Keywords

1 Introduction

According to the 1990 United Nations (UN) International Convention on the Protection of the Rights of All Migrant Workers and Members of their Families (ICRMW), the human rights of labor migrants should be given special attention. This has become part of the general awareness of, sensitivity to, and discussion on human rights in Central Asian societies.

According to World Bank data from 2018, remittances made up a total of 31% of Tajikistan’s GDP. The corresponding figure for Kyrgyzstan was 33.3%, for Uzbekistan it was 9%, and for Kazakhstan 0.2% (Knomad and World Bank 2019). Despite the relatively low shares of remittances in the GDP of Uzbekistan and, notably, Kazakhstan, CAB Survey dataFootnote 1 shows that remittances are equally important for respondents’ financial circumstances in all five Central Asian states. Notably, the results were similar across all five countries, with approximately 70% of respondents reporting that remittances were important for their family’s financial circumstances. This clearly shows not only the importance of remittances for the economies in the region but also illustrates that very large shares of Central Asian citizens are working as labor migrants abroad. Given these statistical indicators, it is worth investigating and initiating a critical discussion on the human rights situation of these migrants.

Russia is the host country for the majority of CA labor migrants as substantial shares of the remittances to Central Asia come from Russia: 83% of Kyrgyzstan’s, 58% of Tajikistan’s, and 55% of Uzbekistan’s in the first three quarters of 2021 according to the World Bank report (Ratha and Kim 2022). Another report, this time from the International Organization for Migration (IOM) states that about four million migrants from CA work in Russia: one million from Tajikistan, the same number from Kyrgyzstan, about two million from Uzbekistan, and 200,000 from Kazakhstan (IOM 2022).

It is a proven fact that labor migrants from CA face immense and numerous difficulties in Russia as well as some other countries, including exploitation by employers, low standards of living, unsafe and exhausting working conditions, and poor health services. As a result, they are at high risk of sustaining injuries and even dying due to poor working conditions, as is regularly reported by numerous news agencies and journalistic investigations covering labor migration in Russia (Redakciya 2021; Current Time 2021). According to Turaeva and Uriboyev, Central Asian labor migrants are often portrayed as “illegal”, “illiterate”, “wild”, and “dangerous” in Russian public discourses and those with work permits still have trouble securing formal employment, leaving them open to abuse (Turaeva and Uriboyev 2021: 19–20).

There are also many concerns about the treatment of labor migrants in Turkey, which became the focus of public discussions in 2019 after the death of a Turkish MP’s Uzbek housecleaner Nadira Kadirova (Eurasianet 2019). According to the Turkish ministry of the interior, Turkmenistan, Uzbekistan, and Kyrgyzstan are among the top ten countries whose citizens have permission to work in Turkey. In addition, in 2017, Uzbekistan also entered the top five countries in terms of the number of illegal migrants detained in Turkey (Russia Today 2019).

Public attitudes in all five CA countries concerning migration issues in general and the discrimination of compatriots working abroad is a topic that has received little attention from scholars conducting either scientific or applied research, in comparison to the breadth of research carried out on the public attitudes of hosting societies toward these migrants. Since 1989, the Levada Center has been regularly measuring xenophobia trends toward different groups within Russian society through its national surveys, which are part of the center’s long-term studies focused on xenophobia, racism, and antisemitism. However, much less research has been conducted on the attitudes of the Russian public concerning the discrimination of labor migrants. A recent Levada study, using the Bogardus social distance scale, shows that the majority of the Russian population continues to maintain ethno-national and racial prejudices toward different groups, including labor migrants. Russians seem to be most distanced from the Roma and people from Africa, and Central Asia. In addition, over the past five years, the proportion of those who believe that the government should prevent the flow of migrants has grown from 58% to 68–73% (Levada Center 2022).

Much of the literature on the nexus of public opinion and discrimination toward labor migrants posits that the attitudes of a host society toward these migrants play an important role in shaping the policies, discriminatory or not, which impact them (Tunon and Baruah 2012). Moreover, although, due to the varying levels of dependency on remittances, sending societies may be expected to have only weak leverage power to influence migration regimes in Russia and other host countries, the attitudes of the sending societies concerning such issues are in fact also important. It may seem that Central Asian migrants, on an individual level, are not in a position of power when negotiating with their job providers or with law enforcement officials. However, in Russia, Central Asian migrants are beginning to form different communities, some of which can be regarded as self-sufficient, based on various institutions including medical clinics, banks, cafes, schools, daycare facilities, and sophisticated money transfer services tailored to labor migrants (Ria Novosti 2013). Such efforts allow them to embrace some form of subjectivity and possibly enable them to develop collective action. In light of all this, discussions on Central Asian public opinion such as this one with a focus on issues of discrimination, have the potential to shed light on the effect that these attitudes have on the levels of discrimination faced by their labor migrants.

