Russian passport may not alleviate the inequality of employment opportunities for a migrant because this inequality is rooted in the perception of ‘otherness’ by Russian natives (Agadjanian et al. 2017, Reeves 2013b). Our conceptualization yields the following hypotheses regarding migrants’ earnings:

H1a: Migrants with a temporary legal status will have lower earnings than migrants with permanent legal status (permanent residence or Russian citizenship), net of other factors.

H1b: Migrants with permanent legal status will have similar earnings as the Russian natives, net of other factors.

Legal Status and Pay Equality Perceptions

Perceptions of pay equality are an important aspect of labor market experiences because they affect job satisfaction (Tortia 2008), effort and turnout (Akerlof and Yellen 1990), and even physical well-being (Falk et al. 2011). Perceptions of wage equality may be influenced by objective factors and subjective experiences. The wage itself and the knowledge of how it is determined are objective factors whereas workers’ interactions with management and coworkers as well as their beliefs that their pay is related to their performance, gender, physique, or migrant status, are examples of subjective experiences. Such subjective experiences may depend on legal, organizational, and social environments of the workplace.

We assume that migrant legal status will affect wage equality perceptions by influencing the information and reasoning that individuals use to determine whether they are paid similar wages for their work as other Russian native women in the same positions. We expect that the likelihood of perceiving wages as similar will be lower among migrants than among natives and particularly low among migrants with a temporary migrant status, controlling for other factors including actual pay. Temporary migrants are likely to have the highest perceptions of pay inequality if they do endure an unequal treatment at the workplace or if they perceive themselves to be treated unfairly. The sensitivities to unequal treatment may be exacerbated by the perceptions of legal vulnerability of temporary migrants in Russia. Migrants with permanent residence/citizenship may also have a stronger perception of wage inequality than Russian natives in the same positions, because they may expect to be treated and rewarded for their work on par with Russian natives yet may experience unfair treatment in the workplace due to their migrant background and racial distinction. Accordingly, we test the following hypothesis on the association of legal status and perceptions of pay equality.

H2a: Migrants with a temporary legal status will have a lower perception of wage equality than migrants with permanent legal status, net of earnings and other factors.

H2b: Migrants with permanent legal status will have lower perceptions of wage equality than Russian natives, net of earnings and other factors.

Legal Status and Job Satisfaction

Prior research suggests that job satisfaction may be shaped by a variety of factors including earnings, social relationships in the workplace (Hulbert 1991), nature of workplace rewards (Linz and Semykina 2012), and cultural characteristics such those associated with nativity (Sabharwal 2011). The valuation of workplace incentives may also depend on individual needs. Income, job security, opportunities for professional development, opportunities for work and life balance as well as social environment at work may be more important for job satisfaction in different circumstances and stages of life (Goldman et al. 2008, Spector 1997). So, for example, opportunities for professional development may be appreciated more by younger workers, flexibility in the workplace may be more important for workers with children, whereas social connections may be particularly valued by workers with stable jobs and satisfied professional aspirations.

Workplace expectations may also be important for job satisfaction (Clark 1997, Liff and Ward 2001, Herrbach & Mignonac 2012). For example, a commonly observed gender differential in job satisfaction, with women being more satisfied than men, may be rooted in women’s lower expectations from the workplace and may disappear after controlling for workplace expectations through worker age, education, professionalism, and occupational environment (Clark 1997). Similar to the differences by gender, job satisfaction may differ by migrant status due to the differences in expectations.