Moreover, the period between the fall of 2021 and the spring of 2022 saw many notable events, both on the regional and global levels. Russia’s war against Ukraine is widely influencing Central Asian migration patterns. Asia-Plus reports that in the first quarter of 2022, more than 60,000 migrants returned to Tajikistan, which is 2.6 times more than last year (Asia-Plus 2022). Azizbek Yunusov, deputy head of the Agency for External Labor Migration of the Republic of Uzbekistan, recently stated that in the first quarter of the year, 133,000 labor migrants returned to Uzbekistan from Russia. Yunusov also shared data from a survey conducted among 150,000 citizens of Uzbekistan, in which 15% of those surveyed said that they had lost their jobs in Russia. Another 25% faced problems due to the unstable exchange rate and would therefore like to return to Uzbekistan (Migration Uzbekistan 2022). It is worth discussing how these events have influenced Central Asian public opinion on migration issues and the level of concern over the treatment of labor migrants abroad based on a comparison of the survey data from fall 2021 and spring 2022. It is also worth noting that, due to various factors, public opinion often fluctuates, and it is best to measure it weekly to be able to accurately monitor the trends. Nevertheless, comparing the results of two surveys conducted using similar methodology in different periods maybe also be useful.

2 Awareness of Migrant Rights in Five Central Asian Countries in Fall 2021 and Spring 2022

First, the study looked into the general awareness of human rights-related matters pertaining to migrants from all five Central Asian countries.Footnote 2 During the fall of 2021 and spring of 2022, in all five CA countries, Central Asia Barometer (CAB) asked a similar set of questions concerning migration also employing a similar methodology and simultaneously conducting fieldwork. This enabled a comparison of survey outcomes between countries. Within this study, simple random sampling (SRS) of mobile phone numbers was used to obtain a nationally representative sample (1,500 interviews per country) in Turkmenistan, Kazakhstan, Kyrgyzstan, Uzbekistan, and Tajikistan. In Kazakhstan, Kyrgyzstan, and Tajikistan, all interviews were conducted via desktop CATIFootnote 3 applications from either in-office or at-home stations. In Uzbekistan and Turkmenistan, all interviews were completed on tablets with calls made via mobile phones from at-home stations. Along with the migration section, questions were asked about media usage, the political and economic situation in each country, public health, attitudes toward the government, and opinions about other countries. In addition, questions aimed at gathering demographic information were included, such as age, education level, and employment status. The data is nationally representative in terms of gender, age group, and urban/rural divide. Assuming a simple random sample, with p = 0.5, at the 95% CI level, the margins of error (MOE) in all five countries are at 2.53%. Incorporating the mean design effects into these estimates yields margins of error of from 3.09% to 3.29% across countries (Central Asia Barometer 2022). In the following, we will discuss the main insights from these two surveys conducted across Central Asia before and after the conflict in Ukraine.

In general, the survey data suggests that the majority of respondents across Central Asia were concerned about how labor migrants from their countries were treated in other countries. When asked “How concerned are you, if at all, about the treatment of our country’s labor migrants abroad?”, around 55% (fall 2021) and 50% (spring 2021) of Uzbek respondents; 60% (fall 2021) and 62% (spring 2022) of Kazakh respondents; 85% (fall 2021) and 84% (spring 2022) of Kyrgyz respondents; 60% (fall 2021) and 70% (spring 2022) of Tajik respondents; 40% (fall 2021) and 57% (spring 2022) of Turkmen respondents answered they were very or somewhat concerned.

If we compare the fall and spring survey data, we can see that in Tajik and Turkmen societies there was a slight increase in concern about the issue, while this was not the case for other Central Asian countries. Consequently, we conclude that the topic of Russia’s war in Ukraine and the consequent sanctions imposed on Russia had no notable influence on public opinion.