Migrants may come into the labor market without considering occupational prestige, opportunities for occupational mobility, and social benefits of employment to the same extent as native earners (Diaz-Serrano 2013, Waldinger and Lichter 2003). For example, in the context of low-skilled labor migration in Russia, a sales clerk’s position at a retail outlet may be viewed by a migrant woman as relatively prestigious because it is more skilled and higher-status than cleaning, cooking, and other menial jobs typically available to migrants. In line with the dual labor market theory, the same position, may be viewed by a native earner as an ‘immigrant job’ with few opportunities for growth, recognition, and social fulfillment compared to other jobs available to native women. Moreover, legal status may affect these views and expectations among migrants: temporary-status migrants will be more likely to be satisfied with their employment than permanent-status migrants. If permanent migrants view and feel themselves equal to natives, they will be more likely to share the natives’ concerns about the benefits and prospects of low-skilled employment. Therefore, permanent migrants will have similar levels of job satisfaction as their Russian native counterparts. With regards to job satisfaction, our hypotheses are summed up as follows:

H3a: Migrants with a temporary legal status will a have higher level of job satisfaction than migrants with permanent legal status, net of the type of occupation and other factors.

H3b: Migrants with a permanent legal status will have the same level of job satisfaction as Russian natives, net of the type of occupation and other factors.

Data and method

We use data from a survey of women aged 18–40 working in the economic sectors that attract the vast majority of female Central Asian migrants: eateries (mainly as waitresses and cleaners), semi-formal produce and clothing bazaars (as stall owners and vendors), and formal retail and grocery stores (as sales clerks and cleaners). The survey was conducted in 2012–2013 mainly in two Russian cities - Moscow, and Novosibirsk; in a supplementary site, the city of Yekaterinburg, only women working in bazaars were interviewed. The survey sample included women from three migrant groups - Kyrgyz, Tajiks, and Uzbeks - and a control group of native women. The survey instrument contained a variety of questions on women’s socio-demographic, economic, cultural characteristics, as well as on their health. The survey also collected information on migrants’ legal status.

In the absence of a sampling frame for a household-based random selection of migrants, alternative approaches had to be implemented. To sample respondents in eateries and retail sector, the time-venue approach was used. First, the territory of each city was divided into squares of 5×5 km. Then, in randomly selected squares, all eateries and retail establishments were enumerated. Next, eateries and retails were randomly selected from the lists and randomly assigned to one of the four nativity groups (Kyrgyz, Tajik, Uzbek, and native Russian). In each chosen establishment, interviewers attempted to identify and interview a female worker of target nativity at about the same time of the day (see Agadjanian and Zotova 2012, for additional details on implementing time-venue sampling of migrants in the Russian urban context). To sample women working in bazaars, first bazaars were randomly selected from the list of all city bazaars and assigned to one nativity group. In each selected bazaar, random-walk algorithm was used to select and interview women of target nativity. Interviews were conducted by interviewers of matching nativity in the language of respondent’s choice. The analytical sample includes 936 women, about three quarters of whom are international migrants (one-fourth from each ethno-provenance group) and one-fourth are Russia’s natives.

To test the proposed conceptual framework, we fit multivariate regression models: ordinary least squares regression for earnings and binomial logistic regression for wage equality and job satisfaction. The outcome variables are operationalized as follows. Income is measured as respondent’s total income in the past month in thousands of Russian rubles. The perception of wage equality is a binary variable: it takes the value of 1 if a respondent believes that in she earns at least as much as other native women doing the same job (native respondents were asked to compare their earnings with those of other natives); and 0 if otherwise. Job satisfaction is a binary variable that equals 1 if a respondent is completely satisfied with their job and 0 if partially satisfied or not satisfied.

The main predictor for all three outcomes is migration status. It includes three categories: a migrant who is not a Russian citizen or permanent resident (temporary-status migrant), a migrant who is a Russian citizen or permanent resident (permanent-status migrant, the reference group in the models), and a non-migrant, or Russian native (regardless of experience of internal migration).

The controls include human and social capital characteristics which have been shown to affect migrants’ adaptation to and navigation of the new environment (Aguilera and Massey 2003; Massey and España 1987) and are operationalized as follows. ‘Some higher education’ equals 1 if the respondent has at least some years of tertiary education, 0 if otherwise. ‘The number of close relatives and friends living elsewhere in the city’ is a continuous variable that measures the size of a respondent’s personal network (immediate relatives living with the respondent are excluded). The sector of work has three categories: retail outlet, eatery, and bazaar. The occupational status variable is a dichotomy that distinguishes between higher-level occupations (typically managerial positions) and lower-level occupations. The survey site is used as a proxy for the type of local economic and sociocultural environment: Novosibirsk and Yekaterinburg, the two provincial cities, are combined into one category and are contrasted to the capital city of Moscow. The models controls for respondent’s age, partnership status and number of children. In addition, the models for job satisfaction and pay equality perception control for monthly income, and the model for job satisfaction controls for the perception of pay equality.

To capture the effects of the characteristics that are unique to migrants, we also fit models that are restricted to the migrant subsample. In this set of models, permanent-status migrants are compared to those with temporary status. In addition to the covariates described above, these models control for migrant ethno-provenance – Kyrgyz, Uzbek, and Tajik. While the group-specific effects may be complex and require special investigation, we assume that differences in ethno-cultural background and collective migration experiences of the three groups may be consequential for the outcomes of interest. Finally, the migrant-specific models also include measures of migrants’ general integration in the host environment: number of years spent in Russia (a continuous variable) and self-assessed fluency in the Russian language (a dichotomy – functionally fluent vs. not).

Results

Descriptive results

Table 1 presents the descriptive statistics for the outcome and predictor variables. The average monthly income in the sample was 23,000 RUR (around 700 USD using the time-of-the-survey exchange rate), with the average difference of 5,000 RUR between migrants and natives. Seventy-three percent of respondents perceived their earnings to be on par with the wages of other native women working in similar positions; the corresponding share was 67 % and 91 % among migrants and native women, respectively. Forty-six percent of respondents were completely satisfied with their job (48% and 41% of migrants and natives, respectively). Among migrants, 27% had Russian citizenship or permanent residency status. Slightly over 30 percent of respondents worked in retail and eateries each, and the remaining 40 percent worked in bazaars, with no differences between migrants and natives. Only 14 percent of respondents were employed in higher-level positions with professional, or management, or supervisory responsibilities, with a larger share of such positions among native women. Sixty percent of respondents were interviewed in Moscow and the remaining forty percent in Novosibirsk and Ekaterinburg. The average age was thirty years, with no difference between migrants and natives. Thirty percent of respondents, migrants and natives alike, had some higher education. Close to sixty percent had at least one child, with a lower share among natives than migrants (49% and 64%, respectively). Close to seventy percent of respondents had a permanent partner, with little difference between migrants and natives. By design, migrants belonged to one of three ethno-provenance groups —Kyrgyz, Tajik, Uzbek— with each of them making up one-quarter of the total sample. The average duration of life in Russia for migrants was 3.7 years. Forty-one percent of migrants reported fluency in the Russian language.

Table 1.

Descriptive Statistics

	Migrant	Native	All
Variable	Mean or Percent	Mean or Percent	Mean or Percent
Total personal monthly income in 1000s RUR (mean)	21	26	23
Perceives wages as equal to wages of native women (%)	67	91	73
Is completely satisfied with her job (%)	48	41	46
Native (%)	−	−	26
Migrant - a citizen or permanent resident (%)	−	−	20
Migrant - a temporary resident (%)	−	−	54
Works in retail (%)	30	31	31
Works in an eatery (%)	30	31	31
Works at a bazaar (%)	40	38	39
Has a higher-level occupation (%)	10	25	14
Resides in Moscow (%)	60	60	60
Age	30	30	30
Has some university education (%)	28	39	30
Has at least one child	64	49	60
Has a permanent partner	68	70	68
Number of relatives and friends living in the city (mean)	14	25	16
Kyrgyz	−	−	25
Tajik	−	−	25
Uzbek	−	−	24
Years spent in Russia (mean)	3.7	−	−
Fluent in Russian	41	−	−
Number of observations	690	246	936

Multivariate results

Income

The results of the OLS models that predict individual monthly income are presented in Table 2.⁶ The first panel of the table shows that controlling for other factors, the earnings of permanent-status migrants are not statistically different from the earnings of Russian natives. In contrast, temporary-status migrants earn about 5,000 RUR less than non-migrants or permanent-status migrants (about 20 percent less than the average earnings), net of other characteristics. Hypothesess 1a and 1b are therefore supported: migrants with a temporary legal status do earn less migrants with permanent legal status (H1a), and migrants with permanent status earn on par with natives (H1b). The effects of legal status are also robust in the migrant-only model. The migrant-only analysis also shows that Kyrgyz respondents, i.e., members of a group with a privileged path to naturalization, have higher average earnings than either Tajiks or Uzbeks, regardless of legal status and other individual characteristics.