Pronounced divides in awareness

There appear to be significant differences in respondents’ opinions on migration within urban and rural areas in Kazakhstan and Kyrgyzstan. Rural respondents showed a high level of concern, around 70% (very and somewhat concerned), about the treatment of their compatriots working abroad, while around 55% of urban dwellers gave this response. In Kyrgyzstan, rural respondents were just slightly more concerned about the topic. In the other countries of Central Asia, there were no significant differences in the opinions of urban and rural respondents. It can be inferred that, despite the fact that this did not apply to other Central Asian countries, communities in Kazakhstan and Kyrgyzstan are more informed about and aware of the issues faced by labor migrants abroad, which may, in turn, serve as an illustration of the urban/rural divide in migration patterns in these countries. Moreover, in Kyrgyzstan, urban respondents tended not to choose Russia as the most attractive country for labor migration. A total of 70% of rural respondents, in contrast to 55% of urban respondents, believe that Russia is the most attractive country.

In Kyrgyzstan, for example, there is a clear correlation between age and perception of Russia as the best country in which to work. The older the respondent is, the more attractive Russia appears to be for labor migration. Only approximately 55% of respondents aged between 18 and 29 years chose this option, compared to around 75% of those over the age of 60. However, this only applies to the Kyrgyz population and shows that the country’s younger citizens are more actively exploring new destinations for labor migration, in contrast to the older generation.

In Kazakhstan, the level of concern about the treatment of labor migrants is clearly correlated with the age of the respondent—the higher the age, the greater the concern. However, in the case of the treatment of labor migrants abroad, only in Kazakhstan was there a significant difference in the level of concern. This may indicate that the younger generation is less informed about the issues these migrants face due to a lack of first-hand information or experience about the hardships of external labor migration. In other Central Asian countries, young people express similar levels of concern about the topic.

Another divide is along gender lines. In Tajikistan, when asked whether they had a family member who works in another country, around 70% of respondents answered in the affirmative. More female respondents reported having a family member abroad than male, with 75 and 55%, respectively. This may be due to the gendered differences in Tajik labor migration patterns, with numerous studies showing that Tajik external migration is male dominated.

When it comes to the questions regarding concern about the topic, there were no significant differences between genders, which may illustrate that women and men across all five nations are more or less equally informed on the topic.

Taking into account that most respondents in Tajikistan believe Russia is the most attractive country in which to work as a labor migrant and that a similar share of the population is concerned about the treatment of their labor migrants abroad, it can be concluded that respondents assess the “attractiveness” of the host country not based on their assumptions of good working conditions or societal acceptance, but based on the feasibility of securing work.

It seems that Tajik respondents see no other option for labor migration besides Russia with 90% reporting this to be the case. This high number could be due to several reasons including migration regimes. The same does not apply to other Central Asian countries, however, which, according to the survey data, have other country options at their disposal. For example, only 65% of Kyrgyz, 40% of Uzbek, 20% of Turkmen, and 15% of Kazakh respondents chose Russia. Tajikistan and Kazakhstan are at opposite ends of the spectrum here, with the vast majority of respondents in the former seeing Russia as the only option, and respondents in the latter being extremely divided on the issue, with only an insignificant share believing Russia to be the most suitable nation for its labor migrants.

Thus, the conclusion we come to is more counterintuitive. Despite all Central Asian countries being viewed as heavily dependent on remittances from Russia, this is largely only the case for Tajikistan and to a lesser degree for Kyrgyzstan. Central Asian populations are actively exploring other countries as destinations for labor migration starting with the nations closer to home, such as Kazakhstan and Turkey, as well as much more unexpected options, such as Canada, the USA, and Korea.

3 Discussion of the Survey Results in Light of Response and Nonresponse Bias in an Authoritarian Context

Each CA country showed a different level of concern about the treatment of their labor migrants abroad, both in fall 2021 and spring 2022, which could be due to a variety of reasons. Of course, the results should not be uniformly interpreted as implying that the less a sending population is concerned, the fewer issues their compatriots routinely experience in a host society. Clearly, there are differences between the Central Asian countries when it comes to migration patterns, destinations, and migration regimes in host countries, which to some extent may explain the differences in the levels of concern on the issue. There are, however, other factors and in the following, we will discuss the effect of the response and nonresponse bias within an authoritarian context, especially in Turkmenistan, Uzbekistan, and Tajikistan.

Response and nonresponse bias within the Central Asian context is one of the most pressing challenges for researchers in the region, due to the authoritarian nature of many of the political regimes and a general lack of awareness concerning survey research. Central Asian countries are some of the hardest in the world in which to conduct opinion polling and the CAB Survey faces various limitations: we cannot ask political, religious, or security-related questions, and before each survey, we need to obtain special permission from a governmental agency. Central Asia Barometer has experienced numerous incidents during fieldwork, such as interviewers being arrested and their survey tablets confiscated—even when they had formal permission letters at hand. Although there are no formal bans in place, these incidents have created a sense of insecurity among the research companies CAB works within these countries, increasing their self-censorship even further. This perceived and real risk of repressive action makes survey design and fieldwork a much longer, ambiguous, and difficult process, during which most political, economic, and religious questions are identified as sensitive. This applies to the different Central Asian countries to varying degrees, as CAB is freer to conduct survey research in Kazakhstan and Kyrgyzstan than in the other three countries.