Table 2.

Total monthly income in thousands of Russian rubles - ordinary least squares parameter estimates, standard errors in parentheses

Predictors	All		Migrants only
Migration Status
Non-migrant	−0.58 (1.06)
[Migrant – a Russian citizen or permanent resident]
Migrant - not a Russian citizen or permanent resident	−4.99 (0.96)	**	−3.02 (0.87)	**
Controls
Retail	−3.80 (0.82)	**	−3.83 (0.74)	**
Eatery	−4.42 (0.87)	**	−3.93 (0.78)	**
[Bazaar]
Higher-level occupation [lower-level occupation]	9.67 (1.05)	**	6.35 (1.08)	**
Moscow [Novosibirsk or Yekaterinburg]	7.27 (0.73)	**	4.78 (0.70)	**
Age	0.13 (0.06)	*	0.17 (0.06)	**
Some university education [no university education]	3.22 (0.75)	**	1.90 (0.74)	*
At least one child [no children]	−0.35 (0.88)		−0.35 (0.85)
Permanent partner [no permanent partner]	0.47 (0.76)		1.05 (0.72)
Number of close relatives & friends in the city	−0.01 (0.01)		0.00 (0.01)
Kyrgyz			3.75 (0.83)	**
Uzbek			−0.93 (0.76)
[Tajik]
Years in Russia			0.06 (0.09)
Fluent in Russian [not fluent in Russian]			0.81 (0.67)
Intercept	17.33 (1.95)	**	14.93 (1.88)
Adjusted R-squared	0.25		0.27
Number of observations	936		690

Note: Reference categories in brackets; significance levels:

^**<.01,

^*<.05,

⁺<.10

The effects of other variables are noteworthy. The models demonstrate strong and robust effects of the employment sector: work at bazaars offers an average earnings premium of 3–4 thousand RUR (about 13–17 percent of the average income) compared to employment in retail or eateries. As could be expected, work in higher-level occupations is associated with higher wages. However, these effects are substantially smaller for migrants. Moscow residence connotes higher earnings than residences in the two provincial Russian cities. Respondent’s age is positively associated with earnings, probably picking up the effect of professional skills and experience on the job market. Having some tertiary education is also positively associated with earnings; however, this association is weaker in the model for migrants. Neither marital status nor having a child show a statistically significant relationship with earnings. The number of close relatives and friends living in the same city as the respondent, is not related to earnings either. Surprisingly, among migrants, the length of stay in Russia and Russian fluency do not show a statistically significant association with earnings, net of other factors, though the coefficients for both variables are positive, as could be expected.

Perceived wage equality

In Table 3, we present the results of logistic regression models that predict perceived equality of wages and complete job satisfaction. As can be seen in the first panel, the odds of native women to view their pay as comparable to that of other Russian women are 206 percent higher than the odds of migrants with permanent status (native/permanent migrant odds ratio - 1=exp (1.12) −1 = 3.06–1=2.06). The odds of temporary-status migrants to perceive their pay as equal to the pay of Russian women are 44 percent lower than those of permanent-status migrants (temporary migrant/permanent migrant odds ratio=1-exp (−0.58) = 1–0.56=0.44).

Table 3.