In the following, we will discuss several “coping mechanisms” resulting from a fear of government sanctions, which respondents may use during interviews and which have a proven effect on response bias within an authoritarian context and as such should be taken into account when comparing the cross-country survey data on migration.

Respondents tend to choose the “don’t know” answer option more frequently

Researchers usually do not read out the “don’t know” answer option, thus omitting it from the questionnaire (Krosnick et al. 2002). This also applies to the CAB Survey. The reason for this approach is that respondents tend to choose this option more often, as it appears to be the safest and easiest when compared to other answer options. By choosing “don’t know” respondents do not have to make an effort or risk their well-being. This is largely the case for other sections of the CAB Survey and, on average, Turkmen respondents tend to choose this option more than respondents from other Central Asian countries. For example, as the survey data shows, of all the CA countries, Turkmenistan seems to be the least concerned about the treatment of their labor migrants abroad. Yet, around 35% of Turkmen respondents chose the “don’t know” option, while in other countries this answer option was generally selected by less than five percent of respondents, with the exception of Tajikistan. Consequently, the number of respondents who expressed their level of concern is lower than in the other countries surveyed. This is why when comparing countries we need to take into account that respondents in more closed countries such as Turkmenistan may choose the “don’t know” option more frequently than others to avoid providing answers to certain questions which they believe would not be safe for their well-being.

Respondents refrain from expressing opinions that they think contradict their government’s stance on a given issue

The question regarding concern about the treatment of labor migrants abroad attempts to shed light on negative manifestations of social life as it is related to the issues of discrimination and the government’s inability to provide jobs at home. The Central Asian authorities face criticism over their inability to provide jobs and better protect their citizens in host countries, especially in Russia.

For example, in 2016, a total of 17 labor migrants working in Moscow died in a fire caused by poor working conditions and this tragic event has fueled public criticism of the Kyrgyz government (Current Time 2016). The fact that this question is related to the government’s performance increases the risk of the results being skewed by a higher response bias and refusal bias (as a component of nonresponse bias).

Numerous studies have discussed the effect of the government’s perceived position on respondents’ answers. Chia (2014), for example, in her article for the International Journal of Public Opinion Research states that “in authoritarian countries, people’s willingness to voice opinions may be a function of their perceptions of the government’s stance, and individuals are likely to refrain from expressing opinions that appear to contradict the authority”.

Respondents tend to try to provide answers that are identical to the perceived stance of the authorities on the given issue

In his recently published article, Marcus Tannenberg (2022) outlines the results of his multilevel analysis of surveys conducted in 37 African countries (with 228,000 respondents) to test for systematic bias. Tannenberg wanted to observe how respondents would answer certain questions when they believed that the government was conducting the survey, not independent researchers. This is also frequently the case in Central Asia. He concludes that the fear of sanctions from the government “induces a significant bias on questions regarding trust, approval and corruption perceptions in more autocratic countries, but not in more democratic countries” (591) and respondents tend to express more positive attitudes about issues raised in the questions asked. As outlined above, the question on the concern about labor migrants abroad is not only political but also directly related to the government’s performance. Respondents in CA may answer that they are not concerned about the issue, because they perceive this to be the government’s stance.

The systematic bias present in authoritarian contexts is a methodological concern that needs to be taken into account at all stages of survey research—especially when interpreting survey results concerning opinions related to the performance of the government. However, these tendencies do not mean that public opinion research is irrelevant in Central Asia or other authoritarian countries. According to Chang (1999) and his case study of public opinion in Singapore, even authoritarian authorities “have to concern themselves with managing perceived public opinion”. In Central Asia, state media regularly highlights the popularity of the leader and the consensus among the people on the direction in which the country is moving and governmental policies in an attempt to influence not only individuals’ opinions but also the people’s perception of public opinion.

4 Conclusion

In sum, and as expected, due to the vast number of reports on discrimination against Central Asian labor migrants in Russia, respondents from Central Asian societies are mostly concerned about the treatment of their labor migrants abroad. In the CAB Survey, respondents in Central Asia expressed varying degrees of concern about this issue, which also needs to be analyzed in light of response and nonresponse bias in an authoritarian context, in addition to other factors.