Perceived Pay Equality and Job Satisfaction – logistic regression parameter estimates, standard errors in parentheses

	Perceived Pay Equality				Complete Job Satisfaction
Predictors	All		Migrants only		All		Migrants only
Migration Status
Non-migrant	1.12 (0.29)	**			−0.78 (0.22)	**
[Migrant – a Russian citizen or permanent resident]
Migrant - not a Russian citizen or permanent resident	−0.58 (0.21)	**	−0.72 (0.25)	**	−0.15 (0.20)		−0.32 (0.24)
Controls
Total monthly income in ‘000s of Russian rubles	0.00 (0.01)		0.00 (0.01)		0.03 (0.01)	**	0.07 (0.01)	**
Perceived pay equality					0.39 (0.16)	*	0.31 (0.18)	+
Retail	0.12 (0.19)		0.05 (0.21)		0.24 (0.17)		0.51 (0.21)	*
Eatery	0.25 (0.20)		0.28 (0.22)		0.48 (0.18)	**	0.71 (0.22)	**
[Bazaar]
Higher-level occupation [lower-level occupation]	−0.01 (0.28)		−0.28 (0.31)		0.64 (0.23)	**	0.72 (0.32)	*
Moscow [Novosibirsk or Yekaterinburg]	0.07 (0.18)		0.25 (0.20)		0.18 (0.16)		0.10 (0.20)
Age	0.01 (0.15)		−0.00 (0.02)		−0.02 (0.01)		−0.03 (0.02)
Some university education [no university education]	−0.09 (0.18)		0.06 (0.20)		−0.19 (0.16)		−0.14 (0.20)
At least one child [no children]	0.32 (0.21)		0.23 (0.23)		−0.04 (0.18)		−0.21 (0.23)
Permanent partner [no permanent partner]	−0.11 (0.18)		−0.18 (0.20)		0.34 (0.16)	*	0.40 (0.20)	*
Number of close relatives & friends in the city	0.01 (0.00)	*	0.00 (0.00)		0.01 (0.00)	+	0.00 (0.00)
Kyrgyz			−0.11 (0.23)				−0.51 (0.23)	*
Uzbek			0.76 (0.22)	**			0.59 (0.21)	**
[Tajik]
Years in Russia			0.04 (0.03)				0.01 (0.03)
Fluent in Russian [not fluent in Russian]			−0.22 (0.19)				−0.24 (0.19)
Intercept	0.62 (0.54)		0.44 (0.46)		−1.03 (0.43)	+	−1.35 (0.56)	*
− 2LL and Chi-square	−504 | 80		−418 | 39		−606 | 81		−431 | 94
Number of observations	936		690		936		690

Note: Reference categories in brackets; significance levels:

^**<.01,

^*<.05,

⁺<.10

In the migrant-only model presented in the second panel of Table 3, the odds of temporary-status migrants to perceive pay equality are 51 percent lower than the odds of migrants with permanent status (1-exp(−0.72)=1–0.49=0.51), net of other factors. These results forcefully illustrate the importance of migrant status and of the legal distinction among migrants in perceptions of wage equality, as predicted by Hypotheses 2a and 2b.

Among other results, the number of close relatives and friends in the city tends to be positively associated with wage equality perceptions in the full-sample model though the effect loses statistical significance in the migrant subsample. Notably, income itself does not have an effect on the perceptions of pay equality and nor do the sector of employment or most individual-level variables.

Job satisfaction

Panels 3 and 4 of Table 3 display the results of a logistic regression model that predicts complete satisfaction with the current job. The model includes the same covariates as the previous model plus the perception of pay equality. No statistically significant difference in job satisfaction exists between migrants with a temporary and a permanent legal status. Hence, H3a is not supported. Moreover, contrary to H3b, which predicted an equal level of job satisfaction for migrants with permanent status and natives, migrants with permanent status, the reference group, are almost twice as likely to be satisfied with their jobs as native women (the coefficient for natives is: 1-native/permanent migrant odds ratio=1-exp (−0.78) = 1–0.46=0.54).

Among other effects, it is noteworthy that human capital is not significantly associated with job satisfaction. All other control variables with statistically significant effects have signs in the predicted direction: higher-level occupation increases the odds of job satisfaction, and so does income. The perception of pay equality increases the likelihood of a woman to be fully satisfied with her job. Curiously, the models demonstrate a positive association of having a permanent partner with a woman’s overall job satisfaction. While this association lies outside the scope of our theoretical interest, we speculate that compared to a single woman, a partnered woman may perceive her job as of secondary importance to that of her partner and therefore less consequential for her overall wellbeing and, therefore, may be more content with it.

Discussion

Guided by the scholarship on migrant economic incorporation into the host society primarily in the US and other Western settings, we have examined earnings as an objective measure of migrant incorporation into the host society and perception of pay equality and job satisfaction as subjective measures of migrant incorporation. With regards to earnings, our findings document a familiar gap between migrants and natives (Berhnardt et al. 2013; Carliner 1980; Chiswick 1979; Dustmann 1993; Hall and Farkas 2008) but also illustrate how permanent legal status erases the migrant-native earnings differentials, regardless of other characteristics. The upward effect of legality on migrants’ income conforms to what has been documented in other settings (e.g., Amuedo-Dorantes 2007; Dustmann 1997, Rivera-Batiz 1999), and the net contrast between migrants with and without permanent legal status that we detected in this relatively low-income segment of the Russian labor market is quite substantial.

Whereas a positive association of permanent legal status with migrant earnings is probably universal, it is of particular importance in Russia where workplace relations between employers and employees are often characterized by high informality and illegality (Ledeneva 1997, 2009; Light 2016), which constricts the role of regulations, due process, and law enforcement more broadly in the operation of businesses and public sector institutions (Hendley 1999; Hendley et al. 2000; Schenk 2013). The effect of legal status on earnings may be muted when it is not uncommon for migrants with regular legal status to receive shadow wages for formal work and for migrants with temporary legal status to receive shadow wages for shadow work due to limited quotas for legal migrant employment (Schenk 2013). The economic benefits conferred by permanent legal status may even be surprising in light of the research on Russian immigration policies that casts doubt on the ability and willingness of the federal government, public officials, law enforcement, and the private sector to respect the rights of individuals with lawful migration statuses (Gerber and Mendelson 2008; Light 2016; Schenk 2010, 2013).

A reconciliation of the positive effect of legal status on earnings with the policy reality may, therefore, be sought in the mechanism by which permanent status exerts this effect. While institutional-level forces may be at play in the relationship of permanent migrant status and earnings, individual-level explanations should not be discounted either. Permanent status may empower migrants to search and negotiate for better labor market opportunities. Having jumped through the hoops of becoming a permanent resident or citizen, migrants may also have relatively high self-efficacy as economic actors who can better compete in the labor market than migrants with a temporary status.

Both temporary and permanent migrants were much less likely to view their wages as equal to the wages of other Russian women in similar occupations, after controlling for earnings and other variables. These differences may be indicative of subtle dimensions of migrant reception in Russia which include overt and masked discrimination (Agadjanian et al. 2017; Alexseev 2015; Regamey, 2010; Reeves, 2013a, 2013b). Though the Russian Federation does have laws to protect ethnic minorities, such laws bear little relevance to how migrants are actually received and treated. Echoing the views on migrants in the Western settings (Bloemraad and Wright 2014; Ceobanu and Escandell 2010; Reitz 2002), the attitudes to migrants in Russia are ambiguous: migrant labor is viewed as both a necessity and a threat (Regamey 2010; Schenk 2013). A significant gap in the perception of pay equality between temporary-status and permanent-status migrants also illustrates the importance of legal status in mitigating adverse experiences in the labor market. In sum, it is in the perception of pay equality that the expected relative positioning of the three groups—natives, permanent-status migrants, and temporary-status migrants—is most clearly manifested.

Finally, in the analysis of overall satisfaction with the current job, we found a pronounced difference between migrants and natives and no difference between the two legal categories of migrants. Migrants with temporary and permanent status alike were significantly more likely to express satisfaction with their jobs than were their native counterparts. Although we have no definitive explanation for the similarity between the two categories of migrants, we suggest that all migrants, regardless of their legal status, might continue assessing their job experiences and opportunities with reference to those in their communities of origin (Stark & Taylor 1989), which typically offer lower earnings and inferior working conditions for similar work. By the same logic, the difference between migrants and Russian natives may be reinforced because Russian natives may have higher reference standards than migrants. Our findings may also reflect the ethno-cultural differences in overall life satisfaction between Central Asians and Russian natives, which have transpired through recent surveys (Di Bartolomeo et al. 2014).

Several limitations of the study must be acknowledged. First, this research is based on a sample of women working in three sectors of urban economy in three Russian cities. Even though these sectors employ the largest shares of female migrants from Central Asia, the results of the study should be generalized across other economic sectors and geographies with caution. Although the survey included uniquely rich data on migrant legal status, no information on the timing of the transition between different statuses was available. Also, as is typical of cross-sectional studies of migrants, we are unable to account for self-selection into migration or group-level migration experiences. Finally, it should be pointed out that Russian natives working the examined segments of the economy, while directly comparable to the migrants working in the same segments, are not representative of all natives who have a broader range of sectoral and occupational choices.

These limitations notwithstanding, our analysis sheds important light on the dynamics of economic inclusion of migrant women in a country with the second largest immigrant population in the world and contributes to a better understanding of these dynamics both in Russia and globally. In a broader sense, our findings illustrate how the forces of unequal development (Wise et al. 2013) constrict legal, social, political, and economic space around migrants and shape their experiences and outcomes.

Biography

Evgenia Gorina is an Assistant Professor at the University of Texas at Dallas. Her research on migration focuses on the determinants and outcomes of migration in the Post-Soviet space, including migration into Russia and internal migration within former Soviet republics.

Victor Agadjanian is a Foundation Distinguished Professor at the University of Kansas. He conducts research on various aspects of social and demographic change in developing and transitional settings. He has studied migration, sexual and reproductive behavior, gender, ethnicity, and religion. He has directed several large projects funded by the National Institutes of Health, United States Agency for International Development, and other agencies in sub-Saharan Africa and Central Eurasia (Russia, the Caucasus, and Central Asia).

Natalya Zotova is a doctoral student at the Department of Anthropology at Ohio State University. She holds advanced degrees in social anthropology from Russia. Her current dissertation research focuses on recent Central Asian immigrants in the U.S., their perceptions of security and insecurity, as well as the role of social networks as mediators of such insecurity.

APPENDIX

Logarithm of total monthly income in thousands of Russian rubles - ordinary least squares parameter estimates, standard errors in parentheses

Predictors	All		Migrants only
Migration Status
Non-migrant	−0.10 (0.04)	*
[Migrant – a Russian citizen or permanent resident]
Migrant - not a Russian citizen or permanent resident	−0.22 (0.04)	**	−0.13 (0.04)	**
Controls
Retail	−0.13 (0.03)	**	−0.13 (0.03)	**
Eatery	−0.15 (0.04)	**	−0.15 (0.04)	**
[Bazaar]
Higher-level occupation [lower-level occupation]	0.36 (0.04)	**	0.28 (0.05)	**
Moscow [Novosibirsk or Yekaterinburg]	0.38 (0.03)	**	0.29 (0.03)	**
Age	0.01 (0.00)	+	0.01 (0.00)	+
Some university education [no university education]	0.11 (0.03)	**	0.08 (0.04)	**
At least one child [no children]	0.02 (0.04)		0.06 (0.04)
Permanent partner [no permanent partner]	0.01 (0.03)		0.06 (0.03)	*
Number of close relatives & friends in the city	−0.00 (0.00)		0.00 (0.00)
Kyrgyz			0.17 (0.04)	**
Uzbek			−0.04 (0.03)
[Tajik]
Years in Russia			0.00 (0.05)
Fluent in Russian [not fluent in Russian]			0.03 (0.03)
Intercept	2.78 (0.08)	**	2.65 (0.09)	**
Adjusted R-squared	0.25		0.27
Number of observations	936		690

Note: Reference categories in brackets; significance levels:

^**<.01,

^*<.05,

⁺<.10